1 s2.0 S0923474816300352 Main

International Journal of Contemporary Hospitality Management
E-commerce performance in hospitality and

tourism Nan Hua,
Article information:
To cite this document:
Nan Hua, (2016) "E-commerce performance in hospitality and tourism", International Journal of
Contemporary Hospitality Management, Vol. 28 Issue: 9, pp.2052-2079,
https://doi.org/10.1108/ IJCHM-05-2015-0247
Permanent link to this document:
https://doi.org/10.1108/IJCHM-05-2015-0247
Downloaded on: 07 January 2019, At: 20:11 (PT)
References: this document contains references to 166 other documents.
Downloaded by Iowa State University At 20:11 07 January 2019 (PT)
To copy this document: permissions@emeraldinsight.com

The fulltext of this document has been downloaded 7482 times since 2016*
Users who downloaded this article also downloaded:
(2017),"Factors affecting desire to negotiate again", Journal of Managerial Psychology, Vol.
32 Iss 1 pp. 16-29 <a href="https://doi.org/10.1108/JMP-10-2015-
0384">https://doi.org/10.1108/ JMP-10-2015-0384</a>
(2017),"Negotiating your way to success: The importance of culture in global negotiation strategy",
Strategic Direction, Vol. 33 Iss 9 pp. 7-9 <a href="https://doi.org/10.1108/SD-06-2017-
0095">https:// doi.org/10.1108/SD-06-2017-0095</a>
Access to this document was granted through an Emerald subscription provided by

emerald-srm:178063 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The
company manages a portfolio of more than 290 journals and over 2,350 books and book series
volumes, as well as providing an extensive range of online products and additional customer
resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative
for digital archive preservation.
*Related content and download information correct at time of download.

The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0959-6119.htm
E-commerce
IJCHM performance in
28,9
hospitality and tourism
Nan Hua
University of Central Florida, Orlando, Florida, USA
2052
Received 21 May
2015 Abstract
Revised 20 October Purpose – This study aims to examine the extant E-commerce performance
2015
literature to derive a coherent framework to further the understanding, identify
3 February 2016
Accepted 12 February research gaps and suggest potential future study directions.
2016 Design/methodology/approach – Based on theoretical sampling (Corbin and
Strauss, 2008), this study followed Greenhalgh et al. (2009) for the paper
sample collection and used exploratory methods before the snowballing
technique to identify key sources to uncover the E-commerce performance

themes and prior findings systematically.
Findings – By reviewing and synthesizing 155 recent articles, this study
proposed an integrated framework of E-commerce performance to organize the
complex literature parsimoniously. This study found that E-commerce
performance exhibits three key dimensions and is influenced by market E-
commerce environment, organization E-commerce environment and the
dynamic and interactive relationships in between.
Practical implications – The proposed framework offers industry practitioners
opportunities to understand determinants and be updated with current practices
of E-commerce performance. The findings of this study further point
practitioners to directions that can lead to better E-commerce performance.
Originality/value – This study produced a cohesive framework of E-commerce
performance based on an extensive review of the literature in both the
mainstream and hospitality and tourism fields, addressing the issue of the
currently fragmented understanding on E-commerce performance in hospitality
and tourism.
Keywords Tourism, Electronic commerce, Performance measurement,
Hospitality industry, Hospitality, E-commerce performance
Paper type Literature review
a Emerald Group Publishing Limited 0959-6119
DOI 10.1108/IJCHM-05-2015-0247
International Journal of
Contemporary
Hospitality
Management
Vol. 28 No. 9, 2016
pp. 2052-2079
1. Introduction through and stay for the hospitality and tourism industry, regardless of the well-
Over the past two recognized reluctance for hoteliers to adopt new technologies (Buhalis, 2003;
decades, the
explosion of research
interest related to E-
commerce
performance has
stimulated a
substantial number
of academic studies
(Chae et al., 2014;
Hua et al., 2015; Law
et al., 2014; Yang et
al., 2015). Although
these studies have
significantly enriched
the body of
knowledge for E-
commerce
performance, new
challenges surface
as the sheer work
volume has become
more or less an
obstacle for
knowledge
development and the
literature has still
remained
fragmented. In
particular, an
overarching
framework that can
provide guidance
and structure to the
swiftly increasing
body of academic
literature is lacking.
Therefore, the
evolution of E-
commerce
performance
literature calls for a
systematic
integration and
synthesis of existing
knowledge to inform
the next wave of
academic endeavors
in this field.
An information
technology (IT) –
intensive
environment (Hua et
al., 2015) has
managed to come
needed.
Law and Jogaratnam, 2005) or the evolving harmony for business stakeholders to Therefore, this
welcome IT progressing with tourism hand in hand (Poon, 1993; Sheldon, 1997). study also
Business processes, such as business-to-business (B2B) and business-to-consumer attempts to
(B2C), are digitized (Beldona et al., 2012), and attentions have shifted from synthesize
promotion (1960s), product development and marketing research (1970s) and extant
revenue management empirical
(1980s) (Dev et al., 2010) to focus on internet- and IT-mediated business studies of E-
processes, such as consumer-to-consumer (C2C) social activities (Morosan commerce
et al., 2014). performance
While IT has increasingly permeated the industry as supported by its own and offer a
advancement and accommodating business models (Law et al., 2014), a coherent and
holistic customer/client experience is a common goal most contemporary informative
digital processes are striving for (Nyheim and Connolly, 2012). In addition, framework of
these digital processes are often characterized by usefulness, playfulness
and ease of use (Morosan and Jeong, 2008). As a result, mobile and social
networking business-related technology (Berkus, 2013), for example, is

gaining momentum over traditional personal computers (Hua et al., 2015).
Electronic commerce (E-commerce), therefore, can be considered as a
sophisticated IT process that supports achieving the holistic customer/client
experience from aspects of design, communication, delivery, fulfillment and
evaluation (Hua et al., 2015), with the underlying structural vehicle evolving
from computer reservation systems (1970s), to global distribution systems
(1980s), to internet (1990s) and to today’s organization-facilitated C2C
virtual network social activities (Buhalis, 2003; eBusiness W@tch, 2006;
Emmer et al., 1993; Morosan et al., 2014; O’Connor, 1999).
While serving as the fundamental media for most customer- and client-
related activities in today’s markets, E-commerce carries a full commercial
connotation because such a process is essentially transactional (Hua et al.,
2015). Its synergistic and accelerating interaction with hospitality and
tourism has brought fundamental changes to the industry (Buhalis and Law,
2008) and cultivated competitiveness in the business environment (UNWTO,
2001). Consequently, E-commerce performance is referred to as the
business value impact of E-commerce (Zhu and Kraemer, 2002) that
includes three types of benefits:
i) marketing and competition benefits;
ii) essential benefits that support strategy and development; and
iii) business efficiency benefits Abou(-Shouk et al., 2013).
Considering the complexity, ubiquity and criticality of E-commerce among B2B,
B2C and organization-facilitated C2C settings in the tourism and hospitality
industry (Hua et al., 2015), this study offers a review of 155 recent articles
published both in mainstream E-commerce literature and the hospitality and
tourism field. Focused on E-commerce performance, this study aims to further
our understanding, identify research gaps and suggest potential future study
directions. It should be noted that, although studies in hospitality and tourism
have approached E-commerce performance from many distinct perspectives, we
have only managed to understand the nature and impacts of E-commerce in a
scattered and fragmented fashion; an integration approach is thus much
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
53
IJCHM E-commerce performance, which should contribute to advancing
28,9 knowledge development and produce a significant long-term impact
(Maclnnis, 2011).
2. Methodology
Because no study has systematically examined the topic of E-commerce
2054 performance in the mainstream and hospitality and tourism field, conducting a
comprehensive and pertinent review of the E-commerce performance literature
requires the method of theoretical sampling (Corbin and Strauss, 2008). In this
case, articles related to E-commerce performance are considered the sample,
whose size is determined by the point of “informational redundancy”. In other
words, the article collection process will stop when finding overlaps start to
emerge from the articles repetitively and no new information is anticipated
(Lincoln and Cuba, 1985, p. 202). Such a point would suggest that both
information depth and breadth are achieved (Bowen, 2008), and a theoretical
framework governing E-commerce performance would emerge. Specifically, this

study followed the methodology by Greenhalgh et al. (2009) for the article
sample collection and used exploratory methods (browsing, asking colleagues)
before the snowballing technique (searching references of references and using
citation-tracking databases such as EBSCOhost (Hung and Law, 2011),
ScienceDirect (Tsang and Hsu, 2011) and ProQuest Business (Tsai et al., 2011).
to identify key sources to uncover the E-commerce performance themes and
prior findings systematically (Greenhalgh et al., 2009) – Greenhalgh and Peacock
(2005) show snowballing techniques are more efficient than both using formal
search strategies to electronic databases and hand searching. The article
collection process started with four of the latest articles related to E-commerce
performance published in premium academic journals – Chae et al. (2014) in
Management Information Systems Quarterly (the impact factor for 2014: 5.311;
five-year impact factor: 8.490), Hua et al. (2015) in International Journal of
Hospitality Management (the impact factor for 2014: 1.939; five-year impact
factor: 2.519), Law et al. (2014) in International Journal of Contemporary
Hospitality Management (the impact factor for 2014: 1.407; five-year impact
factor: 1.963) and Yang et al. (2015) in Technology Analysis and Strategic
Management (the impact factor for 2014: 0.942; five-year impact factor: 1.49) –
and stopped at 155 articles when thematic saturation and finding overlaps
became obvious and a theoretical framework governing E-commerce
performance emerged. In addition, the following two criteria were applied to
determine the 155 articles selected. First, only full-length research papers were
included – this criterion follows from the common belief that other scholarly
works, such as research notes and book reviews, tend to lack in originality and
contribution to knowledge development (Law et al., 2012). Second, the research
paper under consideration needed to be related to E-commerce performance as
defined by Abou-Shouk et al. (2013), Hua et al. (2015) and Zhu and Kraemer
(2002).
The collected articles were then reviewed, organized, analyzed,
synthesized and evaluated to form clear and logical conclusions with regard
to what is learned and yet to be learned (Denyer and Tranfield, 2009).
Specifically, abstracts of the selected articles were first reviewed in a
chronological order, which ensured deduction of idea evolution trends over
time (Mustak et al., 2013). Guided by abstract review insights, this study
then formed the article structural scheme, following the framework
established
by Abou-
Shouk et al.
(2013) – each
article
collected was
categorized
under the
three types
of E-
commerce
benefits:
assumption
(B) “marketing and competition benefits”; that E-
commerce will
(C) “essential benefits that support strategy and development”;
bring the
and
expected
(D) “business efficiency benefits” Abou(-Shouk et al., 2013) for benefits and
mainstream journals and hospitality and tourism journals, lead to
respectively. competitive
advantages,
It is a common practice in the hospitality and tourism field to use “mainstream
which appear
journals” to refer to generic discipline-specific journals, such as Management
at odds
Information Systems Quarterly, and to use “hospitality or tourism journals” to refer to
journals that are contextualized in hospitality and tourism. This categorization further
facilitates research gap identification, article synthesis and theoretical framework
construction.
3. E-commence performance background

A temporal pattern emerged while researchers were closely examining the

relationships between E-commerce and firm performance. E-commerce in the
1990s appeared to have offered advantages to firms with resource slack,
exhibiting characteristics of utilizing proprietary systems (Wang, 2010). For
example, E-commerce facilitated by electronic data interchange (EDI) created
competitive advantages for those that employed it (Mukhopadhyay et al., 1995).
Santhanam and Hartono (2003) showed that EDI was sustainable to a certain
extent in a multi-industry context, suggesting firm IT capabilities could improve
performance (Bharadwaj, 2000). In addition, these stand-alone and proprietary
systems are often costly or difficult for firms to communicate with their trading
partners because of a lack in powerful communicating platforms and constraints
in data (Yang et al., 2015).
Facilitated by abundant IT support, significantly reduced costs and surging
vendor interests, E-commerce started to take center stage in the 2000s (Masli et
al., 2011). Particularly noteworthy was that the extensive standardization of IT
tools (Wang, 2010) and Web-based technologies, which offer two-way, real-time
information exchange on the value chain (Zhu 2004), offered opportunities for
firms that were smaller and less resourceful to compete with those that were
bigger and more resourceful (Masli et al., 2011). As a result, E-commerce
induced large-scale transformations that were both internal and external to an
organization, exerting profound influence on both the organization and its
relationship with stakeholders (Carr, 2003).
3.1 Marketing and competition benefits of E-commerce

From a resource-based view (RBV), E-commerce capabilities can lead to performance
advantages (Nevo and Wade, 2010), such as differentiation or cost advantages
(Bharadwaj, 2000) because unique organizational capabilities can be derived from
proper resources utilization and allocation within a firm (Chae et al., 2014) with
support from other resources such as productive organization cultures (Bharadwaj,
2000). However, firm capabilities are dynamic and continuously evolving to “adapt to
rapidly changing environments” (Elsenhardt and Martin, 2000; Teece et al., 1997),
which led researchers to a dynamic capabilities perspective (DCP) to understand how
a firm gains competitive advantages in today’s dynamic marketplaces (Chae et al.,
2014). Superior performance can result from a firm’s strong E-commerce orientation
because such a firm can be more agile to adapt to dynamic business environments
(Sambamurthy et al., 2003). However, this stream of research has built in an
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
55
IJCHM with the inconclusive results from recent empirical studies about the
28,9 performance impact of E-commerce (Chae et al., 2014) – not all E-commerce
endeavors materialize into resources that are “rare, inimitable, non-
substitutable, and valuable” (RINV) and thus create competitive edges and
command economic rent (Barney, 1992; Amit and Schoemaker, 1993).
Consequently, skepticism started to arise as to whether internet-based
2056 electronic businesses will continue to benefit from the economic impact of IT
(Tallon and Kraemer, 2007; Zhu and Kraemer, 2002), and more general
approaches are called for to better understand potential benefits of E-
commerce in the business setting (Chathoth, 2007).
Several benefits of E-commerce that facilitate gaining a marketing and
competitive edge, however, have been documented. For example, E-
commerce can ensure a stable survival in the ever-changing and
competitive business environment (Stansfield and Grant, 2003). In
particular, E-commerce can improve competitiveness of small and medium
enterprises (SMEs) because it can enable efficient collaboration

(Kvainauskaite et al., 2005). It is widely recognized that E-commerce, which
leads to improved organization image and reputation of SMEs in the global
market (Ayeh, 2006; Beckinsale and Levy, 2004; Collins et al., 2003;
Kajogbola, 2004; Scarborough and Zimmerer, 2003) and technology
diffusion result in SMEs’ improved competitive positions (Beatty et al., 2001;
Daniel and Wilson, 2002; Kartiwi and MacGregor, 2007; MacGregor, 2004).
Moreover, E-commerce can improve productivity and realize economies of
scale, leading to better operating efficiency and performance (Beekhuyzen et al.,
2005; Harindranath et al., 2008; Migiro and Ocholla, 2005; Simpson and
Docherty, 2004). By shortening response times to enquiries and by tracking and
improving customer satisfaction, E-commerce can lead to more repeat
businesses and enhanced customer loyalty (Dyerson and Harindranath, 2007;
Quayle, 2002; Saffu and Walker, 2008; Teo et al., 2009). An “e-competitive
transformation” (Straub and Klein, 2001), if appropriately used, can gain firms
competitive advantage over both online and offline competitors (Dyerson and
Harindranath, 2007; Harindranath et al., 2008; Poon and Joseph, 2001).
3.2 Essential benefits of E-commerce that support strategy and

development
From a strategic and development perspective, turning investment in E-
commerce into firm capabilities that can generate sustainable competitive
advantages and outperformance is playing an increasingly critical role for
business organizations. On the one hand, physical IT assets such as
equipment and even some software are often easily imitable and thus tend
to bring only temporary outperformance (Teece et al., 1997). On the other
hand, the intangible IT assets and capabilities, once successfully integrated
in a business, can bring sustainable superior performance (Christensen and
Overdorf, 2000) because they are considered RINV resources, characterized
by tacitness, idiosyncrasy and often a long development time window
(Dierickx and Cool, 1989), and shielded by factors like social embeddedness
and causal ambiguity from competition imitation (Dierickx and Cool, 1989).
Therefore, studies in the business literature predominantly focus on adoption
and related issues (Wang et al., 2013). Particular attention was directed to
strategic and development implications that follow from E-commerce adoption.
For example, E-commerce adoption was shown to improve financial growth at
both the top
and bottom
lines of
organizations
(Karagozoglu
and Lindell,
2004; Straub
and Klein,
2001)
technology
and enhance a firm’s capability to penetrate international markets through capabilities
easy information access on a global scale (Jin, 2007; MacGregor, 2004; include
Zheng et al., 2004). By improving communications with customers, E- business
commerce strengthens firms’ capability of service customization (Pease and
Rowe, 2005). By improving communications with customers, suppliers and
partners (Azam, 2007; Karanasios, 2008; Simpson and Docherty, 2004) and
efficient use of technology to serve customers, E-commerce leads to
customer satisfaction and loyalty (APEC, 1999), further advancing effective
relationship building. In addition, because E-commerce technology provides
support to organizations with increased and easier access to information
and knowledge, E-commerce adoption can lead to improving decision-
making processes and management productivity (APEC, 1999), as well as
better systems, frameworks and methodologies, to make strategic and
functional decisions (Grandon and Pearson, 2004; Kajogbola, 2004).
In general, IT infrastructure and other physical IT assets, such as internet and
associated technologies, do not create value without being integrated with other
organizational resources (Mata et al., 1995), which are often difficult or costly to
transfer when combined (Amit and Schoemaker, 1993). Only by complementing
tangible technology assets, intangible IT assets such as human and business IT
resources appear to explain performance variances among companies (Siqueira
and Fleury, 2011; Li et al., 2014). Owning IT resources that best complement
technology in attaining innovation appears to be the most powerful way to
create and secure sustainable competitive advantages against competition
imitation (Feeny and Willcocks, 1998). In the E-commerce context, firms do not
have the same E-commerce capabilities; even if they do, they can benefit only
from Web-based technologies that are integrated into organizations and produce
“sustainable resource complementarity” (Lederer et al., 2001; Straub and Klein,
2001; Zhu, 2004).
3.3 Business efficiency benefits of E-commerce

Successful interaction between E-commerce and four intangible assets, i.e.
“alignment of E-commerce with business strategy, financial resources, user
involvement, and external relationship management” (Bharadwaj, 2000; Thong,
2001; Sauer and Willcocks, 2002; Bruneel et al., 2012; Li et al., 2014; Yang et al.,
2015), could result in outperformance (Yang et al., 2015) because of increased
business efficiency (Bharadwaj, 2000) and barriers to imitation (Amit and
Schoemaker, 1993).
Therefore, E-commerce can lead to operations and internal efficiency
improvements (Collins et al., 2003; Dyerson and Harindranath, 2007; Karanasios,
2008; Kuan and Chau, 2001; Teo et al., 2009), which result in better business
performance (Jin, 2007; Poon and Joseph, 2001) and improved supply chain
efficiency (Quayle, 2002). A number of reasons have been explored to
understand business efficiency benefits of E-commerce. For example, adopting E-
commerce technology was cited to improve internal efficiency by enabling better
order processing, employment growth, inventory control, online staff
recruitment, accountability and staff satisfaction (Dyerson and Harindranath,
2007; Harindranath et al., 2008; Kajogbola, 2004; Quayle, 2002), in addition to
supporting improved cooperative efficiency of supply chain management (Daniel
and Wilson, 2002; MacGregor, 2004; Saffu and Walker, 2008). In particular, E-
commerce is recognized for facilitating collaboration and planning among supply
chain partners (Quayle, 2002). Additional benefits of creating E-commerce
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
57
IJCHM knowledge enhancement (APEC, 1999; The e-Regions Trust, 2006), internal
28,9 knowledge sharing improvement (Daniel and Wilson, 2002) and online
transaction execution enhancement (Álvarez et al., 2009; Raymond, 2001),
which includes both processes of order taking and tracking (Ayeh, 2006;
Wesrthner and Klein, 1999).
In short, E-commerce can be a resource that generates competitive value
2058 only when combined with tacit, specific business resources and skills (Yang
et al., 2015). The literature of its performance, as illustrated above, can be
categorized into a production-economics-based stream and a process-
oriented stream (Barua and Mukhopadhyay, 2000) – production functions
are used in the first stream of research to explore the input– output
relationships driven by IT, and, in general, positive empirical evidence was
found between IT and productivity (Chan, 2000); performance analyses are
conducted in the second stream of research to examine the effects of
investment on intermediate operational performance, as well as financial
performance at a higher level (Barua and Mukhopadhyay, 2000).
4. E-commerce performance in hospitality and tourism

E-commerce evolved uniquely in the hospitality and tourism industry over time
(Gaffney, 2013). Early E-commerce practices started in the 1960s, when the
hospitality and tourism industry began to rely on global distribution systems
(Thakran and Verma, 2013). In particular, the programs of frequent flyers in the
1980s provided airlines with detailed customer information (Kim et al., 2009)
and led to successful customer relations management and relationship
marketing, setting a foundation for modern E-commerce (Chen et al., 2002).
While the internet emergence in the 1990s took E-commerce to a new level,
where B2B and B2C relationships were more interactive (Kim et al., 2009), the
year of 2000 observed E-commerce stepping into the SoLoMo era, where social,
location and mobile played dominating roles (Thakran and Verma, 2013).
Starting from 2012 and featuring extensive consumer behavior intermixing
across purchase, channel and device alternatives (Thakran and Verma, 2013),
the hybrid era witnessed rapid evolvement of E-commerce into the mobile
commerce domain (Gaffney, 2013) and offered great opportunities of research
to complement prior studies that addressed technology progress and marketing
evolution (Berkus, 2013; Dev et al., 2010). For example, the majority of hotel
guests tend to bring at least two devices with them and consider paramount the
freedom to use their own devices and control their entertainment (Watkins,
2013). Some cruise lines and upscale hotels, such as Oceania and Kimpton,
have started to make iPads available for guest use and convenience (Hua et al.,
2015).
4.1 Marketing and competition benefits of E-commerce

The resource-based view (Chae et al., 2014; Nevo and Wade, 2010) and the theory of
dynamic capabilities (Elsenhardt and Martin, 2000; Teece et al., 1997) appear to have
served as theoretical foundations for studies related to E-commerce performance in
the hospitality industry. Relationships describing E-commerce and hotel performance
were proposed (Chathoth, 2007); IT-mediated tasks and performance measures were
tested, but, similar to studies in the mainstream, were found to have inconclusive
results. IT, for instance, failed to exhibit a direct influence on company performance;
however, firm capabilities that take full advantage of informalization and networking
lead to improved performance (Sigala, 2003). When mediated by employee
outcomes, E-
commerce that
Consumers
supports customer service was found to influence hotel performance were found to
(Cohen and Olsen, 2013). Recently, Hua et al. (2015) found that E- accept online
commerce expenses in general influence rooms’ revenue positively and purchasing,
significantly, although such expenses appear to have distinct impacts and perceived
across different chain scale types. website
The theory of dynamic capabilities suggests that E-commerce can be control was
considered as a dynamic capability (Hua et al., 2015). Utilizing knowledge, shown to be
routines, simple rules and analytic processes to create customer value, positively
organizations exhibit characteristics of dynamic capabilities, as they use
appropriate E-commerce support to keep improving their resource allocation and
performance (Wheeler, 2002). Early examples can be found in the tourism and
hospitality industry in the 1960s when American Airlines developed SABRE, a
basic but automated E-commerce system, to improve data processing efficiency
of traveler bookings and data. SABRE grew quickly to serve over 10,000 travel
agents all over the world and later on became the biggest private real-time data-
processing system at one time. SABRE Technology Group introduced

Travelocity.com in 1996 and created a new market featuring optimization of
consumer and inventory interaction and aggregation such as search and
customization, changing the landscape of E-commerce (Mamaghani, 2009). The
current shift in guests’ increasing use of mobile devices (Gaffney, 2013) also
highlights some areas where dynamic capabilities can be carefully developed in
the mobile dimension, such as trip planning, real-time information accessing,
information sharing and problem solving (Wang et al., 2012). Geo-based
technology can significantly create differentiating features by offering timely
geographical information and associated services (Tussyadiah and Zach, 2012).
While technology advancement has brought into being virtual reality (VR), VR
appears unlikely to replace a real visit in the near future (Guttentag, 2010).
Understanding how people seek, process and use travel information in both
online and offline contexts can provide managers with benchmarking information
to improve effectiveness of online marketing strategies (Ho et al., 2012),
particularly given responses toward Web advertising were shown to positively
impact website attitude, which leads to improved attitudes about brands, and in
turn to increased intentions to purchase (Hwang et al., 2011). For example,
Xiang and Gretzel (2010) studied the appearance frequency of social media
websites in travel-related information search results, considering that social
media is a critical source of travel information. They found that a large
proportion of search results on travel information came from social media
websites and that search engines have become increasingly sensitive to social
media content. Later on, Hays et al. (2013) showed that social media can
function as a marketing tool for destinations. In addition, given that online
marketing effectiveness positively influences tourism dominance (Woodside et
al., 2011), websites play a critical role in promoting destination image (Jeong et
al., 2012). Furthermore, because both internal and external factors are essential
in determining use of electronic marketing (El-Gohary, 2012), different
approaches should be used to target potential customers to improve marketing
effectiveness in an online environment (Luna-Nevarez and Hyman, 2012) to
improve E-commerce performance.
Online travel communities hosted by business organizations also appear to
behave as effective marketing platforms, as consumers’ intention to participate
in such communities was found to positively impact their intention to use and
recommend products of the host business organizations (Casaló et al., 2010).
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
59
IJCHM related to purchase intention (Liang and Lim, 2011). In addition, online
28,9 purchasing intention appears to be positively related to user innovativeness and
ease of transaction process (San Martín and Herrero, 2012). Interestingly, taking
souvenirs’ purchasing as an example, it may be a good idea to complete a
transaction the traditional way in a local store before disclosing available online
purchasing options, as purchase limitation appears to increase initial purchase
2060 intent regardless of item type (Abendroth, 2011). Along the line of consumers’
repeat purchase and loyalty in online contexts, Kim et al. (2013a, 2013b, 2013c)
showed that navigation functionality, transaction security and cost effectiveness
significantly influence trust, which then positively affect intention to repurchase
online; Llach et al. (2013) offered empirical evidence that quality, decomposed
into the functional and hedonic quality, positively and significantly influences
loyalty through perceived value in an online purchasing context of the airline
industry. And more experience using E-commerce tools appears to positively
affect repurchase intention (Oh et al., 2009). Therefore, it appears that E-
commerce platforms that are user friendly, functional, easy and fun to use and
safe can significantly improve E-commerce performance. In particular, proper E-
commerce decisions can result in competitive advantages (Bilgihan et al., 2011).
4.2 Essential benefits of E-commerce that support strategy and

development
Tracing a similar line of evolvement, hospitality and tourism studies on E-
commerce are predominantly IT adoption-oriented (Wang et al., 2013). Yet,
academic attention has been broadened to address institutional
employment of IT (Herrero and San Martín, 2012; Kim et al., 2008, 2009),
while keeping pace with the increasing number of industry applications of E-
commerce (Law et al., 2014).
IT adoption studies are critical in understanding how to improve E-commerce
performance, as they pinpoint areas of interest that tend to have significant
strategic implications and produce timely and practical improvements for
business development. For example, strengthening market position by
maintaining good relationships with current business partners and securing new
ones is a widely recognized E-commerce benefit (Kim, 2006). E-commerce has
been shown to help SMEs in the tourism industry with collecting and distributing
information on a global scale; by offering easy information access to tourism
services and products, E-commerce was shown to improve stakeholder relations
(Bourgouin, 2002; Kim, 2006). Social media is often considered as a critical
source of information by international tourists to seek solutions for a variety of
problems (Schroeder et al. 2013). And guest-empowerment technologies appear
to be valuable for hotel room services, as shown by Schrier et al. (2010) that
hotel guests consider on-demand services the most attractive features.
In particular, understanding the determinants of IT adoption facilitates
organizations to optimize their choice of IT systems (Tarcan and Varol, 2010),
which, in turn, could lead to improved E-commerce performance. For instance,
tourists who are more familiar with the internet and exhibit high innovativeness
tend to use podcast tours in museums for an enhanced experience (Kang and
Gretzel, 2012). Consumers in restaurants are prepared to use biometric systems
when they are perceived as useful and secure (Morosan, 2011). Potential hotel
guests tend to use mobile hotel reservation services if they offer high
information and system quality (Wang and Wang, 2010). Because customers rely
on E-
commerce
platforms to
facilitate their
decision-
making
role in
process (Law et al., 2014), websites focus on the needs of customers operating and
instead of those of investors (Escobar-Rodríguez and Carvajal-Trujillo, strategic
2013). management
Several studies tried to understand customer needs in an attempt to (Law et al.,
improve E-commerce performance from an online textual content 2014).
perspective and started to explore potentials of text mining in the tourism
and hospitality industry. For example, positive online reviews (eWOM) were
found to significantly improve booking intention and trust (Sparks and
Browning, 2011), intention to travel, tourist attitude and destination image
(Jalilvand et al., 2012). A selection of keywords appears to dominate
destination image because tourists rely on these words as anchors to find
more destination information online (Pan and Li, 2011). In addition, travel
blogs were shown to convey information about bloggers’ travel
expectations, values and beliefs (Banyai and Havitz, 2013). Therefore,
managers can analyze online textural information to understand tourists’
thoughts and work closely with identified online ambassadors (Rong et al.,
2009) to form effective marketing strategies. It was also shown that a price-
network size schedule featuring reducing price with increasing network size
could work to attract potential customers to make timely purchases (Xiong
and Hu, 2010).
Studies addressing the relationship between tourism suppliers and
intermediaries are gaining momentum in the literature. For example, the loyalty
of an airline’s consumers and competition of its own business website with online
travel agency (OTA) platforms appear to determine whether the airline uses OTA
platforms to sell tickets (Koo et al., 2011). Online wholesalers are perceived in
some regions of the USA as partners of hotel operators, with acknowledged
benefits including global distribution of products and services and aggressive
media advertising that increase hotel visibility and exposure (Myung et al.,
2009). And online communication was found to improve online reservation and
e-ticket procurement (Andreu et al., 2010). E-commerce technology applications
were also found to be critical in increasing loyalty and satisfaction of customers
when supplier and travel agencies are concerned (Ruiz-Molina et al., 2010) and
leading to improved organization image and reputation of SMEs in the global
market (Liao and Par, 2006). Note, though, that the distribution channel structure
is affected by the market structure during the process of restructuring caused by
E-commerce technologies (Berne et al., 2012). Of particular importance is that IT
steering committees in hotels appear to exert a significant and positive impact
on IT integration (Cobanoglu et al., 2013). In conjunction with the large number
of benefits that E-commerce can bring (Abou-Shouk et al., 2013), the committee
is thus considered strategically critical to hotel business success because of its
influence in achieving IT strategic goals and gaining a competitive edge over
competition (Cobanoglu et al., 2013); while, a recent study examined impacts of
E-commerce benefits on E-commerce adoption (Abou-Shouk et al., 2013),
instead of focusing on E-commerce benefits as a result of E-commerce adoption
(Bigné-Alcañiz et al., 2009). Moreover, studies started to look at how hotels
assess E-commerce technology investment, and preliminary results suggest that
centrally managed hotels often utilize more assessment tools based on both
financial and nonfinancial information, such as net present value and technical
requirement fit, than do locally managed hotels (Karadag et al., 2009). In short,
E-commerce has started to play an increasingly more important and strategic
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
61
IJCHM 4.3 Business efficiency benefits of E-commerce
28,9 Business efficiency benefits derived from E-commerce can usually be classified
into external and internal benefits. For example, transaction cost reduction was
proposed as a potential internal benefit from using E-commerce technology in a
hotel service operation context – the internal benefit of transaction cost
reduction was found to be positive, yet statistically insignificant, suggesting
2062 resource lack in implementing E-commerce in the hotel industry (Chathoth and
Law, 2011). Using biometric technologies in the hospitality and tourism industry
was found to offer potential external benefits such as consumer convenience and
increase operating efficiency and security which are generally considered
internal to a company (Mills et al., 2010). The concern, though, is how to securely
protect consumers’ biometric data because they are critical private information
(Mills et al., 2010). Community preferences can also be influenced by E-
commerce technologies such as an audience response system for tourism
development (Keske and Smutko, 2010). By facilitating collaboration and
planning among partners of supply chains (Quayle, 2002), E-commerce was

shown to lead to better performance for travel agencies (Heung, 2003). For
SMEs, E-commerce technology applications could lead to improved distribution
channels and reduced operation costs (Abou-Shouk et al., 2013)
Overall, E-commerce appears instrumental in increasing productivity and
operating
efficiency, which in turn leads to the benefit of economies of scale Buhalis,(
2003; Karanasios and Burgess, 2008). Furthermore, indirectly, marketing
benefits derived from E-commerce appear to also bring crucial benefits of
operational efficiency, i.e. sales increase and/or cost reduction (Abou-Shouk
et al., 2013).
*
Discussions and research gaps of e-commerce
performance in hospitality and tourism
E-commerce is taking an increasingly more important role in serving
as the fundamental platform for all business transactions and non-
business interactions between all parties of interests – it provides
critical support to all three domains of an organization:
← the organization’s external market environment;
← the organization’s internal optimal performance; and
← flows between the organization and its market environment.
Understanding the key influencers of the performance of this fundamental
platform and its core dimensions, therefore, is critical for modern business
survival and success.
Based on the findings of this study, the E-commerce performance
framework can be derived as follows in Figure 1. The key dimensions and
influencers of E-commerce performance are identified and relationships
depicted. Specifically, E-commerce performance was found to exhibit three
key dimensions that encompass competitiveness and marketing, strategic
development and efficiency (Abou-Shouk et al., 2013). It appears that E-
commerce performance is influenced by market E-commerce environment,
organization E-commerce environment and the dynamic and interactive
relationships in between (Cohen and Olsen, 2013; Ho et al., 2012; Hua et
al., 2015; Law et al., 2014; Wang et al., 2013).
The
market E-
commerce
environment,
primarily the
E-commerce
infrastructure
and
superstructur
e external to
an
organization
that define,
support and
communicate
E-
commerce
performanc
e
206
3
Figure
1.
The E-
commerce
performance
framework
organizational needs and wants for optimal results and superior performance, is
considered as both stimulating and constraining the organization’s potential for
innovation and better performance. It encompasses suppliers, investors, IT
developers, competitors, consumers and other stakeholders (Carr, 2003) and bears
heavy influence on E-commerce performance by interacting with the organization
through a variety of channels and means, such as the fundamental supply-and-
demand relationship between the organization and suppliers (Wang, 2010),
consumers (Gaffney, 2013) and competitors (Masli et al., 2011), asymmetrical
environmental and social responsibility concerns between the organization and
investors (Escobar-Rodríguez and Carvajal-Trujillo, 2013), and the balance of capital
budget control and innovation push between the organization and IT developers
(Gaston and Botts, 2013).
The organization E-commerce environment, which supports information
synthesis into ideas and solutions to developing and/or existing problems and
issues, is considered as influencing organizational effectiveness in response to
its market environment. These include complex and dynamic interactions
between organizational core competencies (Lederer et al., 2001; Straub and
Klein, 2001; Zhu, 2004), products and services (Liang and Lim, 2011) and
supporting tangible and intangible assets such as internet and associated
technologies (Mata et al., 1995), human and business IT resources (Siqueira and
Fleury, 2011; Li et al., 2014), dynamic capabilities developed in the mobile
dimension – e.g. trip planning, real-time information accessing, information
sharing and problem-solving (Wang et al., 2012) – and appropriate E-commerce
support to continuously improve their resource allocation and performance
(Wheeler, 2002).
O lting from bordering internal and external E-commerce environments between the
p organization and its market, given a set of internal characteristics, can be both
p arousing and stifling the organization’s response. Therefore, the market and
o organization E-commerce environments further interact and deliver a separate, as
r well as combined influence upon E-commerce performance of an organization (Carr,
t 2003; Cohen and Olsen, 2013; Escobar-Rodríguez and Carvajal-Trujillo,
u
n
i
t
i
e
s
a
n
d
b
a
r
r
i
e
r
s
t
o
c
o
m
m
u
n
i
c
a
t
i
o
n
r
e
s
u
IJCHM 2013; Gaffney, 2013; Ho et al., 2012; Hua et al., 2015; Law et al., 2014; Wang et
28,9 al., 2013), resulting in the dynamic model of E-commerce performance depicted
in Figure 1.
Prior studies that addressed issues of E-commerce performance in the
hospitality and tourism field, although fragmented, have made significant
contributions to the literature by providing both theoretical frameworks and
2064 empirical evidence for key relationships as depicted in Figure 1 (Abou-Shouk et
al., 2013; Cohen and Olsen, 2013; Ho et al., 2012; Hua et al., 2015; Law et al.,
2014; Wang et al., 2013). It is important to note, however, that E-commerce-
supported business interactions could significantly contribute to all three aspects
of E-commerce performance, i.e. competitiveness and marketing, strategic
development and efficiency, which, at the same time, are critical considerations
for business and strategic decisions to deploy E-commerce technologies and
infrastructure in the first place (Abou-Shouk et al., 2013).
Researchers in hospitality and tourism appear to have focused primarily on two
stakeholder groups when addressing issues and questions related to E-commerce

performance, i.e. consumers and the organization of interest (particularly, managers
and employees within the organization). Many other stakeholders received limited
attention and disproportionately much less research interest, for example, regulators,
researchers, media, environmentalists, suppliers and investors. This phenomenon
appears to reflect the early stage of E-commerce performance research in hospitality
and tourism, as consumers and the organization of interest tend to be the most
obvious to focus on from an academic perspective. Meanwhile, this phenomenon
seems to also reflect the present initial stage of E-commerce life cycle in hospitality
and tourism, as organizations typically have to prioritize their stakeholders when
initiating strategic and significant new endeavors to ensure those that are most
impactful for organizational survival are satisfied (Chua et al., 2005; Jawahar and
McLaughlin, 2001).
Studies related to E-commerce performance in hospitality and tourism appear
to have focused primarily on social coordination (e.g. social media) and
communication aspects (e.g. web interface), with relatively little attention paid
to transaction-based services such as online payment. Issues of location- and
situation-sensitivity appear to be also understudied. In addition, likely due to
novelty and the explosive speed of popularity gains of mobile apps, only a few
studies in hospitality and tourism focused on services and products in the
mobile app domain. Relative to the large number of adoption studies, few
studies explored E-commerce performance issues of using mobile services. For
example, in-store mobile services, such as the apps created to enhance and
support guest hotel experiences, have still not been well studied and understood
from an impact perspective.
Moreover, despite the surging interest in understanding the E-commerce
performance impact of OTAs (e.g. Priceline and Expedia), generic search engines
(e.g. Google and Bing) and social media platforms (e.g. Facebook and Twitter),
studies in hospitality and tourism appear to only slowly catch up with emphasis
still placed upon various versions of IT adoption frameworks and product
attributes. The critical link between these E-commerce platforms and
performance is still missing or at best unclear. For example, the recent cry from
hoteliers about OTAs burdening hotel operation by heavy commissions has
drawn considerable attention (Hotel News Now, 2015); however, the academic
community has yet to offer a study explaining and tackling the underlying
mechanism
and its
intrinsic
influence on
performance.
focused on
Another research area that appears lacking significantly in hospitality consumers
and tourism concerns E-commerce-driven decision-making, which tends and the
to bear significant direct and indirect influence on performance. For
example, even though the decision-making context has shifted from
quick and single-device-based to more integrated and multiple-
synchronized-device-based (Yadav and Pavlou, 2013), few studies in
hospitality and tourism have approached and addressed this area from
either a theoretical or empirical perspective.
From a research design point of view, extant research related to E-commerce
performance in hospitality and tourism predominantly uses surveys to collect
data; relatively few studies are conducted by secondary data analysis or
experimental design. Two facts are noteworthy, however. On the one hand, the
emergence and explosion of Big Data provided much more powerful analytical
possibilities than ever before. All stakeholders of hospitality and tourism
organizations can use Big Data to improve predictions and decision-making
processes. On the other hand, experimental design is well known for its
advantages on controlling for confounding effects and isolating causality.
Therefore, there appears to be more room for a diverse group of studies that use
different research designs/methodologies to address both theoretical and
empirical questions related to E-commerce performance.
6. Theoretical implications
Researchers have started to address the fundamental question of whether
E-commerce contributes to superior performance. For example, Kim et al.
(2009) examined drivers of individual E-commerce components, while Hua
et al. (2015) tested the overall impact of E-commerce on performance. It
appears that the latest findings suggest that using E-commerce for strategic
purposes contributes to achieving long-term competitive advantage; in
particular, E-commerce appears to be instrumental in improving customer
satisfaction, managerial efficiency, marketing efficacy and financial
performance (Hua et al., 2015; Kim et al., 2009).
In addition, broader, as well as interdisciplinary, implications follow from
studies of E-commerce performance in IT, performance analysis, marketing and
strategic management, providing foundations for future studies examining
consequences of strategic choices. For example, Hua et al. (2015) made a
contribution to the marketing literature by specifically showing the manner in
which combining various components of marketing strategies to form an overall
strategy can help organizations increase revenue. On the other hand, specific
manners in which E-commerce contributes to performance were also revealed by
recent studies, shedding light on superior managerial decision-making
possibilities. For example, even though system beliefs or desire of control tend to
drive E-commerce deployment (Cohen and Olsen, 2013), E-commerce’s
contribution to room revenue provides an opportunity for managers to better
understand potential benefits of E-commerce adoption and, consequently, to
make superior decisions (Abou-Shouk et al., 2013; Hua et al., 2015).
Further fundamental and theoretical questions will likely be explored based on
the proposed framework of E-commerce performance, identified research gaps
and new research directions. For example, different aspects of the framework
(Figure 1) will likely be brought under scrutiny to understand both their individual
roles and mutual relationships against the backdrop of improving E-commerce
performance. Scholarly endeavors on E-commerce performance have primarily
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
65
IJCHM organization(s) of interest in hospitality and tourism largely because E-
28,9 commerce still has a long way to reach its mature stage – maturity tends to
be positively associated with the resources that can be used to satisfy more
stakeholders (Jawahar and McLaughlinn, 2001). Along this line of reasoning,
more studies will attempt to focus on many other stakeholders, such as
governments, activists, environmentalists, technology acquisition
2066 specialists and venture capitalists, to understand their roles with regards to
and impacts on E-commerce performance.
7. Practical implications
Driven by the need to create and maintain competitive advantages,
E-commerce business models have evolved over time to adapt to
the dynamic market and to better produce and implement E-
commerce protocols within a “very competitive operating
environment” (Hua et al., 2015).
Because E-commerce has become a necessity (Hua et al., 2015), E-commerce

performance is critical to business survival and success. Web-based technologies
and an E-commerce environment have significantly transformed the competitive
landscape in the hospitality and tourism industry. With E-commerce’s global
reach, companies can target a potential customer market that covers the entire
world. The non-stop availability feature enables transaction opportunities 24/7
and offers unparalleled business information at all levels (e.g. B2B, B2C and C2C
levels) (Morosan, 2014). The proportion of global population that has internet
access continues to increase from 15 per cent in 2005 to 48 per cent in 2014
(Internet Live Stats, 2014), which, when properly explored, could offer
companies great opportunities to improve business efficiency in a number of
areas, such as customer engagement, acquisition, conversion and E-commerce
performance metrics, such as order value, order frequency, revenues per visit
and order predictability (Gaffney, 2013). Brand awareness, therefore, follows,
and online bookings could increase (Hua et al., 2015).
The criticality of E-commerce performance has led IT developers to see the
need and demand of offering extensive E-commerce services, such as search
engine optimization, social media, email marketing, paid search marketing and
customizable reservation systems (Gaston and Botts, 2013), which in turn create
more opportunities and vehicles for businesses to improve their performance. It
has led businesses to form specialized task forces to ensure execution and
improvement of their E-commerce strategies (Hua et al., 2015). For instance, by
forming a dedicated E-commerce team at the global level, Hilton Worldwide
helps its properties to manage and improve their websites and develops online
performance analytics to ensure their strong online presence (Hilton, 2014).
Marriott has also started an E-commerce global activation summit.
The essentiality of E-commerce performance has led businesses to focus on
further understanding E-commerce behaviors of customers (Hua et al., 2015), as
they evolve continuously with increasing complexity along dynamic changes of
business models and practices (Pan et al., 2013). For example, consumers used
to prefer websites of large online travel agencies with familiar brands, typically
featuring ease of use, playfulness and usefulness, around 2005 (Morosan and
Jeong, 2008). But now they are using more mobile devices (Gaffney, 2013) for
inventory access through “metamediaries” (Gaston and Botts, 2013), and
paying particular attention to automation of services and devices (SmartBrief
Media Services, 2013). Moreover, consumer behaviors appear to show more
variations
across
different E-
commerce
platforms,
exhibiting
characteristic
s of deeper
services and
fragmentation and a strong influence of search technologies on consumers’ social media
decision-making process (Paraskevas et al., 2011). Therefore, businesses, marketing. For
as well as scholars, in the tourism and hospitality industry should consider it example,
critical to introduce and assess E-commerce measures (Perlovich, 2013) social media
while strategizing business performance plans comprehensively (Boyd and integration
Bilegan, 2003). into traditional
Last, but not least, the cohesive framework of E-commerce performance distribution
derived from this study offers industry practitioners an opportunity to have a channels will
holistic understanding on determinants of E-commerce performance. In addition,
the extensive content analysis and synthesization update practitioners on
current practices to improve E-commerce performance. Findings of this study
point industry practitioners to directions that can lead to better E-commerce
performance. For example, research on the mobile dimension appears to have
lagged behind the industry practices significantly in the hospitality and tourism
field. Potential directions to explore matching app interface and function, as well
as achieving competitive advantages by targeting niche markets, could also be

of great interests to practitioners.
8. Future research
Because of currently fragmented research on E-commerce performance in
hospitality and tourism, a stream of future research developing coherent themes
and frameworks is more likely to follow, particularly as E-commerce permeates
businesses inside and out. From an E-commerce performance perspective,
studies are likely to explore the further impact of relationships between all three
domains of an organization:
*
the organization’s external market environment;
*
the organization’s internal optimal performance; and
*
flows between the organization and its market environment.
More in-depth theoretical building and empirical tests are expected
to focus on each of these three domains as well.
As the E-commerce life cycle evolves past the initial stage and into maturity,
organizations will adapt and evolve accordingly and accumulate more resources
to satisfy more stakeholders (Jawahar and McLaughlin, 2001). As a result,
scholarly endeavors on E-commerce performance will also evolve from focusing
primarily on consumers and the organization(s) of interest to encompassing
more constituents. For example, in addressing issues and questions related to E-
commerce performance, corporate governance and market function can be
further explored to seek answers to questions related to criminals (e.g.
unauthorized hacking to steal personal information such as social security
number), as well as competitors; E-commerce infrastructure can be carefully
examined to address and highlight the connection between researchers/
universities and the industry; and a variety of other themes such as acquisition,
evaluation and public governance can be assessed to uncover solutions
challenging to other parties of interest such as technology acquisition
specialists, venture capitalists, activists and governments (Chua et al., 2005).
More targeted and specialized studies will focus on specific aspects of E-
commerce performance with in-depth analyses, considering current studies in
hospitality and tourism tend to be exploratory within the context of certain
service category, such as mobile E-commerce, online marketing, online data
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
67
IJCHM continue to increase (Xiang and Gretzel, 2010). In particular,
28,9 Facebook developed search tools that double as a distribution
channel, providing a natural context to further understand how to
improve E-commerce performance from marketing and revenue
management perspectives.
Study scopes will expand further beyond foundational themes of defining,
2068 describing and exploring into modeling challenges and assessing determinants
of E-commerce performance. Success criteria and evaluation systems would be
critical issues to explore in the future, considering the motto of what gets
measured gets done. Particularly, business models, systems and processes
would draw more attention from both the industry and academia, as they are
potential drivers of E-commerce performance. Although further examining
outcomes of IT adoption will provide an important complementary view and
make a critical contribution to the literature on the institutional adoption of IT
(Hua et al., 2015), future studies will likely cover more diverse topics far beyond
adoption as E-commerce evolves to incorporate more variations and dynamics

in hospitality and tourism organizations and markets.
Research design is likely to be more diverse and inclusive moving
forward on the topics related to E-commerce performance in hospitality
and tourism, evolving to produce more studies based on secondary data
analysis and experimental design. Considering that Big Data has gained
significant momentum because of its vast information content,
researchers are expected to discover more research interests utilizing
secondary data to test theories and develop new frameworks. At the
same time, given the advantages of confounding effects control and
causality isolation and identification, researchers are also likely to
explore more experimental design possibilities.
Mobile apps will come under extensive study in the years to come, as there
appears to be a significant gap between what is being used in the marketplace
and what is under research. This phenomenon is probably because of the
explosive increase of mobile apps in a very short time window, rendering many
studies only “in progress”. Specialized apps that target niche markets appear to
form competitive advantages and deliver better performance, such as
TripAdvisor, Expedia and pPriceline, while branded apps in hospitality and
tourism appear to enjoy competitive advantages over non-branded ones. More
studies are likely emerging to further understand reasons beyond the enhanced
user experience from customized app interface and functions to match services
with the devices in hospitality and tourism.
The mixed findings of extant literature on E-commerce performance in
hospitality and tourism will be explored further, as such inconclusiveness could
result from a number of possibilities, such as inconsistent methodologies,
diverse research variable choices, measurement errors and lack of a uniform
theoretical framework. In addition, competing theories should also be further
explored to understand impacts of E-commerce against the specific backdrop of
a study. For example, it would be surprising to observe that the efforts exerted to
developing “storefront websites or mobile applications (apps)” often fail to
improve consumer adoption that brings in systematically more direct sales,
without understanding that consumers may bypass direct channels for relative
benefits, such as ease of use and playfulness (Morosan and Jeong, 2008). As a
result, measurement issues of effectiveness and specific causes of E-commerce
performance need to be further addressed (Hua et al., 2015). Modeling impact of
E-commerce
can also be
extended into
intangible
dimensions to
incorporate
with articles
more performance measures such as product or service quality, customer value- of the
added and customer loyalty and to understand which intangible benefits literature
manifest themselves better from E-commerce infrastructure and superstructure. review
Further, E-commerce orientation could be further explored, when confounding nature –
factors are controlled for, to understand whether systematic company journal and
performance could follow (Hua et al., 2015). Future studies will probably explore publication
E-commerce performance further from an external perspective, i.e. look into selection,
impacts of external markets and political environment, against a backdrop of timing and
potential industry differences. relevance
criteria and
the
9. Conclusion and limitations researcher’
Although much progress has been made in hospitality and tourism s decision
addressing E-commerce performance, extant research remains and
fragmented, and many gaps need to be bridged. By reviewing and interpretati
synthesizing 155 recent articles, this study proposes a framework of E- on could

commerce performance to organize the complex literature affect the
parsimoniously. The framework is structured around three major generality
domains of an organization: of review
papers.
- the organization’s external market environment;
- the organization’s internal optimal performance; and
- flows between the organization and its market environment,

manifesting intrinsic, dynamic and interactive relationships
that affect E-commerce performance.
The proposed framework was used to identify specific gaps in the

literature and discuss opportunities that lead to future research
frameworks, theoretical developments and emerging trends, in the
hopes of advancing scholarship and practices in this increasingly
more important area of E-commerce performance in hospitality and
tourism.
Key contributions of this study, therefore, can be summarized in four aspects.
First, this study produced a cohesive framework of E-commerce performance
based on an extensive review of literature in both the mainstream and
hospitality and tourism fields, addressing the issue of currently fragmented
understanding on E-commerce performance in hospitality and tourism. Second,
the proposed framework was used to identify specific gaps in the literature and
discuss opportunities that lead to future research frameworks, theoretical
developments and emerging trends in this increasingly more important area of
E-commerce performance in hospitality and tourism. Third, the proposed
framework would also shed light on industry practitioners with regard to
understanding determinants and being updated with current practices of E-
commerce performance. And fourth, the findings of this study point practitioners
to directions that can lead to better E-commerce performance.
Proper caution, though, should be exercised when interpreting
findings of this study because of the intrinsic limitations associated
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
69
IJCHM References
28,9 Abendroth, L.J. (2011), “The souvenir purchase decision: effects of online
availability”, International Journal of Culture, Tourism and Hospitality
Research, Vol. 5 No. 2, pp. 173-183.
Abou-Shouk, M., Megicks, P. and Lim, W.M. (2013), “Perceived benefits and e-
commerce adoption by SME travel agents in developing countries evidence
2070 from Egypt”, Journal of Hospitality and Tourism Research, Vol. 37 No. 4, pp.
490-515.
Alvarez, R.M., Hall, T.E. and Trechsel, A.H. (2009), “Internet voting in comparative
perspective:
the case of Estonia”, Political Science & Politics, Vol. 42 No. 3, pp. 497-
505.
Amit, R. and Schoemaker, P.J. (1993), “Strategic assets and organizational
rent”, Strategic Management Journal, Vol. 14, pp. 33-46.
Andreu, L., Aldas, J., Bigne, J.E. and Mattila, A.S. (2010), “An analysis of e-
business adoption and its impact on relational quality in travel agency

– supplier relationships”, Tourism Management, Vol. 31 No. 6, pp. 777-
787.
Asia Pacific Economic Cooperation, Telecommunications Working Group, Business
Facilitation Steering Group (1999), “SME electronic commerce study
(TEL05/97T)”, Final report, 24 September, PricewaterhouseCoopers, London, pp.
1-36.
Ayeh, J. (2006), “Determinants of internet usage in Ghanaian hotels: the
case of the Greater Accra Region (GAR)”, Journal of Hospitality and
Leisure Marketing, Vol. 15 No. 3, pp. 87-109.
Azam, M. (2007), “Internet adoption and usage in Bangladesh”, Japanese
Journal of Administrative Science, Vol. 20 No. 1, pp. 43-54.
Banyai, M. and Havitz, M.E. (2013), “Analyzing travel blogs using a realist evaluation
approach”, Journal of Hospitality Marketing and Management, Vol. 22 No. 2, pp. 229-
241.
Barney, J.B. (1992), “Integrating organizational behavior and strategy formulation
research: a resource based analysis”, Advances in Strategic Management, Vol.
8, pp. 39-61.
Barua, A. and Mukhopadhyay, T. (2000), “Information technology and
business performance: past, present and future”, in Zmud, R. (Ed.),
Framing the Domains of IT Management: Projecting the Future Through
the Past, Pinnaflex Educational Resources, Cincinnati, OH.
Beatty, R., Shim, J. and Jones, M. (2001), “Factors influencing corporate web site
adoption: a time-based assessment”, Information and Management, Vol. 38 No.
6, pp. 337-354.
Beckinsale, M. and Levy, M. (2004), “SMEs and internet adoption strategy: who do SMEs
listen
to?”, available at: http://is2.lse.ac.uk/asp/aspecis/20040016.pdf
Beekhuyzen, J., Hellens, L. and Siedle, M. (2005), “Cultural barriers in the adoption of
emerging technologies”, Proceedings of HCI International, Las Vegas, NV, 22-27
July.
Beldona, S., Lin, K. and Yoo, L. (2012), “The roles of personal innovativeness and
push vs pull delivery methods in travel-oriented location-based marketing
services”, Journal of Hospitality and Tourism Technology, Vol. 3 No. 2, pp.
86-95.
Berkus, D. performance: an empirical investigation”, Management Information
(2013), Systems Quarterly, Vol. 24 No. 1, pp. 169-196.
“Riding
the
prevailin
g
winds”,
in Hotel
Yearboo
2014,
Wade
and
Compan
y SA,
Grandva
ux, pp.
10-11.
Berne, C.,
Garcia-
Gonzalez
, M. and
Mugica, J.
(2012),
“How ICT
shifts the
power
balance
of
tourism
distributi
on
channels
”,
Tourism
Manage
ment,
Vol. 33
No. 1, pp.
205-214.
Bharadwaj,
A.S.
(2000),
“A
resourc
e-based
perspec
tive on
informat
ion
technol
ogy
capabilit
y and
firm
Chua, C.E.H.,
Bigné-Alcañiz, E., Aldas-Manzano, J., Andreu-Simó, L. and Ruiz-Mafé, C. (2009), “Business- Straub,
to-business E-commerce adoption and perceived benefits: evidence from small and D.W.,
medium Spanish enterprises”, International Journal of Electronic Business, Vol. 7 No. Khoo,
6, pp. 599-624. H.M.,
Bilgihan, A., Okumus, F., “Khal” Nusair, K. and Joon-Wuk Kwun, D. (2011), “Information Kadiyala
, S. and
technology applications and competitive advantage in hotel companies”,
Kuechler
Journal of Hospitality and Tourism Technology, Vol. 2 No. 2, pp. 139-153.
, D.
Bourgouin, F. (2002), “Information communication technologies and the (2005),
potential for rural tourism SMME development: the case of the Wild “The
Coast”, Development Southern Africa, Vol. 19 No. 1, pp. 191-212. evolutio
Bowen, G. (2008), “Naturalistic inquiry and the saturation concept: a n of E-
research note”, Qualitative Research, Vol. 8 No. 1, pp. 137-142. commer
Boyd, A. and Bilegan, I. (2003), “Revenue management and E-commerce”, ce
Management Science, Vol. 49, pp. 1363-1386. research
: a
Bruneel, J., de Velde, E.V., Clarysse, B. and Gemmel, P. (2012), “Improving stakehol
the success of radical innovation projects within established firms: der

engaging employees across different hierarchal levels”, Technology perspect
Analysis and Strategic Management, Vol. 24 No. 4, pp. 951-965. ive”,
Buhalis, D. (2003), ETourism: Information Technology for Strategic Tourism Journal
Management, Financial Times Prentice Hall, Harlow. of
Buhalis, D. and Law, R. (2008), “Progress in information technology and Electron
tourism management: 20 years on and 10 years after the Internet – ic
the state of eTourism research”, Tourism Management, Vol. 29 No. 4, Commer
pp. 609-623. ce
Researc
Carr, N.G. (2003), “IT doesn’t matter”, Harvard Business Review, Vol. 81, pp. 41-
h, Vol. 6
49.
No. 4,
Casaló, L.V., Flavián, C. and Guinalíu, M. (2010), “Determinants of the intention pp. 262-
to participate in firm-hosted online travel communities and effects on 279.
consumer behavioral intentions”, Tourism Management, Vol. 31 No. 6, pp.
898-911.
Chae, H.C., Koh, C.E. and Prybutok, V.R. (2014), “Information technology
capability and firm performance: contradictory findings and their possible
causes”, Management Information Systems Quarterly, Vol. 38 No. 1, pp.
305-326.
Chan, Y.E. (2000), “IT value: the great divide between qualitative and
quantitative and individual and organizational measures”, Journal of
Management Information Systems, Vol. 16 No. 4, pp. 225-261.
Chathoth, P.K. (2007), “The impact of information technology on hotel operations,
service management and transaction costs: a conceptual framework for full-
service hotel firms”, International Journal of Hospitality Management, Vol. 26 No.
2, pp. 395-408.
Chathoth, P.K. and Law, R. (2011), “Managerial perceptions of information
technology and their impact from a transaction cost perspective”,
Journal of Travel and Tourism Marketing, Vol. 28 No. 8, pp. 787-803.
Chen, L., Gillenson, M.L. and Sherrell, D.L. (2002), “Enticing online
consumers: an extended technology acceptance perspective”,
Information and Management, Vol. 39 No. 8, pp. 705-719.
Christensen, C.M. and Overdorf, M. (2000), “Meeting the challenge of
disruptive change”, Harvard Business Review, Vol. 78 No. 2, pp. 66-77.
E
-
c
o
m
m
e
rc
e
p
e
rf
o
r
m
a
n
c
e
20
71
IJCHM Cobanoglu, C., Ayoun, B., Connolly, D. and Nusair, K. (2013), “The effect of information
technology steering committees on perceived IT management sophistication in
28,9 hotels”, International Journal of Hospitality and Tourism Administration, Vol. 14 No. 1,
pp. 1-22.
Cohen, J.F. and Olsen, K. (2013), “The impacts of complementary information technology
resources on the service-profit chain and competitive performance of South African
2072 hospitality firms”, International Journal of Hospitality Management, Vol. 34, pp. 245-
254.
Collins, C., Buhalis, D. and Peters, M. (2003), “Enhancing SMTEs business performance
through the Internet and e-learning platforms”, Education Training, Vol. 45 Nos 8/9, pp.
483-494.
Corbin, J. and Strauss, A. (2008), Basics of Qualitative Research, Sage, Thousand
Oaks, CA.
Daniel, E. and Wilson, H. (2002), “Adoption intensions and benefits realised:
a study of E-commerce in UK SMEs”, Journal of Small Business and
Enterprise Development, Vol. 9, pp. 331-348.

Denyer, D. and Tranfield, D. (2009), “Producing a systematic review”, in
Buchanan, D.A. and Bryman, A. (Eds), The SAGE Handbook of
Organizational Research Methods, SAGE Publications, London, pp. 671-
689.
Dev, S., Buschman, J., Bowen, J. (2010), “Hospitality marketing: a
retrospective analysis (1960-2010) and predictions (2010-2020)”,
Cornell Hospitality Quarterly, Vol. 51 No. 4, pp. 459-469.
Dierickx, I. and Cool, K. (1989), “Asset stock accumulation and sustainability
of competitive advantage”, Management Science, Vol. 35 No. 12, pp.
1504-1511.
Dyerson, R. and Harindranath, G. (2007), “ICT Adoption and Use by SMEs in
the UK: a survey of South East”, 2014 Portland International
Conference on Management of Engineering and Technology,
Kanazawa, 27-31 July.
eBusiness W@tch (2006), ICT and E-Business in the Tourism Industry,
Sector Impact Study, No. 08/2006, European Commission, available at:
www.ebusiness-watch.org/resources/ tourism/SR08-2006_Tourism.pdf
El-Gohary, H. (2012), “Factors affecting E-marketing adoption and
implementation in tourism firms: an empirical investigation of Egyptian
small tourism organizations”, Tourism Management, Vol. 33 No. 5, pp.
1256-1269.
Elsenhardt, K.M. and Martin, J.A. (2000), “Dynamic capabilities: what are
they”, Strategic Management Journal, Vol. 21, pp. 1105-1121.
Emmer, R.M., Tauck, C., Wilkinson, S. and Moore, R.G. (1993), “Marketing
hotels using global distribution systems”, Cornell Hotel and Restaurant
Administration Quarterly, Vol. 34 Nos 5/6, pp. 80-89.
Escobar-Rodríguez, T. and Carvajal-Trujillo, E. (2013), “An evaluation of
Spanish hotel websites: informational vs relational strategies”,
International Journal of Hospitality Management, Vol. 33, pp. 228-239.
Feeny, D.F. and Willcocks, L.P. (1998), “Core IS capabilities for exploiting
information technology”, Sloan Management Review, Vol. 39 No. 3, pp.
9-21.
Gaffney, T.
(2013), “Hot
travel trends”,
paper
presented at
HITEC 2013,
Minneapolis,
MN.
Gaston, G.
and
Botts, T.
(2013),
“The
rise of
metame
diaries”,
paper
present
ed at
HITEC
2013,
Minneap
olis, MN.
Grandon, E.
and
Pearson,
J.
(2004),
“Electro
nic
commer
ce
adoptio
n: an
empiric
al study
of small
and
medium
US
busines
ses”,
Informa
tion and
Manage
ment,
Vol. 42
No. 1,
pp. 197-
216.
Jawahar, I.
Greenhalgh, T. and Peacock, R. (2005), “Effectiveness and efficiency of and
search methods in systematic reviews of complex evidence: audit of McLaugh
primary sources”, BMJ, Vol. 331 No. 7524, pp. 1064-1065. lin, G.L.
Greenhalgh, T., Potts, H.W., Wong, G., Bark, P. and Swinglehurst, D. (2009), “Tensions (2001),
and paradoxes in electronic patient record research: a systematic literature “Toward
review using the meta-narrative method”, Milbank Quarterly, Vol. 87 No. 4, pp. a
729-788.
descripti
ve
Guttentag, D.A. (2010), “Virtual reality: applications and implications for stakehol
tourism”, Tourism Management, Vol. 31 No. 5, pp. 637-651. der
Harindranath, G., Dyerson, R. and Barnes, D. (2008), “ICT adoption and use theory:
in UK SMEs: a failure of initiatives?” Electronic Journal of Information an
Systems Evaluation, Vol. 11 No. 2, pp. 91-96. organiza
Hays, S., Page, S.J. and Buhalis, D. (2013), “Social media as a destination marketing tool: tional
its use by national tourism organizations”, Current Issues in Tourism, Vol. 16, pp. life cycle
211-239.
approac
h”,
Herrero, Á. and San Martín, H. (2012), “Developing and testing a global model to Academ
explain the adoption of websites by users in rural tourism y of
accommodations”, International Journal of Hospitality Management, Vol. 31 Manage
No. 4, pp. 1178-1186. ment
Heung, V.C.S. (2003), “Barriers to implementing E-commerce in the travel industry: a Review,
practical perspective”, International Journal of Hospitality Management, Vol. 22 No. 1, Vol. 26
pp. 111-118. No. 3,
pp. 397-
Hilton (2014), The Hilton Worldwide eCommerce Program, available at: 414.
www. hiltonmanagementservices.com/en/revenue/ecommerce.html
Jeong, C.,
Ho, C.I., Lin, M.H. and Chen, H.M. (2012), “Web users’ behavioural patterns of tourism
Holland, S.,
information search: from online to offline”,Tourism Management, Vol. 33 No. 6, pp.
Jun, S.H.
1468-1482.
and
Hotel News Now (2015), “Report: OTA commissions a burden on margins”, available Gibson, H.
at: www. hotelnewsnow.com/Article/16235/Report-OTA-commissions-a-burden-
(2012),
on-margins
“Enhancing
Hua, N., Morosan, C. and DeFranco, A. (2015), “The other side of technology
destination
adoption: examining the relationships between E-commerce expenses and hotel
image
performance”, International Journal of Hospitality Management, Vol. 45, pp.
through
109-120.
travel
Hung, K. and Law, R. (2011), “An overview of internet-based surveys in
website
hospitality and tourism journals”, Tourism Management, Vol. 32 No. 4,
pp. 717-724. information
”,
Hwang, J., Yoon, Y. and Park, N. (2011), “Structural effects of cognitive and
affective responses to web advertisements, website and brand Internation
attitudes, and purchase intentions: the case of casual-dining al Journal
restaurants”, International Journal of Hospitality Management, Vol. 30 of Tourism
No. 4, pp. 897-907. Research,
Internet Live Stats (2014), Internet Live Stats, available at: Vol. 14 No.
www.internetlivestats.com 1, pp. 16-
Jalilvand, M.R., Samiei, N., Dini, B. and Manzari, P.Y. (2012), “Examining the 27.
structural relationships of electronic word of mouth, destination image, Jin, N. (2007), “A
tourist attitude toward destination and travel intention: an integrated study of
approach”, Journal of Destination Marketing and Management, Vol. 1 Nos informatio
1/2, pp. 134-143.
n
technolog
y Conference on Wireless Communications, Networking and Mobile Computing,
ad Shanghai. E-
op
tio
commerce
n performanc
for e
s
m
all
an
d 207
m
ed
3
iu
m
siz
ed
en
ter
pri
se
s:
str
at
eg
ic
co
m
pe
titi
ve
ne
ss
”,
Pa
pe
r
pr
es
en
te
d
at
th
e
20
07
Int
er
na
tio
na
l
IJCHM Kajogbola, D. (2004), “The impact of information technology on the Nigerian
economy: a study of manufacturing and services sectors in the South
28,9 Western and South Eastern Zones of Nigeria”, ATPS Working Paper
Series No. 39, available at: www.atpsnet.org/Files/
working_paper_series_39.pdf
Kang, M. and Gretzel, U. (2012), “Perceptions of museum podcast tours: effects of
2074 consumer innovativeness, internet familiarity and podcasting affinity on
performance expectancies”, Tourism Management Perspectives, Vol. 4, pp. 155-
163.
Karadag, E., Cobanoglu, C. and Dickinson, C. (2009), “The characteristics of IT
investment decisions and methods used in the US lodging industry”,
International Journal of Contemporary Hospitality Management, Vol. 21 No. 1,
pp. 52-68.
Karagozoglu, N. and Lindell, M. (2004), “Electronic commerce strategy,
operations, and performance in small and medium-sized enterprises”,
Journal of Small Business and Enterprise Development, Vol. 11 No. 3,
pp. 290-301.
Karanasios, S. (2008), “An E-commerce framework for small tourism enterprises
in developing countries”, Unpublished doctoral dissertation, School of
Information Systems, Faculty of Business and Law, Victoria University,
Melbourne.
Karanasios, S. and Burgess, S. (2008), “Tourism and Internet adoption: a developing
world perspective”, International Journal of Tourism Research, Vol. 10 No. 2, pp. 169-
182.
Kartiwi, M. and MacGregor, R. (2007), “Electronic commerce adoption barriers in small to
medium-sized enterprises (SMEs) in developed and developing countries: a cross
country comparison”, Journal of Electronic Commerce in Organization, Vol. 5 No. 3,
pp. 35-51.
Keske, C. and Smutko, S. (2010), “Consulting communities: using audience response
system (ARS) technology to assess community preferences for sustainable
recreation and tourism development”, Journal of Sustainable Tourism, Vol. 18
No. 8, pp. 951-970.
Kim, C. (2006), “E-tourism: an innovative approach for the small and medium-
sized tourism enterprises (SMTEs) in Korea”, in Innovation and growth in
tourism, Organisation for Economic Co-operation and Development, Paris,
pp. 135-146.
Kim, H.B., Kim, T.T. and Shin, S.W. (2009), “Modeling roles of subjective
norms and eTrust in customers’ acceptance of airline B2C eCommerce
websites”, Tourism Management, Vol. 30 No. 2, pp. 266-277.
Kim, H.B., Lee, D.S. and Ham, S. (2013c), “Impact of hotel information security on system
reliability”, International Journal of Hospitality Management, Vol. 35 No. 3, pp. 369-
379.
Kim, J.S., Farrish, J. and Schrier, T. (2013a), “Hotel information technology
security: do hoteliers understand the risks?”, International Journal of
Hospitality and Tourism Administration, Vol. 14, pp. 282-304.
Kim, M.J., Lee, C.K. and Chung, N. (2013b), “Investigating the role of trust and
gender in online tourism shopping in South Korea”, Journal of Hospitality
and Tourism Research, Vol. 37 No. 3, pp. 377-401,
doi:10.1177/1096348012436377.
Kim, T.G., Lee, J.H. and Law, R. (2008), “An empirical examination of the
acceptance behaviour of hotel front office systems: an extended
technolo environment framework”, Information and Management, Vol. 38 No. 8,
gy pp. 507-521.
accepta
nce
model”,
Tourism
manage
ment,
Vol. 29
No. 3,
pp. 500-
513.
Koo, B.,
Mantin,
B. and
O’Conno
r, P.
(2011),
“Online
distribut
ion of
airline
tickets:
should
airlines
adopt a
single or
a multi-
channel
approac
h?”,
Tourism
Manage
ment,
Vol. 32
No. 1,
pp. 69-
74.
Kuan, K. and
Chau, P.
(2001),
“A
percepti
on-
based
model
for EDI
adoption
in small
business
es using
technolo
gy-
organiza
tion-
Manage
Kvainauskaite, V., Sarapovas, T. and Cvilikas, A. (2005), “Selection and ment
assessment of E-commerce models in SMEs”, Emerging Markets, Vol. 4 No. Informat
4, pp. 64-70. ion
Law, R., Buhalis, D. and Cobanoglu, C. (2014), “Progress on information and Systems
communication technologies in hospitality and tourism”, International Quarterl
Journal of Contemporary Hospitality Management, Vol. 26 No. 5, pp. 727-
y, Vol.
750.
19 No. 4,
pp. 487-
Law, R. and Jogaratnam, G. (2005), “A study of hotel information technology applications”,
505.
International Journal of Contemporary Hospitality Management, Vol. 17 No. 2, pp.
Migiro, S.O.
170-180.
and
Law, R., Leung, R. and Cheung, C. (2012), “A systematic review, analysis, Ocholla,
and evaluation of research in the Cornell Hospitality Quarterly”, Cornell D.N.
Hospitality Quarterly, Vol. 53 No. 4, pp. 365-381. (2005),
Lederer, A.L., Mirchandani, D.A. and Sims, K. (2001), “The search for strategic advantage “Informa
from the World Wide Web”, International Journal of Electronic Commerce, Vol. 5 No. tion and
commun
4, pp. 117-133.
ication
Li, C.R., Lin, C.J. and Huang, H.C. (2014), “Top management team social technolo
capital, exploration-based innovation, and exploitation-based gies in
innovation in SMEs”, Technology Analysis and Strategic Management, small
Vol. 26 No. 1, pp. 69-85. and
Liang, A.R. and Lim, W.M. (2011), “Exploring the online buying behavior of specialty food medium
shoppers”, International Journal of Hospitality Management, Vol. 30 No. 4, pp. 855- scale
865. tourism
enterpris
Liao, Y. and Par, R. (2006), “The effect of E-commerce on travel agent in es in
Taiwan”, Journal of Sport, Leisure and Hospitality Research, Vol. 4 No. Durban,
1, pp. 102-114. South
Lincoln, Y.S. and Cuba, E.G. (1985), Naturalistic Inquiry, Sage, Beverly Hills, Africa”,
CA. Informat
Llach, J., Marimon, F., del Mar Alonso-Almeida, M. and Bernardo, M. (2013), ion
“Determinants of online booking loyalties for the purchasing of airline develop
tickets”, Tourism Management, Vol. 35, pp. 23-31. ment,
Vol. 21
Luna-Nevarez, C. and Hyman, M.R. (2012), “Common practices in destination website
No. 4,
design”, Journal of Destination Marketing and Management, Vol. 1 Nos 1/2, pp. 94- pp. 283-
106. 294.
MacGregor, R. (2004), “The role of small business strategic alliances in the Mills, J.E.,
adoption of E-commerce in small-medium enterprises (SMEs)”, Meyers,
Unpublished Doctoral Dissertation, University of Wollongong, New M. and
South Wales. Byun, S.
Maclnnis, D.J. (2011), “A framework for conceptual contributions in (2010),
marketing”, Journal of Marketing, Vol. 75 No. 4, pp. 136-154. “Embrac
ing
Mamaghani, F. (2009), “Impact of E-commerce on travel and tourism: an
broad-
historical analysis”, International Journal of Management, Vol. 26 No. 3,
scale
p. 365. applicati
Masli, A., Richardson, V.J., Sanchez, J.M. and Smith, R.E. (2011), “Returns to ons of
IT excellence: evidence from financial performance around information biometri
technology excellence awards”, International Journal of Accounting c
Information Systems, Vol. 12 No. 3, pp. 189-205. technolo
Mata, F.J., Fuerst, W.L. and Barney, J.B. (1995), “Information technology and gies in
sustained competitive advantage: a resource-based analysis”, hospitali
ty Journal of Hospitality and Tourism Technology, Vol. 1 No. 3, pp. 245-
a 256. E-
n
d commerce
to performanc
ur e
is
m
:
is
th
e 207
b 5
us
in
es
s
re
a
d
y?
”,
IJCHM Morosan, C. (2011), “Customers’ adoption of biometric systems in
restaurants: an extension of the technology acceptance model”,
28,9 Journal of Hospitality Marketing and Management, Vol. 20 No. 6, pp.
661-690.
Morosan, C. (2014), “Toward an integrated model of adoption of mobile
phones for purchasing ancillary services in air travel”, International
2076 Journal Contemporary Hospitality Management, Vol. 26 No. 2, pp. 246-
271.
Morosan, C. and Jeong, M. (2008), “Users’ perceptions of two types of hotel
reservation web sites”, International Journal of Hospitality Management, Vol. 27
No. 2, pp. 284-292.
Morosan, C., T. Bowen, J. and Atwood, M. (2014), “The evolution of marketing
research”, International Journal of Contemporary Hospitality Management, Vol.
26, pp. 706-726.
Mukhopadhyay, T., Kekre, S. and Kalathur, S. (1995), “Business value of
information technology: a study of electronic data interchange”,

Management Information Systems Quarterly, Vol. 19 No. 2, pp. 137-
156.
Mustak, M., Jaakkola, E. and Halinen, A. (2013), “Customer participation and
value creation: a systematic review and research implications”,
Managing Service Quality, Vol. 23 No. 4, pp. 341-359.
Myung, E., Li, L. and Bai, B. (2009), “Managing the distribution channel
relationship with e-wholesalers: hotel operators’ perspective”, Journal
of Hospitality Marketing and Management, Vol. 18 No. 8, pp. 811-828.
Nevo, S. and Wade, M.R. (2010), “The formation and value of it-enabled resources:
antecedents and consequences”, Management Information Systems Quarterly, Vol.
34 No. 1, p. 10.
Nyheim, P.D. and Connolly, D.J. (2012), Technology Strategies for the
Hospitality Industry, 2nd ed., Prentice Hall, Upper River Saddle, NJ.
O’Connor, P. (1999), Electronic Information Distribution in Tourism and
Hospitality, CAB, Wallingford.
Oh, S., Lehto, X.Y. and Park, J. (2009), “Travelers’ intent to use mobile
technologies as a function of effort and performance expectancy”,
Journal of Hospitality Marketing and Management, Vol. 18 No. 8, pp.
765-781.
Pan, B. and Li, X.R. (2011), “The long tail of destination image and online
marketing”, Annals of Tourism Research, Vol. 38 No. 1, pp. 132-152.
Pan, B., Zhang, L. and Law, R. (2013), The Complex Matter of Online Hotel
Choice, Sage Publications, available at:
https://login.ezproxy.net.ucf.edu/login?auth shibbandurl
http://search.ebscohost.com/login.aspx?direct trueanddb edsgeaandAN
edsgcl.33700 9174andsite eds-liveandscope site
Paraskevas, A., Katsogridakis, I., Law, R. and Buhalis, D. (2011), “Search
engine marketing: transforming search engines into hotel distribution
channel”, Cornell Hospitality Quarterly, Vol. 52 No. 2, pp. 200-208.
Pease, W. and Rowe, M. (2005), “An overview of information technology in the tourism
industry”, Conference Proceedings-ICT Networks-Building Blocks for Economic
Development, Communication Economics and Electronic Markets Research
Centre.
Perlovich, D.
(2013), “3
proven
strategies to
help your E-
commerce
sales”, Forbes,
available at:
http://on
forb.es/
16zADG
N
Poon, A. (1993),
Tourism,
Technology and
Competitive
Strategies, CAB
International,
Oxford.
Poon, S. and
Joseph,
M.
(2001),
“A
prelimina
ry study
of
product
nature
and
electroni
c
commerc
e”,
Marketin
g
Intellige
nce and
Planning,
Vol. 19
No. 7,
pp. 493-
499.
Siqueira, A.C.O.
Quayle, M. (2002), “E-commerce: the challenge for UK SMEs in the twenty- and
first century”, International Journal of Operations and Production Fleury,
Management, Vol. 22 No. 10, pp. 1148-1161. M.T.L.
Raymond, L. (2001), “Determinants of web site implementation in small business”, (2011),
Internet Research: Electronic Networking Applications and Policy, Vol. 11 No. 5, pp. “Complem
411-422. entarities
Rong, J., Li, G. and Law, R. (2009), “A contrast analysis of online hotel web of human
service purchasers and browsers”, International Journal of Hospitality capital
Management, Vol. 28 No. 3, pp. 466-478. and
Ruiz-Molina, M.E., Gil-Saura, I. and Moliner-Velazquez, B. (2010), “The role of informatio
information technology in relationships between travel agencies and their n
suppliers”, Journal of Hospitality and Tourism Technology, Vol. 1 No. 2, pp. 144- technolog
162. y: small
businesse
Saffu, K. and Walker, J. (2008), “Strategic value and electronic commerce
adoption among small and medium-sized enterprises in a transactional s,
economy”, Journal of Business and Industrial Marketing, Vol. 23 No. 6, emerging

pp. 395-404. economy
context
Sambamurthy, V., Bharadwaj, A. and Grover, V. (2003), “Shaping agility through
digital options: reconceptualizing the role of information technology in and the
contemporary firms”,Management Information Systems Quarterly, pp. 237- strategic
263. role of
firm
San Martín, H. and Herrero, Á. (2012), “Influence of the user’s psychological
factors on the online purchase intention in rural tourism: integrating resources”
innovativeness to the UTAUT framework”, Tourism Management, Vol. 33 ,
No. 2, pp. 341-350. Technolog
Santhanam, R. and Hartono, E. (2003), “Issues in linking information technology capability y Analysis
&
to firm performance”, Management Information Systems Quarterly, Vol. 27 No. 1, pp.
Strategic
125-153.
Managem
Sauer, C. and Willcocks, L. (2002), “The evolution of the organizational ent, Vol.
architect”, MIT Sloan Management Review, Vol. 43 No. 3, pp. 41-49. 23 No. 6,
Scarborough, N. and Zimmerer, W. (2003), Effective Small Business pp. 639-
Management: An Entrepreneurial Approach, 7th ed., Pearson Education, 653.
Upper Saddle River, NJ.
SmartBrief
Schrier, T., Erdem, M. and Brewer, P. (2010), “Merging task-technology fit and Media
technology acceptance models to assess guest empowerment technology Services
usage in hotels”, Journal of Hospitality and Tourism Technology, Vol. 1 No. (2013),
3, pp. 201-217.
“10
Schroeder, A., Pennington-Gray, L., Donohoe, H. and Kiousis, S. (2013), “Using social media Hospitalit
in times of crisis”, Journal of Travel and Tourism Marketing, Vol. 30 Nos 1/2, pp. 126- y
143. technolog
Sheldon, P. (1997), Tourism Information Technologies, CAB, Oxford. y trends
Sigala, M. (2003), “The information and communication technologies you need
productivity impact on the UK hotel sector”, International Journal of to know
Operations and Production Management, Vol. 23 No. 10, pp. 1224- about”,
1245. available
Simpson, M. and Docherty, A. (2004), “E-commerce adoption support and advice for UK at:
SMEs”, Journal of Small Business and Enterprise Development, Vol. 11 No. 3, pp. 315- www2.sm
328. artbrief.c
om/hoste
d/ sues influencing the use of the internet and electronic commerce
a among small-medium sized enterprises”, Journal Electronic Commerce E-
d Research, Vol. 4 No. 1, pp. 15-33.
commerce
2
1 performanc
8 e
7/
H
os
pi
ta 207
lit 7
yT
re
n
ds
2
0
1
3.
p
df
sfi
el
d,
M
.
a
n
d
G
ra
nt
,
K.
(2
0
0
3)
,
“A
n
in
v
e
st
ig
at
io
n
in
to
is
IJCHM Straub, D. and Klein, R. (2001), “E-competitive transformations”, Business
Horizons, Vol. 44 No. 3, pp. 3-12.
28,9
Tallon, P.P. and Kraemer, K.L. (2007), “Fact or fiction? A sensemaking
perspective on the reality behind executives’ perceptions of IT
business value”, Journal of Management Information Systems, Vol. 24
No. 1, pp. 13-54.
2078 Tarcan, E. and Varol, E.S. (2010), “Role of the demographic factors in the
process of hotel information systems adoption”, Turizam: znanstveno-
strucˇni cˇasopis, Vol. 58 No. 2, pp. 127-144.
Teece, D.J., Pisano, G. and Shuen, A. (1997), “Dynamic capabilities and
strategic management”, Strategic Management Journal, Vol. 18 No. 7,
pp. 509-533.
Teo, T., Lin, S. and Lai, K. (2009), “Adopters and non-adopters of e-
procurement in Singapore: an empirical study”, Omega, Vol. 37 No. 5,
pp. 972-987.
Thakran, K. and Verma, R. (2013), “The emergence of hybrid online distribution channels in
travel, tourism and hospitality”, Cornell Hospitality Quarterly, Vol. 54 No. 3, pp. 240-
247.
Thong, J.Y.L. (2001), “Resource constraints and information systems
implementation in Singaporean small businesses”, Omega, Vol. 29 No. 2,
pp. 143-156.
Tsai, H., Pan, S. and Lee, J. (2011), “Recent research in hospitality financial
management”, International Journal of Contemporary Hospitality
Management, Vol. 23 No. 7, pp. 941-971.
Tsang, N.K. and Hsu, C.H. (2011), “Thirty years of research on tourism and
hospitality management in China: a review and analysis of journal
publications”, International Journal of Hospitality Management, Vol. 30 No.
4, pp. 886-896.
Tussyadiah, I.P. and Zach, F.J. (2012), “The role of geo-based technology in
place experiences”, Annals of Tourism Research, Vol. 39 No. 2, pp. 780-
800.
UNWTO (2001), eBusiness for Tourism: Practical Guidelines for Destinations
and Businesses, World Tourism Organization, Madrid.
Wang, D., Park, S. and Fesenmaier, D.R. (2012), “The role of smartphones in
mediating the touristic experience”, Journal of Travel Research, Vol. 51 No.
4, pp. 371-387.
Wang, E.T.G., Hu, H.F. and Hu, P.J.H. (2013), “Examining the role of
information technology in cultivating firms’ dynamic marketing
capabilities”, Information Management, Vol. 50 No. 6, pp. 336-343.
Wang, H.Y. and Wang, S.H. (2010), “Predicting mobile hotel reservation
adoption: insight from a perceived value standpoint”, International
Journal of Hospitality Management, Vol. 29, pp. 598-608.
Wang, P. (2010), “Chasing the hottest IT: effects of information technology
fashion on organizations”, Management Information Systems Quarterly,
Vol. 34, pp. 63-85.
Watkins, E. (2013), “Trends in hotel technology”, available at:
www.hotelnewsnow.com/Article/ 10764/10-trends-in-hotel-technology
rthner, H. and
Klein, S. (1999),
“ICT and the
changing
landscape of
global tourism
distribution”,
Electronic
Markets, Vol. 9 No.
4, pp. 256-262.
eler, B.C. (2002),
“NEBIC: a dynamic
capabilities theory
for assessing net-
enablement”,
Information
Systems Research,
Vol. 13 No. 2, pp.
125-146.
dside, A.G., Ramos,
M.V. and Duque,
M. (2011),
“Tourism’s
destination
dominance and
marketing website
usefulness”,
International
Journal of
Contemporary
Hospitality
Management, Vol.
23 No. 4, pp. 552-
564.
Xiang, Z. and Gretzel, U. (2010), “Role of social media in online travel E-
information search”, Tourism management, Vol. 31 No. 2, pp. 179-188.
commerce
Xiong, L. and Hu, C. (2010), “Harness the power of viral marketing in hotel
industry: a network discount strategy”, Journal of Hospitality and performanc
Tourism Technology, Vol. 1 No. 3, pp. 234-244. e
Yang, T., Xun, J. and He, X. (2015), “British SMEs’ E-commerce technological
investments and firm performance: an RBV perspective”, Technology
Analysis & Strategic Management, Vol. 27 No. 5, pp. 586-603.
Zheng, J., Caldwell, N., Harland, C., Powell, P., Woerndl, M. and Xu, S. (2004),
“Small firms and E-business: cautiousness, contingency and cost- 207
benefit”,Journal of Purchasing and Supply Management, Vol. 10 No. 1, pp.
27-39. 9
Zhu, K. (2004), “The complementarity of information technology
infrastructure and E-commerce capability: a resource-based
assessment of their business value”, Journal of Management
Information Systems, Vol. 21 No. 1, pp. 167-202.
Zhu, K. and Kraemer, K.L. (2002), “E-commerce metrics for net-enhanced

organizations: assessing the value of E-commerce to firm performance in
the manufacturing sector”, Information Systems Research, Vol. 13 No. 3,
pp. 275-295.
Corresponding author
Nan Hua can be contacted at: nan.hua@gmail.com
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
This article has been cited by:
- Ruggero Sainaghi, Mehmet Ali Köseoglu, Francesca d’Angella, Issaka Lawerh Tetteh. 2019. Foundations
of hospitality performance measurement research: A co-citation approach. International Journal of
Hospitality Management 79, 21-40. [Crossref]
- Xusen Cheng, Shixuan Fu, Jianshan Sun, Anil Bilgihan, Fevzi Okumus. 2019. An investigation on online
reviews in sharing economy driven hospitality platforms: A viewpoint of trust. Tourism Management
71, 366-377. [Crossref]
- SarmahBijoylaxmi, Bijoylaxmi Sarmah, KambojShampy, Shampy Kamboj, KandampullyJay, Jay Kandampully. 2018.
Social media and co-creative service innovation: an empirical study. Online Information Review 42:7, 1146-
1179. [Abstract] [Full Text] [PDF]
- SainaghiRuggero, Ruggero Sainaghi, BaggioRodolfo, Rodolfo Baggio, PhillipsPaul, Paul Phillips,

MauriAurelio G., Aurelio G. Mauri. Hotel performance and research streams: a network cluster
analysis. International Journal of Contemporary Hospitality Management, ahead of print. [Abstract]
[Full Text] [PDF]
- Garry Wei-Han Tan, Keng-Boon Ooi. 2018. Gender and age: Do they really moderate mobile tourism shopping
behavior?. Telematics and Informatics 35:6, 1617-1642. [Crossref]
- Natalia Daries, Eduard Cristobal-Fransi, Berta Ferrer-Rosell, Estela Marine-Roig. 2018. Maturity and development of high-
quality restaurant websites: A comparison of Michelin-starred restaurants in France, Italy and Spain. International Journal
of Hospitality Management 73, 125-137. [Crossref]
- Ruggero Sainaghi, Paul Phillips, Rodolfo Baggio, Aurelio Mauri. 2018. Cross-citation and authorship
analysis of hotel performance studies. International Journal of Hospitality Management 73, 75-84.
[Crossref]
- Fuentes-BlascoMaría, María Fuentes-Blasco, Moliner-VelázquezBeatriz, Beatriz Moliner-Velázquez, Gil-
SauraIrene, Irene Gil-Saura. 2017. Exploring relationship variables and Information and
Communication Technologies use in industrial segmentation. Management Decision 55:7, 1441-1459.
[Abstract] [Full Text] [PDF]
- OkumusFevzi, Fevzi Okumus, BilgihanAnil, Anil Bilgihan, OzturkAhmet Bulent, Ahmet Bulent Ozturk, ZhaoXinyuan
(Roy), Xinyuan (Roy) Zhao. 2017. Identifying and overcoming barriers to deployment of information technology
projects in hotels. Journal of Organizational Change Management 30:5, 744-766. [Abstract] [Full Text] [PDF]
- Gonzalo Díaz-Meneses. 2017. A multiphase trip, diversified digital and varied background approach to
analysing and segmenting holidaymakers and their use of social media. Journal of Destination
Marketing & Management . [Crossref]

IJBE: Integrated Journal of Business and Economics
e-ISSN: 2549-3280
DOI: https://doi.org/10.5281/zenodo.1258437
Date of Publication: 4 June 2018
Volume 2 Issue 2, pp. 73-79
Electronic Commerce: Theory and Practice

Margarita Išoraitė, Neringa Miniotienė Vilniaus kolegija/The University Applied Sciences
misoraite@gmail.com
Abstract
Electronic commerce in the world is becoming an increasingly popular form of trade. Most
shoppers start looking for products, descriptions and quality features online before buying a
product. In order to provide customers with more convenience, more and more companies and
existing stores are setting up their own online stores where a person can buy at a convenient
time, even at night when regular stores are no longer working. Online stores allow you to save
time spent by a person searching for a particular product and driving through shops. The article
analyzes the concept of e-commerce, the advantages and disadvantages of electronic commerce,
and the situation of electronic commerce in Lithuania.
Keywords: Electronic commerce, Internet, E-commerce advantages and disadvantages
1. Introduction
Electronic commerce is a business in which information technology is used to increase sales,
business efficiency and provide a basis for new products and services. Through its activities,
each company communicates with many other entities: it may be private or corporate clients,
business partners, suppliers. When communicating with each other, these entities exchange
various types of information: they inform themselves about their products and services, negotiate
the terms of transactions, exchange documents, transmit and accept orders for goods, complain
about inappropriate services, distribute press releases, etc. The flow of information is also carried
out among all of these entities: managers and subordinates communicate with the company,
marketing specialists talk about vendors and product managers. Internet popularity has provided
many new opportunities for business. Accelerated and simplified communication and
unimagined technical capabilities have enabled smart entrepreneurs to create advertisements,
stores, and other promising items on their own, on the Internet. Moreover, the growing number
of Internet users has increased the volume and profitability of the use of commerce and services.
2. E-commerce Concept
A stated Wikipedia, electronic commerce - the purchase and sale of goods or services on the
internet. E-commerce is the pursuit of commercial activity through electronic tools. It is based on
the electronic processing and transmission of information (text, video, audio). E-commerce
involves many activities - e-commerce of goods and services, electronic delivery of digital
73
© 2018 IJBE Publishing.
Available Online at: http://ijbe-research.com
e-ISSN: 2549-3280
information, electronic auctions, direct marketing to consumers. Electronic commerce can be

widely applied in the following areas: e-trade; financial transactions in the provision of banking,
financial leasing, insurance and other services, investments, speculative operations in currency
and securities; other services markets: hotels, tourism, education, consulting, payment for
utilities, advertising and other; between various business, public, public and other institutions,
legal and natural persons, households and individuals (Misevičiūtė, 2001).
Table 1. Electronic Commerce Concept

Author Definition
Nanehkaran, Y. A. Electronic commerce is a powerful concept and process that has
(2013) fundamentally changed the current of human life. Electronic commerce
is one of the main criteria of revolution of Information Technology and
communication in the field of economy.
Shahriari, S., Electronic commerce, commonly known as E-commerce, is trading in
Shahriari, M., products or services using computer networks, such as the Internet.
Ggheiji, S. (2015) Electronic commerce draws on technologies such as mobile commerce,
electronic funds transfer, supply chain management, Internet
marketing, online transaction processing, electronic data interchange
(EDI), inventory management systems, and automated data collection
systems.
Gangeshwer, D. K. The effects of e-commerce already appear in all areas of business, from
customer service to new product design. It facilitates new types of
(2013)
information based business processes for reaching and interacting with
customers like online advertising and marketing, online order taking
and online customer service, etc.
Khan, A. G. (2016) Electronic commerce, or e-commerce, is the buying and selling of
goods and services on the Internet. Other than buying and selling,
many people use the Internet as a source of information to compare
prices or look at the latest products on offer before making a purchase
online or at a traditional store.
Fichter, K. (2003) E-commerce is understood as part of the business, which also includes,
for example, video conferencing and teleworking. By definitions
available so far, the term “e-business” can be defined as follows:
business processes, commercial activities, or other economic tasks
FORUM Fichter, Environmental Consequences of E-Commerce 27
conducted over the Internet or computer-mediated networks (Intranet,
etc.).
Ritter, J. B. (1992) International electronic commerce is not simply defined by the absence
of paper documents. Indeed, though an important objective of many
commercial and administrative users is the elimination of paper, most
users will also admit that current implementation efforts remain
heavily inter-dependent upon the continued use of paper documents.
Ngai, E., W. T. and , There is a growing interest in the use of electronic commerce (EC) as
Wat, F. K. T. (2002) a means to perform business transactions. For many businesses, it has
become a priority Through using EC; companies can connect with
their trading partners for „„just in time production‟‟ and „„just in time
74
e-ISSN: 2549-3280
Author Definition
delivery‟‟, which improves their competitiveness globally. „
Mansell, R, (2003) B2B electronic commerce is widely believed to have implications for
the complete value chain of business processes that firms become
involved in when they trade. B2B electronic commerce, enabled by the
Internet, is expected to increase the efficiency of internal firm
processes and to streamline inter-firm linkages. Firms of all sizes are
expected to benefit from reduced information asymmetries and
strengthened business relationships as a result of their introduction of a
range of electronic commerce applications
Gunasekarana, A., EC provides new channels for the global marketing of tangible goods
Marrib, H. B., and presents opportunities to create new businesses providing
McGaugheyc, R. E., information and other knowledge-based intangible products.
Nebhwanib, M. D.
(2002)
E-business can be supported by all market participants (Misevičiūtė, 2001):

← Business (B) with business (B): (B2B);
← Business (B) with government agencies (G): (B2G);
← Private and other business enterprises (B) with the budget, state-owned capital companies
(G): (B2G);
← Public Authorities and Budget Firms (G) with Public Institutions and Budget Firms:
(G2G);
← Business (B), public authorities and budget firms (G) with individual consumers and
households (C), as well as by identifying buyers and recipients of electronic models:
(B2C and G2C);
← Individual consumers and households with each other: (C2C).
b E-commerce Advantages and Disadvantages

Electronic commerce is a tool for promoting new forms and dimensions of business. After all,
you can sell your online store seven days a week, 24 hours a day, and you do not need any
special premises, sellers, complicated accounting systems or stores in each country. Your online
store is accessible, visible and visited everywhere and always, and your customer does not even
need to leave home to buy a product that he likes, he only has to click on the "buy" button and
the courier will deliver his desired item directly to his home (see-marketing.lt, accessed 2017).
E-commerce offers the user the opportunity at any time, regardless of where it is located (as long
as the user has access to the Internet) information on goods and services, as well as purchasing
them. This removes the barriers associated with time and space constraints when acquiring goods
or services. The user has the electronic space the ability to compare offers from different traders
regarding price, quality aspect, and assessment seller's credibility. The user can personalize his
purchase and keep himself from subjective criteria for determining the product's choice, such as
seller's persuasion, alleged discounts shares.
75
e-ISSN: 2549-3280
Table 2. E-commerce Advantages and Disadvantages

(Based on alinosbloggas.blogspot.co.id, webo.lt, and mususavaite.lt)
Advantages Disadvantages
Convenience - You save time by physically The ability to be cheated - Your money
shopping. You can shop anywhere you security depends on your vigilance. So
have an Internet connection and at any before submitting your personal
time. information, check that the online store is
trusted. When you buy online, you risk not
getting or getting the same as the one shown
in the photo.
Information details and recommendations - It is easy to make a mistake on the price -
Detailed information is provided for each Keep in mind that the product will have to
item in the e-shop. Here you can find pay more than specified.
product reviews and recommendations.
Lower prices and better choices - Online Lack of privacy - Some online stores even
sellers do not have to pay rent or wages to unsolicited send you emails about various
employees. It is convenient for you to stocks and new items. On the one hand, this
compare prices for the same product in may seem useful, but on the other hand, it is
different e-shops in a short period. a way to get you more money.
Starting an e-business requires less When ordering a product online, it often
investment than traditional commerce takes a long time before it is delivered, and
A wider circle of consumers is attracted shipping charges sometimes exceed the
(disabled, foreign clients) price of the item, especially if the product is
ordered from abroad. Repaying your online
purchase may be more difficult than buying
a traditional store
More varied and more convenient supply Often money will not be refunded for
of goods and services shipping, and in some cases, it will even
Comfortable billing (bank charges, bank have to cover the cost of returning the
link integration). product itself. Also, when shopping online,
it is important to make sure the website is
trusted, and your credit card or online
banking data will be protected. Finally,
shopping online will not be able to see the
goods (like clothes).
The main advantage of e-commerce is to provide its current and potential customer with all
information about the product, to offer to order and pay for it online, at any time of the day and
anywhere, on the Internet, advice on matters of concern (Misevičiūtė, 2001).
Electronic commerce also has some failures. Studies have shown that 35% of expensive goods
sold in virtual stores are pirated. Investigations have also shown buyers a lot of confidence
(Misevičiūtė, 2001).
76
e-ISSN: 2549-3280
4. Statistical Data on E-commerce in Lithuania

In the EU Member States, e-commerce in 2016 accounted for 9.4% of all retail sales. Compared
to previous years, e-commerce grew by about 11%, while trading areas shrank by about 1.5%.
Still more significant changes are recorded in those countries where e-commerce is already a
relatively significant part of all retail sales. One such country is the United Kingdom. In this
country's retail market, the share of online shopping is about 17% and is proliferating. At that
time, the area of trade fell by more than 4% over the past year (Sagatauskas, A., 2018)
In Lithuania, the share of online shopping is still a relatively small part of all retail trade - in
2016 it was about 3.5%. A Eurobarometer survey shows that every third person in the country
has been online at least once in the EU. Thus, at present, a larger proportion of Lithuanians still
enjoy visiting physical stores, and the volume of online sales is too small to have a real impact
on the decline of traditional retail outlets. Vilnius is worth mentioning individually, with a trade
area (679 sq. M) per capita now considerably lower than in Riga or Tallinn (998 sq. M and 817
sq. M respectively). Therefore, in the future, we should still hear about the projects of new
shopping centres in the capital of Lithuania.
According to the data of the EU statistical agency Eurostat, the indicators related to e-commerce
have rapidly risen in Lithuania:
iv) In Lithuania, online sales of companies (excluding financial firms) grew from 18% in 2015
up to 22% in 2017.
v) There is also a noticeable increase in the number of online purchases by businesses (in 2014,
Internet purchases made up 25% and in 2017 it was already 28%).
vi) The largest growth in corporate turnover from e-commerce - in 2017 rose almost double
compared with 2014. (In 2014, the turnover of e-commerce companies was 7%, and in 2017
it was already 13%).
vii) When analyzing the statistics important for the development of e-commerce in Lithuania,
it is important to mention that (based on The Lithuanian Department of Statistics at
osp.stat.gov.lt):
o Between 2015 and 2017, the percentage of households with a personal computer
increased. 2015 it was 67.6%, while in 2017 it grew to 73%.
o Major cities in 2017 it has grown to 82.4%.
o Internet access has also increased. 2015 it reached 68.3%, and in 2017 it has grown to
75%. Statistics show an increasing interest in the Internet.
viii) Analyzing the proportion of companies that have sold goods or services (received orders)
by e-communications, it is noted that (based on The Lithuanian Department of Statistics at
osp.stat.gov.lt):
(E) Wholesale and retail networks in Lithuania grew from 25.2 percent in 2014 to 32.4%
- 2016
*
Computer programming, consultancy, and related activities; data processing, web
server services (hosting) and related activities; Internet gateway service activities also
increased from 23.9% in 2014 to 40.8% 2016
*
Sales and retail sales increased from 23.2%. In 2012, to 33.8 percent. – 2016.
77
e-ISSN: 2549-3280
- Conclusion
Electronic commerce is usually defined as the automation of routine business procedures and
operations and the transfer to virtual space. This process greatly increases business efficiency
and simplifies everyday routine work. Up to now, it has been assumed that most e-commerce
transactions are conducted across the globe, but recent research has shown that businesses are
moving fast between local companies. Many businesses keep internet and e-commerce in a more
convenient way to market their products on the local market (news.lt, accessed 2018). Most
online retailers are cheaper than the same products offered in regular stores. This is the main
reason why today's online stores are gaining market share. Some online stores have exceptional
conditions where you do not have to pay for delivery at home for a certain amount. When trading
online, less time is spent on paperwork, as in the e-commerce sales, as reports are automated. It
is convenient for small businesses to go online because they can easily compete with bigger
ones. Also, online sales are cost-saving - there is no need to rent an outlet to the store, and
warehousing is not always necessary. It is much easier to access foreign markets on the Internet.
The biggest shortage of online stores is that it is not possible to see a product before buying it
(tv3.lt, accessed 2018).
References
- Apsipirkimo Vieta - Elektroninė Parduotuvė. (n.d.). Retrieved 2018, from http://alinosbloggas.
blogspot.lt/2010/09/pirkimo-internetu-privalumai.html13.
- Elektroninė komercija. (2018, February 20). Retrieved 2018, from https://lt. wikipedia.org
/wiki/Elektroninė_komercija
- Elektroninė komercija ir globalizacija. (n.d.). Retrieved 2018, from http://www.news.lt/IT-
naujienos/Elektronine-komercija-ir-globalizacija.im?id=95369&f=c
- Elektroninės parduotuvės: Pliusai ir minusai. (n.d.). Retrieved 2018, from http://www.
mususavaite.lt/elektronines-parduotuves-pliusai-ir-minusai/
- Elektroninės parduotuvės privalumai. (n.d.). Retrieved 2018, from http://www.webo.lt/elektronines-
parduotuves-privalumai.htm
- Fichter, K. (2003). E-Commerce Sorting Out the Environmental Consequences. Journal of Industrial
Ecology, (6)2.
- Gangeshwer, D. K. (2013). E-Commerce or Internet Marketing: A Business Review from Indian
Context. International Journal of u- and e-Service, Science, and Technology, (6)6, pp.187-194.
- Gunasekarana, A., Marrib, H., B., McGaugheyc, R., E., Nebhwanib, M., D. (2002). E-commerce and
its impact on operations management. Int. J. Production Economics, (75), pp. 185–197.
- Households have personal computers, Internet access. (2017). Retrieved 2018, from
https://osp.stat.gov.lt/statistiniu-rodikliu-analize?indicator=S4R029#/
- Kas Yra E-Komercija. (n.d.). Retrieved 2018, from http://www.seo-marketing.lt/kas-yra-
marketingas/kas-yra-e-komercija.html
- Khan, A., G. (2016). Electronic Commerce: A Study on Benefits and Challenges in an Emerging
Economy. Global Journal of Management and Business Research: B Economics and Commerce,
(16)1.
- Internetinė prekyba: Nauda ir pavojus. (2010, May 03). Retrieved 2018, from
https://www.tv3.lt/naujiena/387629/internetine-prekyba-nauda-ir-pavojus
- Mansell, R. (2003). Electronic commerce: conceptual pitfalls and practical realities. Prometheus,
(21)4, pp. 429-447.
- Misevičiūtė, B. (2001). Elektroninė komercija [PPT]. Retrieved 2018, from
http://kopustas.elen.ktu.lt /studentai/lib/exe/fetch.php?media=elektronine_komercija.ppt .
78
e-ISSN: 2549-3280
- Nanehkaran, Y., A. (2013). An Introduction To Electronic Commerce. International journal of scientific

& technology research, (2)4, pp. 190-193.
- Ngai, E., W., T. Wat, F., K., T. (2002). A literature review and classification of electronic commerce
research. Information & Management, (39), pp. 415-429.
- Ritter, J., B. (1992). Defining International Electronic Commerce. Northwestern Journal of International
Law & Business, (13)1, pp. 3-30.
- Sagatauskas, A. (2018, January 01). Kada e. prekyba Lietuvoje rim. Retrieved from
https://www.vz.lt/prekyba/2017/07/31/kada-e-prekyba-lietuvoje-rimciau-susigrums-su-tradicinemis-
parduotuvemis
- Shahriari, S., Shahriari, M., Ggheiji, S. (2015). E-commerce and it impacts on global trend and market.
International journal of research-Granthaalayah. (3)4, pp. 49-55.
- The proportion of enterprises which sold (received orders for) goods or services via electronic networks.
(2017). Retrieved 2018, from https://osp.stat.gov.lt/statistiniu-rodikliu-analize? indicator= S4R027#/17
79
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/321176869
E-Commerce: A Short History Follow-up on Possible Trends
Article · October 2017

DOI: 10.5430/ijba.v8n7p130
CITATIONS READS
0 1,797
4 authors, including:
Valdeci Ferreira dos Santos Leandro Ricardo Sabino

Universidade FUMEC 2 PUBLICATIONS 0 CITATIONS
12 PUBLICATIONS 1 CITATION
SEE PROFILE
SEE PROFILE
Carlos Alberto Gonçalves

FUMEC e UFMG
198 PUBLICATIONS 1,202 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Comércio Eletrônico e Mídias Digitais e Suas Relações Com os Consumidores View project
Neuromessenger View project
All content following this page was uploaded by Valdeci Ferreira dos Santos on 01 October 2018.
The user has requested enhancement of the downloaded file.

http://ijba.sciedupress.com International Journal of Business Administration Vol. 8, No. 7; 2017
E-Commerce: A Short History Follow-up on Possible Trends

Valdeci Ferreira dos Santos1, Leandro Ricardo Sabino2, Greiciele Macedo Morais3 & Carlos Alberto Gonçalves4
cFUMEC University; Degree in Business Administration by PUC/MG; Graduate in Strategic Business Management
by FUMEC University; Worked as a Logistic Support Analyst at Companhia de Saneamento de Minas Gerais
(COPASA); Currently holds a Business Administrator position at Federal University of Ouro Preto; Ongoing
Master’s in Business Administration by FUMEC University.
d FUMEC University; Degree in Information Technology by the Centro Universitário Newton Paiva, Graduate in
Project Management by Fundação Getúlio Vargas, Currently holds the Coordination on Logistic Engineering at CMP
– Componentes e Módulos Plásticos; Ongoing Master’s in Business Administration by FUMEC University
e Federal University of Ouro Preto; Degree in Business Administration and Accounting by PUC/MG; Graduate in
Strategic Business Management by FUMEC University; Worked as Assistant Dean of Planning at Federal University
of Ouro Preto (UFOP); Currently holds a Business Administrator position at Federal University of Ouro Preto
(UFOP); Holds a Master’s Degree in Business Administration by FUMEC University
f FUMEC and UFMG; Professor at both FUMEC University and UFMG. Holds a PhD in Business Administration
by USP/SP and a Master’s Degree in Computing by PUC/RJ. Currently teaches the subjects ‘Strategy and Planning’
and ‘Innovation and Entrepreneurship’. He runs a research project in Strategy and Marketology. Coordinator of the
Department of Research on Marketology and Strategy – NUME.
Correspondence: Greiciele Macedo Morais, Federal University of Ouro Preto, Brazil.
Received: November 2, 2017 Accepted: November 15, 2017 Online Published: November 20, 2017
doi:10.5430/ijba.v8n7p130 URL: https://doi.org/10.5430/ijba.v8n7p130
Abstract
The aim of this work is to think on the state-of-the-art of e-commerce and its trends for the future. E-commerce has
been developing since the 1990’s and its evolution is directly linked to the advancement of information technology.
Early e-commerce began with the simple dissemination of goods and services by digital means, going from the
issuance of orders, then the delivery of products to achieving interaction between traders and consumers via the
Internet. Some e-commerce tools enable users to perform transactions even without leaving home – with transactions
ranging from purchasing to paying bills. This can be done 24 hours a day, including weekends and holidays. The
demand for convenience and, even, privacy are the main responsible reasons for the increased use of electronic
commerce by consumers. Despite the advances of e-commerce in recent times it still requires larger investment,
especially regarding safety, which is appointed as major deficiency in this trade modality, along with logistics.
Investments will also be necessary to enable e-commerce to keep up with the current technological advances and
future development prospects, such as the adoption of virtual intelligence, the expansion of globalization with
language translators, adaptive interfaces that take into account the specific characteristics of user groups and the
mobile commerce, and the experimentation in 3D model.This study aims to – by means of literature review – draw a
brief history of the emergence and evolution of e-commerce, also highlighting the tools in use, the trends and
challenges of this modern business model.
Keywords: e-commerce, internet, information technology, consumers
1. Introduction
Given the advances on the transmission, processing and storage technologies, presentation forms, ergonomics,
information availability, speed and trustworthiness, more precisely with the development of the Internet, from 1990’s,
a large range of possibilities were noted – with a particular emphasis to communication. Some companies, having
noticed the potential of that tool, stated using it to interact with their customers – at first, to divulgate information on
their products, then to receive orders and, after that, to distribute products and services – up to arrival of the open
arrival of e-commerce, which would cover even receipt and reverse logistics.
According to e-commerce in Brazil (EC) had its beginning approximately eight years ago, with a growing number of
new companies to it, as well as a clientele which grows on a daily basis. Increased easiness to access the Internet was
Published by Sciedu Press 130 ISSN 1923-4007 E-ISSN 1923-4015

the driving force that pushed the increase in the number of new consumers to this new form of commerce. Authors
say that trust in using the tools which enable electronic commerce usually comes after the experience of the first
purchase.
According to (Tesche, 2013) some examples of products widely negotiated using electronic media, and purchased
mainly by young people, are electronics, flight and show tickets, among other events. The same author highlights
that the main attractive for consumers to use electronic media for shopping are access easiness and the fact that there
is no presence of physical sellers, what makes consumers more comfortable while shopping.
Currently EC finds itself on a much consolidated phase, bearing in mind the steady development of IT, the easiness
to access the Internet and the profile of this new consumer. Demands from that new consumer make companies
develop themselves to keep competitive in the market where they operate. For that end they seek to keep up with
market trends, the current one being the search for comfort by means of electronic commerce.
The present study aims at, by means of a theoretic review, outlining a short history of the beginning and evolution of
electronic commerce, also highlighting the tools used and the trends and challenges of this modern model of
commerce.
2. Electronic Commerce
In the next sections some aspects of EC are introduced. This explanation will enable the reader, giving more
information on the emergence, evolution, concepts, tools, challenges and trends of this modality, as well as
information on its relationship with the consumer.
2.1 Emergence and Evolution of Electronic Commerce
The evolution of EC has been studied and followed up by several researchers and stakeholders in the field. Given the
advances on Information Technology, more precisely with the development of the Internet, from 1990’s, a large
range of possibilities were perceived, with a noted emphasis to communication.
Nevertheless, (Galinari et al., 2015) advocate that e-commerce has its first phase on the 1970’s, when e-commerce
was restricted to operations among large corporations which established among themselves private communication
networks and, by means of electronic fund transfer systems, which electronically made financial transactions and
document exchanges.
According to (Albertin, 2012) the evolution of EC can be divided in four phases. In Phase One, organizations used
the functionalities of the Internet for processes of information divulgation regarding their product and services. That
was the initial stimulus for EC development.
Still according to the author, the Phase Two was receiving orders and sending information and instructions on the
utilization of their products and services. In this phase, logistics caused its first impact onto companies.
Phase Three of the evolution, according to (Albertin, 2012) was the distribution of products and services by using
Information Technology (IT). In this phase, some products began to be commercialized digitally as, e.g., music and
software.
For the last comes the phase which consolidates EC, with the interaction between seller and consumer, no more
transmitting data or delivering products and services only. With the advance of IT and widespread use of the Internet,
such interaction enabled the simple internet user to become a potential consumer, given the possibilities of EC. That
tool allowed for a true revolution on how to commercialize products, services and information, bringing more
comfort and large variety of offers and options for the consumer, but also for the seller who is inserted in that market
practice.
For it is just normal that companies undergo transformations in their structure, and globalization contributed for that
strong tendency. Increased competitiveness, the need of producing innovation and increasing consumer demands –
all items brought by globalization – have culminated with the emergence of more modern forms of company
management.According to the authors, the advance in the access of broadband Internet in last years has become
important for the development of EC. The appearance of a technology called 3G and 4G, from 2012 in Brazil, gave
access to high speed Internet through mobile devices, such as smartphones and tablet PCs.
Such technology enables the consumer with more easiness to do price research for various locations. Some
consumers use that resource even when inside physical stores, which they use to get in touch with the product, and
there they choose the purchase channel of sales which better fits their shopping needs.

The development of EC follows the stages of evolution of the digital environment – an evolution which must be
understood and guaranteed through the aspects that must be taken into consideration when using the EC – aiming to
assuring the use of its contributions. Thus, the development of information technology is impacted by the evolution
on EC. Another dimension which completes the analysis on EC phases is its application in business processes
(Albertin, 2000).
This way, electronic commerce became, much more than a trend, a reality. Modern companies seek to change their
structure to comply with new consumers’ demands, once the latter more and more value comfort. For such end, using
new technologies is a very common strategy.
2.2 Concepts and Types of Electronic Commerce
Electronic commerce is the achievement/realization of the entire chain of value of the business processes by means
of intense use of communication and information technologies, thus reaching the business’s objectives (Albertin,
2000).
According to the author, the processes may be achieved, entirely or in part, in any of the transaction types set forth in
Table 1, below. For him, such definition allows to understand that the achievement of that chain of values must
include from distribution of information on products and services to the performance of transactions between the
parties that compose the business environment.
According to the authors (Galinari et al., 2015) the Organization for Economic Cooperation and Development
(OECD) defines, in 2011, the EC as purchasing and selling products or services managed through a digital
environment and utilizing methods specifically idealized for the receipt or performance of commercial transactions.
Still according to the authors, that such definition sets, as e-commerce, transactions whose orders are made using
those methods, even when payments or deliveries are made without using the digital environment, restricted to
operations performed using a computer network which provides a controlled external access.
The great advance of EC, as stated by (da Silveira Coelho et al., 2013), permits a large amount of transactions, which
occur electronically, either between companies and consumers, between companies, between consumers and may
also involve government organs. Table 1 below shows some information on the types of transactions possible by EC.
Table 1. Types of transactions possible in electronic commerce

Acronym Name Description
B2B Business-to-business Transactions between companies.
B2C / C2B Business-to-consumer/Consumer-to-business Companies making transactions between
companies and the end consumer.
C2C Consumer-to-consumer Transactions between end consumers.
G2C/C2G Government-to-consumer/ Transactions between government and end
consumer-to-government consumers.
B2G/G2B Business-to-government/ Transactions between governmentand
government-to-business companies.
G2G Government-to-government Transactions between government
departments.
Note. Source: Adapted from da Silveira Coelho et al.(2013).
Authors appoint that among the transactions mentioned in Table 1, the categories most done through electronic
commerce are: B2B (Business-to-business), B2C (Business-to-consumer) and C2C (Consumer-to-consumer). For the
major highlight goes for B2C, which has the larger number of transactions. 2CB Still according to the authors, in
Brazil, the B2B commerce model is mostly comprised by commerce between industrial and commercial companies.
For the relevance of e-commerce expansion in current business scenarios is eminent, what requires the companies
inserted in this model to operate differentiated and up-to-date strategies and approaches so they may reach their
goals.

2.3 Main Tools in Electronic Commerce

According to (Luciano, Testa, & Freitas, 2003) the Internet is constantly improved, thus several applications often
appear seeking to better utilization of the electronic commerce services.
E-procurement, according to (Luciano et al., 2003) is the automation of the purchase of products and services –
besides dynamism in purchasing, a considerable cost decrease can be noted, also a shorter decisory time can be
reached.
Another important tool mentioned by (Luciano et al., 2003) is e-learning, which is a tool that shortens distances
between educators and students, providing them with conditions to access courses and training sessions wherever
they are. Important examples may be mentioned, such as: language schools, virtual universities, etc.
An important and successful tool is e-banking, which, as mentioned by (Luciano et al., 2003), is good both for
banking institutions and their clients. It enables clients to do, from home, what they would need to go to a bank
branch to do, before. Besides bringing practicity to consumers, a considerable cost is reduced to banks.
E-gambling is the tool for online casinos, with bets on money, which can be accessed by users located in any
country, including those which do not allow traditional casinos, according to (Luciano et al., 2003).
E-auctioning are online auctions, which reduce costs with commuting and make traditional auctions more accessible
and faster, as cited by (Luciano et al., 2003)
Many other tools were mentioned by (Luciano et al., 2003) as important ones for the development of electronic
commerce, despite having a minor expression. For example: e-trade, which is buying and selling stock shares online;
e-drugs, which are online drugstores, adding value to traditional drugstores, among others.
2.4 Electronic Commerce Advantages
According to (Lefebvre & Lefebvre, 2002), electronic commerce has become an imperative environment for
business. Opportunities it provides are so great there is no going back. As approached by the authors, electronic
commerce provides competitive advantages, such as cost reduction, because it offers gains in precision and speed for
the transactions on the Internet.
According to (Lefebvre & Lefebvre, 2002) still another advantage of electronic commerce is giving visibility in the
business world at low cost, making it possible to offer clients the opportunity of personalizing products and services.
Still according to authors, electronic commerce allows for marketing or micro marketing by means of data mining
techniques, higher capacity of adapting to changes in the competitive conditions, and the capacity of optimizing
supply chain in real time and of obtaining advantages through business partners’ skills.
For (da Silveira Coelho et al., 2013) the industrial and commercial companies using B2B in Brazil are those which
offer the clients the opportunity of product purchase and delivery at a location defined by the client.
According to (Kiang & Chi, 2001), when it comes to B2B transactions, a great advantage electronic commerce
provides is decreased processing time, which by consequence online companies get to keep lower levels of storage,
thus reducing management cost and overhead maintenance. Another advantage mentioned by the authors is that the
Internet provides fast adjustment to market conditions, which means it becomes possible to customize promotions
and sales for individual clients.
According to (Kiang & Chi, 2001), for products and services such as software, music, news, consultancy services,
issuance of online tickets and reservations, banking services, among others, which use the Internet as a distribution
channel, one can have not only delivery costs reduce but also immediate availability of products and services.
In Table 2 (Kiang & Chi, 2001) propose a framework which puts together advantages electronic commerce provides,
regarding marketing, within three groups: communication, transaction and distribution.
Table 2. Advantages of electronic commerce in marketing approach

Channel Advantages
Communication Better information on products
More information on Pricing
Service availability, 24 hours, 7 days per week
Lower communication costs

Interactivity and trustworthiness for on-demand information

Stock updates in real time
Online technical support
Fast response for clients’ requests
Customized orders
After-sales service
No personal contact
Transaction Virtual storefront viewable by all Internet users
Lower transaction cost
Allows micro transactions
Human error reduction
Purchase cycle time reduction
Lower level of storage and other associated costs
Possibility of customizing promos and sales for individual clients
Price flexibility
Relatively low initial investment and establishment cost
Distribution Lower wait time for receiving digital products and services
Lower cost for delivery of digital products and services
Allows clients to follow-up orders
Reduces the number of representatives
Note: Source: Adapted from Kiang e Chi (2001).
2.5 Challenges and Trends of Electronic Commerce

To (Lefebvre & Lefebvre, 2002), the electronic commerce causes a deep revolution in the way people do business; in
such deep way an adaptation and learning period becomes necessary, both for companies and for consumers. There
are obstacles to be overcome and a series of challenges for all involved stakeholders for e-commerce to be entirely
developed.
According to the authors, forecasting how the new industries will develop is still a difficult task. Some newly-created
segments are already changing, particularly Internet providers, where small ones are having more and more difficulty
in competing with larger ones, which can offer their clients a wider range of services (modem or cable connection,
ADSL [Asymmetric Digital Subscriber Line] technology, satellite links, etc.).
Regarding privacy, according to (Luciano et al., 2003), reconciling good service quality and privacy is still a
challenge to be met for the Internet.
Still according to the authors, another challenge for commerce is delivery, because there is no trust on the EC
customers’ side on a perspective as to if products are to be delivered with the expected quality and, if that quality is
not met, refunding will not be a hindrance.
According to (Diniz, 2008) the company-consumer relationship suggests availability of the technology related to
access points and cost and usability, also related to user interface and communication, which are fundamental items
for the evolution of the technology used in electronic commerce.
Another worrying factor according to (Diniz, 2008) is related to telecommunications where the Internet and its
transactions have had a higher impact. According to the author, despite the increase in the revenue of telecom
companies, bearing in mind their increased demand, their concern is that that demand is increasing at a faster pace
than investments.
In the company-company relationship, electronic commerce needs to march forward having in mind there still is a
trustworthiness gap as to safety systems, declares (Diniz, 2008).
According to (Cernev & Leite, 2005), safety has been appointed by specialists as the greatest hindrance for the
growth of electronic commerce in the world. And it has been responsible for billionary losses. That may be seen not
only as a restriction, but also as an opportunity for growth.

Authors appoint that those who are in charge of the development of electronic commerce should invest both on
safety and on trustworthiness of their users. They are responsible for assuring that consumers, potential consumers
and other users notice the existence and commitment to safety.
For (Diniz, 2008), with massive utilization of the Internet, another important factor which should be addressed is
regulation, which according to the author must be more effective and explicit, to assure commercial transactions at a
cost coherent with the traditional forms of commerce.
According to (da Silveira Coelho et al., 2013) the industrial and commercial companies in Brazil which use B2B are
those which need investment the most for their conception, development and maintenance.
According to (Demo et al., 2014) there are relevant logistic challenges for the practice of electronic commerce with
the growth of e-commerce in Brazil. According to the authors, logistics managers are being forced to implement
diverse operational practices with the purpose of meeting the new form of consumer demand.
Authors also highlight that electronic commerce consumers tend to be more demanding than their traditional
commerce counterparts. Those extra requirements and demands have required significant changes on online
company operationalization dynamics, as well as bringing new solutions for client satisfaction.
For (Novak, Hoffmam and Yung, 2000) the characteristics of the online environment exert positive influences on the
client's emotional and cognitive state during purchases. Corroboram (Ferreira et al., 2000) emphasizing that
consumers are increasingly submerged and focused on their shopping experience when the virtual environment is
more attractive.
To that end, they add (Trotti et al, 2017) that knowing the technology, functional logistics, providing a good
relationship with the customer at all stages of the sales cycle in the virtual environment are crucial points for e-
commerce and that need to be improved.
In addition to overcoming the challenges identified as providing greater security in purchases and payments, easy
navigation layout with greater level of entertainment and interactivity, and other improvements that may impact on
an improvement in the shopping experience, it is understood that a tendency for the continuous development and
evolution of e-commerce runs through new technologies such as virtual 3D experimentation.
The experience of having access to all dimensions of the products in 3D model can be a source of benefits to the
consumer in relation to the greater knowledge about the product, increase the possibility of finding a product or
alternative more assertive, saving time, thus generating a greater pleasure and reliability in the online shopping
process (Ferreira et al., 2015).
E-commerce is also about immersing itself in other technologies such as the adoption of virtual intelligence, the
expansion of globalization with language translators, adaptive interfaces contemplating the characteristics of specific
groups of users and mobile commerce as a way to improve the purchasing environment on-line and to improve the
consumer shopping experience (Ferreira et al., 2015, Trotti et al., 2017, Nienow et al., 2017).
2.6 Relationship with the Consumer
According to (Kotler & Keller, 2012), client-focused companies develop relationships, not only products or services.
Still according to the authors, the success of organizations depends on the clients, from acquisition to maintenance to
expansion. Clients are the only reason for developing products, installing factories, hiring employees, etc. With no
clients, there is no business.
According to (Demo et al., 2014), it is necessary to highlight how the purchase experience takes place in the context
of electronic commerce. According to (Kotler & Keller, 2012), the latest forms of interactive marketing are the
electronic channels. In this context there is a new form of client-company relationship, one in which the client starts
and controls the process more and more. Clients are the ones who define what information they need, what offers
they are interested on and which prices they are willing to pay, mainly considering how easy it is to compare prices
on the Internet.
For (Demo et al., 2014), with the purpose of stimulating new visits to their virtual stores, companies must assess the
context (layout) and content of their websites, given that, according to (Kotler & Keller, 2012), visitors evaluate the
performance of a virtual company basing on their website easiness to use and attractiveness.
According to (Anderson & Srinivasan, 2003) facing the strong competition and growing expectation on the clients’
side, electronic commerce companies are more and more interested in the identifying, understanding, taking care of
and maintaining their profitable existing clients.

Authors highlight that apparently many of their online clients compare the benefit package to be obtained to the costs
to reaching the noted value. Clients, in the long run, also assess the cost of moving onto another competitor. This
way, for a company to keep competitive it must continuously work on the noted value by the client and discourage
his moving to any other competitor.
Still according to (Anderson & Srinivasan, 2003), building trust relationships is a great challenge for companies,
even coming to giving up part of their profits to differentiate themselves from their competitors. This way companies
showing their clients that they are important and that the given companies want to help them, no matter what the
short term profit consequences are, helps create or reinforce the type of trust which results in client loyalty.
For (Gefen, 2002), reaching client loyalty is an important objective for almost every profit-oriented business. As the
author highlights, results of case studies show that reaching client loyalty depends on the capacity electronic
commerce companies have to build and keep client trust by providing good quality service.
Good quality service of an electronic commerce company may be composed by three dimensions: Tangibility: a
combination of response capacity, trustworthiness and safety and empathy. According to the author, the combination
of responsiveness, trustworthiness and guarantee is the main dimension to increase client trust (Gefen, 2002).
According to (Kotler & Keller, 2012), client satisfaction is the feeling of pleasure or disappointment on the
performance of the value noted versus his expectation. If performance falls short of his expectation, the client will be
dissatisfied; if performance meets the expectations, the client is satisfied. If performance is beyond the expectations,
the client will be charmed.
Still according to the authors, a highly satisfied or charmed client stays loyal for the long run, and even more when
the company comes to develop new products or services, besides speaking well of the company and spending less
money on competitors, including being less sensitive to price changes.
3. Final Provisions
Electronic commerce history is connected to the evolution of information technology, but more specifically to the
advances of the Internet. In this scenario, given the great advance of the Internet in recent years, any internet use
becomes a potential consumer, and portable devices as mobile phones and tablet PCs began to more used for price
researches and even for purchasing.
This new format of commerce has been successful due to the strong social demand for comfort. By its means it is
possible doing transactions 24 hours a day, 365 day per year, across demographic frontiers, supported by the
globalization phenomenon.
Despite the growing EC development, scholars state that for higher safety in this modality of commerce more
investments become necessary to provide transactions with more safety and, in this scope, get closer to that of
traditional commerce.
According to some authors, trust in this type of commerce emerges with the experience of the first purchase, and
may be maintained, increased or reduced according to the meeting of expectations by suppliers.
E-commerce has become a reality and has been developing with information technology, so some of its limitations
tend to be minimized, given the great technological advance in the present and great possibilities for development in
the future, such as the adoption of virtual intelligence, expansion of globalization with language translators, adaptive
interfaces contemplating the specific characteristics of groups of users and mobile commerce, and experimentation in
3D model.
This phenomenon is transforming the structures of organizations which are focused on becoming more competitive,
because the business world currently demands from organizations a large capacity of adequating themselves to
customers’ needs. So, the utilization of information technology as a commerce tools becomes the source of a
competitive differential, while opposition in maintaining an extremely conservative structure – resisting the
implementation of IT tools – may be damaging to organizations’ results.
This research demonstrates how e-commerce has developed in recent years, following developments in information
technology and consumer needs. This rapid development generates an expectation that for the future e-commerce
will be improved even more in a short time, by means of new technologies that stimulate the increase of the number
of fans to this modality of commerce. For organizations, being aware of e-commerce trends for the next few years,
can be a differential for their consolidation in the market.

In this way the present work, by means of rhetorical review, seeks to contribute for the formation of potential
consumers, those who may want to know and make e-commerce transactions, as well as to transmit this knowledge
to traders on EC practices, its advantages, tools, challenges and trends. Nevertheless, this article presents as a
limitation – its little depth on the theme – it suggests new works to come new researches to assess the impact of
electronic commerce evolution onto organizations’ routine and planning.
References
Albertin, A. L. (2000). O comércio eletrônico evolui e consolida-se no mercado brasileiro. Revista de Administração
de Empresas, 40(4), 94–102.
Albertin, A. L. (2012). Comércio eletrônico: da evolução para as novas oportunidades. GVexecutivo, 11(2), 66–70.
Anderson, R. E., & Srinivasan, S. S. (2003). E-satisfaction and e-loyalty: A contingency framework. Psychology &
Marketing, 20(2), 123-138.
Ban, C. A. B., Bauab, G. C., Lepiscopo, G. C. D., de Brito, J. A. N., Silva, L. L. G., & de Moraes, C. A. (2016). E-
COMMERCE E BENEFÍCIOS PROCURADOS: UM ESTUDO EXPLORATÓRIO. Jovens
Pesquisadores-Mackenzie, 12(1). Retrieved from
http://www.mackenzie.br/dhtm/seer/index.php/jovenspesquisadores/article/view/1307
Cernev, A. K., & Leite, J. C. (2005). Seguranç na Internet: a Percepção dos Usuários como Fator de Restrição ao
comércio eletrônico no Brasil. ENCONTRO ANUAL DA ASSOCIAÇÃO NACIONAL DE PROGRAMAS DE
PÓS-GRADUAÇÃO EM ADMINISTRAÇÃO, 29, 2005.
da Silveira Coelho, L., Oliveira, R. C., & Alméri, T. M. (2013). O CRESCIMENTO DO E-COMMERCE E OS
PROBLEMAS QUE O ACOMPANHAM: a identificação da oportunidade de melhoria em uma rede de
comercio eletrônico na visão do cliente. Revista de Administração do UNISAL, 3(3). Retrieved from
http://revista.unisal.br/sj/index.php/RevAdministracao/article/viewArticle/235
Demo, G., Alvarenga, B., & Guarnieri, P. (2014). Validação de uma Escala de Avaliação do E-commerce (EAE): a
Importância do Sistema Logístico e da Experiência de Compra. XVIII Encontro do ANPAD. Retrieved from
https://www.google.com.br/search?q=Valida%C3%A7%C3%A3o+de+uma+Escala+de+Avalia%C3%A7%C3
%A3o+do+E-commerce+(EAE)%3A+a+Import%C3%A2ncia+do+Sistema+Log%C3%ADstico+e+da+Experi
%C3%AAncia+de+Compra&ie=utf-8&oe=utf-8&client=firefox-b&gfe_rd=cr&ei=bIw2V52_IuWp8weW37f4
AQ
Diniz, E. (2008). Comércio eletrônico: fazendo negócios através da Internet. em manutenção, 5(2), 56–61.
Ferreira, J. B., Freitas, A. S., Giovanni, C. J., Kurtz, R. K., Pina, F. (2015). Tecnologias Interactivas: efeitos no
Varejo de Vestuário Online. Revista Eletrônica de Estratégia & Negócios, Florianópolis, 8(2), 245-262.
Galinari, R., Cervieri Júnior, O., Júnior, T., Rodrigues, J., Rawet, E. L., & others. (2015). Comércio eletrônico,
tecnologias móveis e mídias sociais no Brasil. BNDES Setorial, Rio de Janeiro, (41). Retrieved from
https://web.bndes.gov.br/bib/jspui/handle/1408/4285
Gefen, D. (2002). Customer loyalty in e-commerce. Journal of the Association for Information Systems, 3(1), 2.
Jacob, D. A., & Silva, L. A. (2015). COMÉRCIO ELETRÔNICO: ATRAÇÃO E CONQUISTA DE
E-CONSUMIDORES. INTERLINK, 2(2). Retrieved from http://187.45.244.77/ojs-
2.4.6/index.php/InterLink/article/view/14
Kiang, M. Y., & Chi, R. T. (2001). A Framework for Analyzing the Potential Benefits of Internet Marketing. J.
Electron. Commerce Res., 2(4), 157–163.
Kotler, P., & Keller, K. L. (2012). Administração de Marketing (14th ed.). São Paulo: Pearson Education do Brasil.
Lefebvre, L. A., & Lefebvre, E. (2002). E-commerce and virtual enterprises: issues and challenges for transition
economies. Technovation, 22(5), 313–323.
Lobosco, A., Alakija, A. F. da C., Zilber, S. N., & Maccari, E. (2014). A INFLUÊNCIA DO COMÉRCIO
ELETRÔNICO NAS PEQUENAS EMPRESAS DO SETOR DE MODA. Retrieved from
https://repositorio.uninove.br/xmlui/handle/123456789/560
Luciano, E. M., Testa, M. G., & Freitas, H. (2003). As tendencias em comercio eletrónico com base em recentes
congressos. ASAMBLEA ANUAL DEL CONSEJO LATINOAMERICANO DE ESCUELAS DE
ADMINISTRACIÓN-CLADEA, 38, 21-24.

Nienow, A. L. (2017). INTERFACES ADAPTATIVAS NO COMÉRCIO ELETRÔNICO COMO FACILITADORAS

DA INCLUSÃO DIGITAL DE IDOSOS. Tecnologia e Tendências, 9(2), 116-136
Novak, T.P., Hoffman, D.L., & Yung, Y.-F (2000). Measuring customer experience in online environments: a
structural modeling approach. Marketing Sciences, 19(1), 22-42.
Tesche, B. G. (2013). Estudo sobre as motivações na compra de produtos pelo comércio eletrônico. Retrieved from
http://www.lume.ufrgs.br/handle/10183/132139
Trotti, F., Burgos, G., Júnior, G., Pavão, M. D. O. P. D., & Pavão, O. (2017). Comércio eletrônico: como conquistar
clientes e aumentar o faturamento em um ambiente virtual. Revista IT-Inovação & Tecnologia, 1(1).

View publication stats
E-Commerce Product Recommendation Agents: Use, Characteristics, and

Impact
Article in MIS Quarterly · March 2007
DOI: 10.2307/25148784 · Source: DBLP
CITATIONS READS
500 5,152
2 authors:
Bo Xiao Izak Benbasat

University of Hawaiʻi at Mānoa University of British Columbia - Vancouver
54 PUBLICATIONS 908 CITATIONS 284 PUBLICATIONS 29,578 CITATIONS
SEE PROFILE SEE PROFILE
Understanding and Designing Trust in Information Systems View project
Actor Behavior and Technology Affordances in Collaborative Innovation Networks View project
All content following this page was uploaded by Bo Xiao on 17 June 2014.

Xiao & Benbasat/E-Commerce Product Recommendation Agents
THEORY AND REVIEW
E-COMMERCE PRODUCT RECOMMENDATION AGENTS:

USE, CHARACTERISTICS, AND IMPACT1
By: Bo Xiao lated to product, user, and user–RA interaction, which influ-
Sauder School of Business ence users’ decision-making processes and outcomes, as well
University of British Columbia as their evaluation of RAs. It goes beyond generalized models,
2053 Main Mall such as TAM, and identifies the RA-specific features, such as
Vancouver, British Columbia V6T 1Z2 RA input, process, and output design characteristics, that
CANADA affect users’ evaluations, including their assessments of the
bo.xiao@sauder.ubc.ca usefulness and ease-of-use of RA applications.
Izak Benbasat
Based on a review of existing literature on e-commerce RAs,
this paper develops a conceptual model with 28 propositions
Sauder School of Business
derived from five theoretical perspectives. The propositions
University of British Columbia
help answer the two research questions: (1) How do RA use,
2053 Main Mall
RA characteristics, and other factors influence consumer
Vancouver, British Columbia V6T 1Z2
decision making processes and outcomes? (2) How do RA
CANADA
use, RA characteristics, and other factors influence users’
izak.benbasat@sauder.ubc.edu
evaluations of RAs? By identifying the critical gaps between
what we know and what we need to know, this paper identifies
potential areas of future research for scholars. It also pro-
Abstract vides advice to information systems practitioners concerning
the effective design and development of RAs.
Recommendation agents (RAs) are software agents that elicit
the interests or preferences of individual consumers for pro- Keywords: Product recommendation agent, electronic
ducts, either explicitly or implicitly, and make recommenda- com-merce, adoption, trust, consumer decision making
tions accordingly. RAs have the potential to support and im-
prove the quality of the decisions consumers make when
searching for and selecting products online. They can reduce Introduction
the information overload facing consumers, as well as the
complexity of online searches. Prior research on RAs has Recommendation Agents Defined
focused mostly on developing and evaluating different under-
lying algorithms that generate recommendations. This paper Recommendation agents2 (RAs) are software agents that elicit
instead identifies other important aspects of RAs, namely RA the interests or preferences of individual users for products,
use, RA characteristics, provider credibility, and factors re-
2
1 In the literature reporting on previous research, the labels recommendation
Dov Te'eni was the accepting senior editor for this paper. Joey George agents (the terminology adopted in this paper), recommender systems,
was the associate editor. Moez Limayem, Patrick Chau, and Mark Silver recommendation systems, shopping agents, shopping bots, and comparison
served as reviewers. shopping agents have been used interchangeably.
MIS Quarterly Vol. 31 No. 1, pp. 137-209/March 2007 137

either explicitly or implicitly, and make recommendations exhibit similarities to knowledge-based systems (KBS) inas-
accordingly. RAs have been used in different areas, including much as they need to explain their reasoning to their users so
e-commerce, education, and organization knowledge manage- that they will trust the RAs’ competence and accept their
ment. In the context of e-commerce, a distinction can be made recommendations (Gregor and Benbasat 1999; Wang and
between RAs involved in product brokering (i.e., finding the Benbasat 2004a), a functionality supported by analysis and
best suited product) and merchant brokering (i.e., finding the reasoning techniques (Benbasat et al. 1991; Zachary 1986).
best suited vendor) (Spiekermann 2001). In this paper, we Additionally, RAs are designed to understand the individual
focus our attention on e-commerce product bro-kering RAs 3 needs of particular users (customers) that they serve. Users’
for supporting product search and evaluation. beliefs about the degree to which the RAs understand them
and are personalized for them are key factors in RA adoption
The study of RAs falls within the domain of information (Komiak and Benbasat 2006). Moreover, there is an agency
systems. An information technology artifact (Benbasat and relationship between the RAs and their users; users (princi-
Zmud 2003), RAs are characterized as a type of customer pals) cannot be certain whether RAs are working solely for
decision support system (DSS) (Grenci and Todd 2002) based them or, alternatively, for the merchants or manufacturers that
on the three essential elements of DSS proposed by Mallach have made them available for use. Therefore, users may be
(2000): (1) DSSs are information systems, (2) DSSs are used concerned about the integrity and benevolence of the RAs,
in making decisions, and (3) DSSs are used to support, not to beliefs that have been studied in association with trust in IT
replace, people. Similar to other types of DSS, when em- adoption models (Wang and Benbasat 2005).
ploying RAs, a customer provides inputs (i.e., needs and
constraints concerning the product desired and/or ratings on
previously consumed products) that the RAs use as criteria for Motivation, Scope, and Contribution
searching products on the Internet and generating advice and
recommendations for the customer. Due to the rapid growth of e-commerce, consumer purchase
decisions are increasingly made in an online environment. As
RAs are distinguished from traditional DSSs by several unique forecasted by Forrester research,5 online retail will reach $331
features. Specifically, the users of traditional DSSs are billion by 2010. The growing population of online shopping
generally managers or analysts employing the systems for households combined with retailer innovations and site im-
assistance in tasks such as marketing planning, logistics provements will drive e -commerce to account for 13 percent
planning, or financial planning. As such, process models 4 of total retail sales in 2010, up from 7 percent in 2004.
(which assist in projecting the future course of complex pro- Between 2004 and 2010, online sales will grow at a 15 per-
cesses) (Benbasat et al. 1991; Zachary 1986) are the primary cent compound annual growth rate. Digital marketplaces offer
decision support technologies employed by such DSSs. In the consumers great convenience, immense product choice, and a
case of RAs, in contrast, the decision makers are customers significant amount of product-related information. However,
facing a class of problems called preferential choice problems as a result of the cognitive constraints of human information
(Todd and Benbasat 1994). Thus, choice models, which processing, identifying products that meet cus-tomers’ needs is
support the integration of decision criteria across alternatives not an easy task. Therefore, many online stores have made
(Benbasat et al. 1991; Zachary 1986), are the primary decision RAs available to assist consumers in product search and
support technologies employed by RAs. RAs also selection as well as for product customi-zation (Detlor and
Arsenault 2002; Grenci and Todd 2002; O’Keefe and
McEachern 1998). By providing product advice based on user-
3
specified preferences, a user’s shopping history, or choices
In this paper, the terms RAs, product RAs, e-commerce RAs, and e- made by other consumers with similar profiles, RAs have the
commerce product RAs will be used interchangeably.
potential to reduce consumers’ infor-mation overload and
4 search complexity, while at the same time improving their
Zachary (1986) has proposed a functional taxonomy of decision support
techniques representing the six general kinds of cognitive support that decision quality (Chiasson et al. 2002; Hanani et al. 2001;
human decision makers need. The six classes are process models (which Haubl and Trifts 2000; Maes 1994).
assist in projecting the future course of complex processes), choice models
(which support integration of decision criteria across alternatives), In addition, we have seen the establishment of many success-
information control techniques (which help in storage, retrieval,
organization, and inte-gration of data and knowledge), analysis and ful comparison shopping websites, such as Shopping.com and
reasoning techniques (which support application of problem-specific
expert reasoning procedures), repre-sentation aids (which assist in
expression and manipulation of a specific representation of a decision 5
problem), and judgment amplification/refinement techniques (which help http://www.forrester.com/Research/Document/Excerpt/0,7211,34576,00.
in quantification and debiasing of heuristic judgments). html.
g MIS Quarterly Vol. 31 No. 1/March 2007

BizRate. Equipped with different types of high-end RAs, these 1.1 How does RA use influence consumer decision-
websites facilitate access to a wide range of products across making processes and outcomes?
many merchants, promising to help consumers locate the best 1.2 How do the characteristics of RAs influence con-
product at the lowest price. According to a survey by Burke sumer decision-making processes and outcomes?
(2002), over 80 percent of the respondents were enthusiastic 1.3 How do other factors (i.e., factors related to user,
about using the Internet to search for product information and product, and user–RA interaction) moderate the
to compare and evaluate alternatives. As reported by effects of RA use and RA characteristics on con-
BusinessWire,6 during the 2003 holiday season, 22 percent of sumer decision-making processes and outcomes?
shoppers began their shopping at comparison shopping sites.
The Economist (June 4, 2005, p. 11) reported that EBay (F) How do RA use, RA characteristics, and other factors
recently bought Shopping.com for $620 million, further influence users’ evaluations of RAs?
indicating the importance of recommendation tech-nologies to 2.1 How does RA use influence users’ evaluations of
e-commerce leaders. Regardless of the type of websites RAs?
(online retailers’ websites or comparison shopping websites) in 2.2 How do characteristics of RAs influence users’
which the RAs are embedded, RAs “hold out the promise of evaluations of RAs?
making shopping on the Internet better not just by finding 2.3 How do other factors (i.e., factors related to user,
lower prices but by matching products to the needs and tastes product, and user–RA interaction) moderate the
of individual consumers” (Patton 1999). effects of RA use and RA characteristics on users’
evaluations of RAs?
The design of RAs consists of three major components: input 2.4 How does provider credibility influence users’
(where user preferences are elicited, explicitly or implicitly), evaluations of RAs?
process (where recommendations are generated), and output
(where recommendations are presented to the user). Since the This review makes the following contributions to research
advent of the first RA, Tapestry, about a decade ago, research and practice:
on RAs has focused mostly on process, which consists of
developing and evaluating the different underlying algorithms *
It develops a conceptual model with supporting
that generate recommendations (Cosley et al. 2003; Swearin- proposi-tions derived from five theoretical perspectives
gen and Sinha 2002), while failing to focus on and adequately con-cerning (1) the effects of RA use on consumer
understand input and output design strategies. Similarly, the decision-making processes and outcomes, as well as
majority of the review articles regarding RAs (Herlocker et al. the effects on users’ evaluations of RAs, (2) how such
2004; Montaner et al. 2003; Sarwar et al. 2000; Schafer et al. effects are moderated by RA characteristics and other
2001; Zhang 2002) provide either evaluations of different
contingency factors, and (3) the effect of provider
recommendation-generating algorithms (focusing primarily on
credibility on users’ evaluations of RAs.
criteria such as accuracy and coverage) or taxonomies of cur-
rently available RAs (mostly in terms of the underlying algo- *
It not only compiles and synthesizes existing
rithms and techniques), without paying much attention to other knowledge from multiple disciplines that contribute to
design issues. However, from the customers’ perspec-tive, the
and shape our understanding of RAs, but also identifies
effectiveness of RAs is determined by many factors aside from
critical gaps between what we know and what we need
the algorithms (Swearingen and Sinha 2002), including the
to know, thereby alerting scholars to potential
characteristics of RA input, process, output, source credibility,
opportunities for key contributions.
and product-related and user-related factors. This study
reviews the literature on e-commerce RA design beyond *
It offers suggestions about how additional propositions
algorithms to derive propositions for iden-tifying promising
can be developed and how they can be empirically
areas for future research. This review is organized along the
investigated.
following research questions:
*
It provides advice to IS practitioners concerning the
ix) How do RA use, RA characteristics, and other factors effective design and development of RAs.
influence consumer decision-making processes and
outcomes? The unit of analysis for this review is an instance of using the
RA. This paper reviews previous theoretical and empirical
studies of RAs, as defined in this paper, in both online and
6
http://www.websearchguide.ca/newsletter/031129.htm. offline shopping environments. The review covers a period
MIS Quarterly Vol. 31 No. 1/March 2007 139

from the early 1990s, when the first RAs came into being, to Conceptual Model and Constructs
the present. A search of literature in information systems,
marketing, consumer research, computer science, decision Conceptual Model
science, and human computer interaction was undertaken to
identify the relevant studies. Databases for published journal The conceptual model for this review is depicted in Figure 1. 10
articles, conference proceedings, and unpublished disserta- The key constructs of the conceptual model, from right to left,
tions, theses, and working papers were searched by using are outcomes of RA use (consisting of consumer decision
relevant keywords. Leading journals and conference pro- making and users’ evaluations of RAs), product, user, user–
ceedings were also scanned by their table of contents. The RA interaction, RA characteristics, provider credibility, and
criteria for including a particular empirical study7 are RA use. They are identified based on previous conceptual and
empirical research in information systems in general, and
*
The study must be “empirical” in the sense that the study decision support systems and RAs in particular, as discussed in
has to involve actual use of an RA (prototype or opera- the following sections. To explicitly indicate the connect-ion
tional, web-based or stand-alone) by human users (similar between the conceptual and the research questions stated
to the definition of empirical studies in Gregor and earlier, relationships among the constructs are marked with the
corresponding research question number(s). This high level
Benbasat 1999) in either online or physical settings.8
conceptual model will be decomposed into more focused,
*
The study must have dependent variables that go lower level models in the “Propositions” section to map the
relationships among different constructs (and ele-ments of the
beyond “accuracy” and “coverage,”9 which are the
constructs).
variables commonly investigated as dependent
variables in algorithm-focused research. This paper focuses on two classes of outcomes associated with
RA use. First, it intends to enquire into the differences in
In the next section, the conceptual model is introduced and decision making (1) between consumers who are assisted by
the constructs that make up the model are defined. Propo- RAs and those who are not, as well as (2) between con-sumers
sitions concerning the relationships among the constructs using different types of RAs or RAs with differing design
are then presented and evidence from empirical studies is characteristics. Hence, the construct consumer deci-sion
intro-duced to test the propositions. In the final section, making is included in the conceptual framework to investigate
recom-mendations to practitioners and researchers are how the different groups of consumers differ in their decision
provided and directions for future work are suggested. making processes and outcomes. Second, this paper aims to
examine users’ perceptions of RAs, represented by the
construct users’ evaluations of RAs. Since RAs are a particular
type of information system, we focus on the two dominant
approaches to examining user perceptions of infor-mation
system success—user satisfaction and IT acceptance (Wixom
7
The empirical studies reviewed in this paper are listed in alphabetical order by
and Todd 2005)—hence including satisfaction and TAM
authors’ names in Appendix A. For each study, the theoretic perspectives (technology acceptance model) constructs (perceived
drawn, the context of the study (e.g., research method, RA used, tasks), the usefulness and perceived ease of use) in the conceptual model.
variables included, and the results that were obtained are described. In addition to being information technology artifacts, RAs are
8 also trust objects (Gefen et al. 2003; Wang and Benbasat
Many researchers (e.g., Ariely et al. 2004; Breese et al. 1998; Herlocker et al.
2004) in the field of RA have published results evaluating the accuracy and/or
2005), that is, consumers must have confidence in the RAs’
coverage of different recommendation algorithms. However, most of these product recommendations, as well as in the processes by
studies make conclusions based on either simulations or post hoc analysis which these recommendations are generated (Haubl and
(offline studies based on preexisting databases) and do not include users in the Murray 2003). Furthermore, the relationship between RAs and
evaluation process. For the purpose of this paper, simulations and offline
studies on preexisting datasets are excluded from the review. their users can be described as an agency relationship. By
delegating the task of product screening and evaluation to
9
Accuracy and coverage are the two most widely used metrics for
evaluating recommendation algorithms, particularly for collaborative-
filtering algo-rithms. Accuracy measures how close the RA’s predicted 10
ratings are to the true user ratings. Coverage of an RA is a measure of the Two additional constructs, namely, intention for future use and future use of
domain of items in the system over which the system can form predictions RA, as well as the relationships (1) between the two groups of outcome
or make recom-mendations (Herlocker et al. 2004). Both metrics measure variables, (2) among the outcome variables in each group, (3) between outcome
what system designers believe will affect the utility of an RA to the user variables and intention for future use, and (4) between intention for future use
and affect the reaction of the user to the system, instead of measuring and future use of RA are not part of our conceptual model. They are, however,
directly how actual users react to an RA. included in Figure 1 to show compatibility with prior research.
- MIS Quarterly Vol. 31 No. 1/March 2007

User User-RA
Interaction Outcomes of RA Use
- Product
expertise ƒ User-RA similarity
- Perceived ƒ User’s familiarity Consumer Decision
product risks with RA Making
ƒ Confirmation/
disconfirmation
of expectations Decision Processes
Q1.3 Q1.3
Q2.3 Q2.3 Decision Outcomes
RA Use Q1.1
Q2.1
Intention for
Future Use
RA
Characteristics User Evaluation of RAs
Q1.2
Future Use
- RA Type Q2.2 Trust
of RA
- Input
- Process Perceived Usefulness
- Output
Q1.3
Q2.3
Perceived Ease of
Provider Use
Product Credibility Q2.4
Satisfaction
- Product type ƒ Type of RA
- Product provider
complexity ƒ Reputation of
RA provider
Note: Solid lines indicate relationships and constructs investigated in this paper.
Figure 1. Conceptual Model
RAs (i.e., the agents), users assume the role of principals. spondence between the constructs proposed in this review and
Thus, as a result of the information asymmetry underlying those included in the two earlier frameworks. Abstracting from
any agency relationship, users usually cannot determine the two studies, we include RA characteristics, product, user,
whether RAs are capable of performing the tasks delegated user–RA interaction, and provider credibility11 in the model as
to them, and whether RAs work solely for the benefit of the important determinants of both RA-assisted consumer decision
users or the online store where they reside. As such, trust is making and user evaluation of RAs.
also an integral part of the conceptual framework.
Eierman et al. (1995) identified six broad DSS constructs (i.e.,

11
environment, task, implementation strategy, DSS capability, As advice-giving information systems, RAs are effective to the extent that the
users follow their recommendations. Prior research has recognized the
DSS configuration, and user) that affect user behavior and
importance of source credibility in determining users’ acceptance of the advice
DSS performance. Brown and Jones (1998) also recognized from consultative knowledge based systems (Fitzsimons and Lehmann 2004;
decision aid features, decision-maker characteristics, and Mak and Lyytinen 1997; Pornpitakpan 2004). Insomuch as current RAs are
provided by and embedded in websites, the credibility of the websites is an
decision-task characteristics as important factors influencing
indicator of source credibility. It is therefore included in our conceptual
users’ reliance on decision aids. Table 1 illustrates the corre- framework and represented by the construct provider credibility.

Table 1. Selection of Constructs
Sources
Constructs Eierman,
Included in Brown and Niederman,
Conceputal Jones and Adams Reasons for Including the Constructs
Model (1998) (1995) in the Conceptual Model
The correspondence between RA characteristics (in this study),
RA Decision aid DSS
decision aid features (Brown and Jones 1998), and DSS con-
Characteristics features configuration
figuration (Eierman et al. 1995) is apparent.
The correspondence between user (in this study), decision-
maker characteristics (Brown and Jones 1998), and user
(Eierman et al. 1995) is apparent.
User;
Decision-maker
User–RA User The construct user–RA interaction (in this study), as the name
characteristics
Interaction suggests, captures the relationship between RA and user (e.g.,
user–RA similarity) as well as users’ experience with RA (e.g.,
user’s familiarity with RA, confirmation/disconfirmation of
expectation)
In this study, “task” is fixed, namely, to purchase products with/
Decision task Task without an RA. Since “product” is an integrated component of
Product
characteristics the buying task, its type (i.e., search or experience) and com-
plexity may affect the effectiveness of the RAs.
According to Eierman et al. (1995), environment is the setting in
which the DSS is developed and used. Since RAs are typically
Provider
Environment embedded in different websites, referred to as RA provider in
Credibility
this paper, the credibility of such RA provider will influence
users’ evaluations of the RA.
Finally, the use of RAs is a necessary condition that affects beliefs (based on information provided by an outside source).
consumers’ decision making effectiveness and users’ evalua- In line with Fishbein and Ajzen’s account, in the context of
tions of the RAs. Although the intent to use IS is modeled as a RAs, beliefs about RAs may come from two sources:
main dependent variable in most IS adoption research, based 348 from perceptions acquired from other sources,
on TAM, we explicitly include RA use as an independent colleagues, newspapers, etc. (if people have not used RAs yet),
variable in the conceptual model for two reasons. First, initial or
or trial RA use serves as a context for the majority of empirical 349 from individuals’ direct experiences based on RA
studies included in this review. TAM specifies that perceived use. In this review paper, we are dealing with the second
usefulness (PU) and perceived ease of use (PEOU) are two case only. In almost all of the studies reviewed in this
salient beliefs determining users’ behavioral inten-tions (i.e., paper, the participants are using an RA for the first time.
about future use). Although TAM is silent about where these This use of RAs constitutes direct experience through
two beliefs come from, the basic concept under-lying TAM as which descriptive beliefs about the RAs are formed.
well as other user acceptance models is that an individual’s Indeed, in many prior studies validating TAM (e.g., Davis
reactions to using IT/IS contribute to the inten-tion to use 1989; Davis et al. 1989), subjects were allowed to interact
IT/IS, which, in turn, determines the actual use of IT/IS with a system for a brief period of time before their beliefs
(Venkatesh et al. 2003). Fishbein and Ajzen (1975) have about the system and intention to use the system were
differentiated three types of beliefs based on three dif-ferent measured, although such interaction with the system (or
belief formation processes: descriptive beliefs (based on direct system use) was not explicitly included in the conceptual
experiences), inferential beliefs (based on prior descriptive model. This paper explicitly models RA use as an
beliefs or a prior inference), and informational independent variable to account for the results of the many
experimental studies included in our review.
*
MIS Quarterly Vol. 31 No. 1/March 2007
Second, prior IS research (e.g., Bajaj and Nidumolu 1998; processes and outcomes and their evaluations of RAs
Kim and Malhotra 2005a; Limayem et al. 2003) has investi- (Swaminathan 2003).
gated the effects of initial or prior IS use on users’ evaluations
and their subsequent use or adoption of the IS. Bajaj and Factors related to user (see Table 4 for definitions) include
Nidumolu (1998) have demonstrated that past use itself could consumers’ characteristics, such as their product expertise
be a basis for the formation of user evaluations (i.e., perceived and perceptions of product risks. Similar to product-related
ease of use) at a subsequent stage. Limayem et al. (2003) fac-tors, user -related factors have been investigated as an
explicitly incorporate initial IS use in an integrative model ante-cedent of both objective and subjective measures of
explaining subsequent IS adoption and post-adoption. The RA effec-tiveness (Senecal 2003; Spiekermann 2001;
results of a multistage online survey revealed that initial IS use Swaminathan 2003; Urban et al. 1999).
confirmed or disconfirmed users’ prior expectations, thus
affecting users’ satisfaction with the IS and, subsequently, their Factors related to user–RA interaction (see Table 4 for defi-
intention to continue the use of the IS. Kim and Mal-hotra nitions) include user–RA similarity (i.e., the user’s similarity
(2005b) have developed a longitudinal model of how with RAs in terms of ratings, goals, and needs, decision
individuals form (and update) their evaluations of an IT strategy, and attribute importance), user’s familiarity with RAs
application and adjust their system use over time. The model through repeated use, and confirmation/disconfirmation of the
includes two instances of the IS use construct (use at t = 0 – 1 user’s prior expectations related to RAs.
and Use at t = 1 – 2), representing the use of the IS during two
different time periods. In a two-wave survey in the context of In Figure 1, factors related to product, user, and user–RA
web-based IS use, they found that prior use of a personalized interaction are modeled as moderators of the effects of RA
web portal influenced users’ evaluations of the portal (i.e., the use and RA characteristics on outcomes of RA Use (i.e.,
perceived usefulness, perceived ease of use, and future use consumer decision making and users’ evaluations of RA).
intention), which in turn influenced the users’ future use of the
portal. Provider credibility (see Table 4 for definitions) refers to
users’ perception of the credibility of the providers of RAs.
The two reasons presented above justify our treatment of This is influenced by the type of websites in which the RAs
RA use as an independent variable in the model. are embedded (e.g., sellers’ websites, third party websites
commercially linked to sellers, or third party websites not
commercially linked to sellers) (Senecal 2003; Senecal and
Nantel 2004), as well as by the reputation of the websites.
Definitions of Constructs
The credibility of RAs’ providers has a direct impact on
users’ evaluations of the RAs.
Although the constructs in Figure 1 have been labeled,
mea-sured, and applied differently in different studies, RA characteristics (see Table 5 for definitions) include RA
some com-mon definitions of the constructs are provided
type as well as features and characteristics of RAs related
below to facilitate the comparison and synthesis of the
to the following stages:
results from various studies.
350 Input (the stage where users’ preferences are elicited)
Outcomes of RA use includes two general classes of conse-
quences associated with the application of RAs in decision 351 Process12 (the stage where recommendations are gener-
making. Consumer decision making (see Table 2 for defini- ated by the RAs)
352 Output (the stage where RAs present recommendations
tions) is a broad construct referring to decision processes,
to the users).
including consumers’ product evaluations, preference func-
tions, decision strategies, and decision effort, and decision RA characteristics have been shown to influence the
outcomes, including consumers’ choice and decision customers’ decision-making processes and outcomes, as well
quality (both subjective and objective). Users’ evaluations
as their evaluation of RAs, which in turn influence their
of RAs (see Table 3 for definitions) refers to users’
behavioral intention to adopt RAs. While a majority of the
perceptions and assessment of RAs, including their trust in
factors relate to both content-filtering and collaborative-
RAs, perceptions of RAs’ usefulness and ease of use, and
filtering RAs, some (e.g., product attributes included in the
satisfaction with RAs.
Factors related to product (see Table 4 for definitions) include

product type and product complexity. Product-related factors 12
RA recommending algorithms are not a focus of this paper and therefore
have been shown to influence both users’ decision-making not included in Table 5.

Table 2. Variables Associated with Consumer Decision Making

Variables Definitions
Decision processes
Decision effort The amount of effort exerted by the consumer in processing product information, evaluating
product alternatives, and arriving at choice decision.
• Decision time • The time taken for the consumer to search for product information and make a purchase
decision.
• Extent of product search • The number of product alternatives that have been searched, for which detailed information
is acquired, and have seriously been considered for purchase by the consumer.
• Amount of user input • The amount of preference information provided by the user prior to receiving
recommendations.
Preference functions The consumer’s attribute-related preferences (e.g., product attribute importance weight).
Product evaluation The consumer’s evaluation (e.g., ratings) of the product alternatives recommended by the
RA.
Decision strategies The strategies with which the consumer evaluates product alternatives to arrive at product
choice.
Decision Outcomes
Choice The consumer’s final choice from the products in the alternative set.
Decision quality (objective) The objective quality of the consumer’s purchase decision, indicated by such measures as:
• Preference matching • Calculated score of how the chosen alternative matches the consumer’s preferences
score
• Quality of consideration • Averages quality of the alternatives that the consumer considers seriously for purchase.
set
• Choice of non-dominated • Whether the product chosen by the consumer is a dominant or dominated alternative within
alternative(s) the context of the whole set of products she selects (when there exists a dominant product)
or on a particular attribute dimension (when there is no dominant product).
• Product switching • Whether the consumer, after making a purchase decision, changes her mind and switches
to another alternative when given an opportunity to do so.
Decision quality (subjective) The subjective quality of the consumer’s purchase decision, indicated by such measure as
• Confidence • The degree of a consumer’s confidence in the RA’s recommendations.
Table 3. Variables Associated with Users’ Evaluations of RAs

Variables Definitions
Trust The user beliefs in the RA’s competence, benevolence, and integrity. The beliefs that
• Competence • the RA has the ability, skills, and expertise to perform effectively
• Benevolence • the RA cares about the user and acts in the user’s interest
• Integrity • the RA adheres to a set of principles (e.g., honesty and promise keeping) that the user finds
acceptable
Satisfaction The user’s satisfaction with RA and her decision-making process aided by the RA.
Perceived usefulness The user’s perceptions of the utility of the RA or the RA’s recommendations.
Perceived ease of use The user’s perceptions of the effort necessary to operate the RA.
352 MIS Quarterly Vol. 31 No. 1/March 2007

Table 4. Factors Related to Product, User, User–RA Interaction, and Provider Credibility
Factors Definitions
Product type Whether a product is a search product or an experience product. Search product is
characterized by attributes that can be assessed based on the values attached to their
attributes, without necessitating the need for the user to experience the products directly.
Produc
Experience product is characterized by attributes that need to be experienced prior to

t
purchase.
Product complexity Product complexity is defined along four dimensions: the number of product alternatives,
the number of product attributes, variability of each product attribute, and inter-attribute
correlations.
Product expertise The user’s knowledge about or expertise with the intended product.
User
Perceived product The user’s perceptions of uncertainty and potentially adverse consequences of buying a
risks recommended product.
User–RA similarity Similarity between the user and the RA in terms of past opinion agreement, goals, decision
User–RAInteraction
strategies, and attribute weighting.
User’s familiarity with The user’s familiarity with the workings of RAs through repeated use.
RAs
Confirmation/disconfir- Consistency/inconsistency between the user’s pretrial expectations about the RA and the
mation of expectations actual performance of the RA.
Provider credibility The user’s perception of how credible the provider of an RA is.
ProviderCredibilit
y
• Type of RA provider • The type of website in which the RA is embedded.

• Reputation of RA • The reputation of the website in which the RA is embedded.
provider
RAs’ preference-elicitation interface) pertain only to consumers shopping for products as well as users of an IT
content-filtering ones. RA characteristics are modeled as artifact. Accordingly, there have been two streams of empirical
having direct effect on outcomes of RA use.13 research on RAs, one focusing on consumers’ deci-sion-
making processes and outcomes with the assistance of RAs,
RA use refers to the application of RAs to assist in and the second focusing on users’ subjective evaluation of
shopping decision making. In most of the empirical studies RAs. This paper also adopts this dual focus. In the following
reviewed in this paper, RA use is an independent variable subsections, propositions concerning the effects of RA use, RA
implemented by comparing use to nonuse of RAs.14 RA use characteristics, and other factors (i.e., those related to product,
is also binary in our research model. user, user–RA interaction, and provider credibility) on
consumers’ decision making processes and outcomes, as well
as on their evaluation of RAs are derived from five theoretical
Propositions perspectives15: (1) theories of human information processing,
(2) the theory of interpersonal simi-larity, (3) the theories of
When individuals engage in online or offline shopping with trust formation, (4) the technology acceptance model (TAM),
and (5) the theories of satisfaction. Whereas propositions
the assistance of web-based RAs, they are simultaneously
related to RA-assisted consumer deci-sion making are
primarily derived from theories of human in-formation
processing, those concerning users’ evaluations of RAs and
13
It should be noted that the effects of RA characteristics on the outcomes their adoption intention are developed based on the
of RA use can only be realized when the RAs are actually used.
14
Typically, subjects were randomly assigned into two different groups: 15
one group was instructed to use an RA to assist with the shopping task A brief introduction to the five theoretical perspectives is provided in
while the other group was not provided an RA. Appendix B.

Table 5. Recommendation Agent Characteristics

Factors Definitions
RA type Type of RAs
Content-filtering vs. Content filtering RAs generate recommendations based on product attributes the consumer likes;
collaborative filtering vs. collaborative filtering RAs mimic “word-of-mouth” recommendations and use the opinions of like-
hybrid minded people to generate recommendations; hybrid RAs integrate content filtering and
collaborative filtering methods in generating recommendations.
Type
Compensatory vs. non- Compensatory RAs allow tradeoffs between attributes. All attributes simultaneously contribute to
compensatory computation of a preference value; non-compensatory RAs do not consider tradeoffs between
RA
attributes.
Feature-based vs. Feature-based RAs ask the consumer to specify what features she wants in a product and then
needs-based vs. hybrid help the consumer narrow the available choices and recommend alternatives; needs-based RAs
ask the consumer to provide information about herself and how she will use the product, and then
recommend alternatives from which the consumer can make a choice; hybrid RAs allow the
consumer to specify both desired product features and product-related needs.
Preference elicitation Whether feature-based or needs-based preference elicitation method is used.
method Whether explicit or implicit preference elicitation method is used.
Included product What attributes of the product are included in the RA’s preference-elicitation interface
attributes
Input
Ease of generating Ease for the user to generate new/additional recommendations (e.g., is the user required to repeat
new/additional the entire rating or question-answer process to see new recommendations? Can the user modify
recommendations their previous answers/ratings to generate new recommendations?).
User control The amount of control the user has over the interaction with the RA (e.g., the user may choose
what and how many preference-elicitation questions to answer).
Information about Whether or not the RA provides information about search progress (e.g., what percentage of the
Process
search progress database has the search engine reviewed and what percentage remains to be reviewed).
Response time Amount of time elapsed for the user to receive recommendations (after providing ratings or
answers to preference-elicitation questions).
Recommendation What is presented to the users at output stage?
content
• Recommendations • The product recommendation generated by the RA
• Utility scores or • Whether or not the RA displays utility scores or predicted product ratings (calculated on the
predicted ratings basis of the user’s profile) for the products recommended by the RA
• Detailed information • Whether or not the RA provides detailed information about recommended items (e.g., detailed
about recommended item-specific information, pictures, community ratings)
products
• Familiar recom- • Whether or not the list of recommendations contains familiar products
mendations
Output
• Composition of the
dations
• The composition of the list of recommendations presented by the RA (e.g., whether there is a
list of recommen- balance of both familiar and novel recommendations)
• Explanation • Whether or not explanations are provided on how the recommendations are generated by the
RA
Recommendation How is the recommendation content presented to the users output stage?
format
• Recommendation • Whether the recommendations are displayed in sorted or non-sorted list
display method
• Number of • The number of recommended products displayed by the RA
recommendations
• Navigation and layout • Navigational path to product information and layout of the product information

theories of trust formation, TAM, and satisfaction. The The Impact of RA Use on Decision Processes. In complex
theories of interpersonal similarity are drawn upon to decision-making environments, individuals are often unable to
explain phenomena in both streams of research. evaluate all available alternatives in great depth prior to
making their choices due to their limited cognitive resources.
The discussion in each subsection follows a common struc- According to Payne (1982; see also Payne et al. 1988), the
ture: complexity can be reduced with a two-stage decision-making
process, in which the depth of information processing varies
356 A brief introduction of the subsection is provided. by stage. At the first stage (i.e., the initial screening stage),
individuals search through a large set of available alternatives,
357 Each proposition is advanced as a triplet of (1) acquire detailed information on select alternatives, and iden-
theoretical explanations and past empirical findings (in tify a subset (i.e., the consideration set) of the most promising
areas other than RAs) that provide rationale for the candidates. Subsequently (i.e., at the in-depth comparison
proposition; (2) the proposition (and sub-propositions); and stage), they evaluate the consideration set in more detail,
(3) existing empi-rical findings in RA research (if available) performing comparisons based on important attributes before
that are used to support (or qualify) the proposition. committing to an alternative (Edwards and Fasolo 2001; Haubl
and Trifts 2000). Since typical RAs facilitate both the initial
358 The research question relevant to the discussion in screening of available alternatives and the in-depth
that subsection is answered. Propositions and relevant comparison of product alternatives within the consideration
empiri-cal findings for the subsection are summarized in a set, they can provide support to consumers in both stages of
table. the decision-making process.
Table 6 illustrates the two-stage process of online shopping

decision making, with and without RAs, as well as the tasks
Consumer Decision Making performed by RAs and by consumers at each stage. It also
shows the different alternative sets16 involved in the two-stage
Consumer behavior is defined as the “acquisition, consump-
decision-making process: (1) the whole set of products
tion, and disposition of products, services, time and ideas by
available in the database(s) (i.e., the complete solution space),
decision making units” (Jacoby 1998, p. 320). Major domains
[35] the subset of products that is searched (or search
of research in consumer behavior include information
set),
processing, attitude formation, decision making, and factors
[36] the subset of alternatives for which detailed
(both intrinsic and extrinsic) affecting these processes (Jacoby
information is acquired (or in-depth search set), (4) the subset
1998). This paper focuses on consumer decision making
of alterna-tives seriously considered (or consideration set),
related to the acquisition or buying of products, and separates
and the
the outcomes of decision making from the processes of deci-
[39] the alternative chosen. As one moves from each
sion making. What follows is a discussion of how RA use, RA
set to the next—that is, from (2) to (3) or from (3) to (4)—
characteristics, and other contingencies affect consumer
some form of effortful information-processing occurs.
decision making processes and outcomes. The guiding theories
for the development of propositions P1 to P13 are the theories RA use affects the consumer’s decision-making process by
of human information processing and the theory of influencing the amount of effort exerted during the two stages
interpersonal similarity. The propositions provide answers to (as well as the individual phases in each stage) of the shopping
our first research question: How do RA use, RA charac- decision-making process as illustrated in Table 6.
teristics, and other factors influence consumer decision- Dictionary.com defines effort as “the use of physical or mental
making processes and outcomes? The relationships investi- energy to do something.” Consumer decision effort, in online
gated in this section are depicted in Figure 2.
shopping context, refers to the amount of effort exerted by the
consumer in processing information, evaluating alternatives,
and arriving at a product choice. It is usually measured by
RA Use
decision time and the extent of product search. Decision time
refers to the time consumers spend searching for product
The application of RAs to assist in consumers’ shopping task
information and making purchase decisions. Since RAs
influences their decision-making processes and outcomes. As
assume the tedious and processing intensive job of
predicted by the theoretical perspective of human information
processing, and represented in propositions P1 and P2, when
supported by RAs in decision making, consumers will enjoy
improved decision quality and reduced decision effort. 16
The categorization of alternative sets is based on the comments provided
by one of the anonymous reviewers.

User-RA User Product

Interaction • Product • Product
• User-RA expertise type
similarity • Perceived • Product
product risk complexity
Outcomes of RA Use
Consumer Decision
P13 P12 P9 Making
P1
P8 Decision Processes
RA Use P13 P12 P9
P2 • Decision effort
• Preference function
• Product evaluation
RA Characteristics
• Decision strategies
RA Type
[43] Content- P3, P4, P5, P6, P7
filtering/
collaborative-
filtering vs. hybrid
[44] Compensa Decision Outcomes
tory vs. non-
P11 P10
compensatory • Decision quality
[45] Feature- P3, P4, P5, P6, P7
based vs. needs-
• Choice
based
Input
[47] Preference
elicitation method
[48] Included
product attributes
Process
Output
• Recommendation
content
• Recommendation
format
Figure 2. Effects of RA Use, RA Characteristics, and Other Factors on Consumer Decision Making
Table 6. Two-Stage Process of Online Shopping Decision Making

Alternative
Tasks Performed by Consumers Tasks Performed by RA Sets
Searches through a large set of relevant
products, without examining any of them in Search set
OnlineShoppingWithoutRA
great depth
Stage 1 Acquires detailed information on a select In-depth search
Initial Screening set of alternatives set
Identifies a subset of alternatives that
includes the most promising alternatives
for in-depth evaluation Consideration
Performs comparisons across products (in set
Stage 2 previously identified subset) on important
In-Depth attributes
Comparison
Makes a purchase decision Final choice
Indicate preferences in terms of product
features or ratings
Screens available products in the
database to determine which ones
are worth considering further and
Complete
presents a list of products sorted
by their predicted attractiveness to solution space
with RA
Stage 1 the customer (based on the

Initial Screening customer’s expressed preferences)
Searches through the list of
Shopping
Search set
recommendations
Acquires detailed information on a select In-depth search
set of alternatives set
Identifies a subset of products (that
Online
includes the most promising alternatives)

to be included in comparison matrix
The comparison matrix organizes Consideration
Performs comparisons across products (in attribute information about multiple set
Stage 2 previously identified subset) on select products and allow for side-by-side
In-Depth attributes comparisons of products in terms
Comparison of their attributes
Makes a purchase decision Final choice
screening and sorting products based on consumers’ ex- shop without RAs, those who use RAs are expected to
pressed preferences, consumers can reduce their information search through and acquire detailed information on fewer
search and focus on alternatives that best match their pre- alternatives (i.e., only those close to the top of the ordered
ferences, resulting in decreased decision time. list), resulting in a smaller search set, in-depth search set,
and consideration set. Thus, the use of RAs is expected to
The extent of product search refers to the number of product reduce the extent of consumers’ product search by reducing
alternatives that have been searched, for which detailed the total size of alternative sets as well as the size of the
information is acquired, and have seriously been considered search set, in-depth search set, and consideration set.
for purchase by consumers. Thus, a good indicator for the
extent of product search is the size of the alternative sets (as An additional indicator of consumer decision effort, the
illustrated in Table 6) and, in particular, the size of the search amount of user input, occurs during RA-assisted online
set, the in-depth search set, and the consideration set. Since shopping (see Table 6). The amount of user input refers to the
RAs present lists of recommendations ordered by predicted amount of preference information (e.g., desired product
attractiveness to consumers, compared to consumers who features, importance weighting, and product ratings) provided

by the users prior to receiving recommendations. Such dominated (an optimal decision) or dominated (a suboptimal
deci-sion effort will not be incurred by consumers shopping decision) alternative (Diehl 2003; Haubl and Trifts 2000;
without RAs. It is therefore proposed that Swaminathan 2003), (2) as a calculated preference matching
score of the selected alternatives, which measures the degree
P1: RA use influences users’ decision effort. to which the final choice of the consumer matches the
preferences she has expressed (Pereira 2001), (3) by the
P1a: RA use reduces the extent of product search by quality of consideration set (Diehl 2003), averaged across
reducing the total size of alternative sets processed individual product quality, (4) by product switching, that is,
by the users as well as the size of the search set, in- after making a purchase decision, when given an opportunity
depth search set, and consideration set. to do so, if a customer wants to change her choice and trade
her initial selection for another (Haubl and Trifts 2000;
P1b: RA use reduces users’ decision time. Swaminathan 2003), and (5) by the consumer’s confidence in
her purchase decisions or product choice (Haubl and Trifts
P1c: RA use increases the amount of user input. 2000; Swaminathan 2003).
Dellaert and Haubl (2005) found that RA use reduced the The typical decision maker often faces two objectives: to
number of products subjects looked at in the course of their maximize accuracy (decision quality) and to minimize effort
search (i.e., the total size of the alternative sets). Similar (Payne et al. 1993). These objectives are often in conflict,
results were obtained by Moore and Punj (2001). Haubl and since more effort is usually required to increase accuracy.
Trifts (2000) observed that consumers who shopped with the Since RAs perform the resource-intensive information pro-
assistance of RAs acquired detailed product information on cessing job of screening, narrowing, and sorting the available
significantly fewer alternatives (i.e., the in-depth search set) alternatives, consumers can free up some of the processing
than did those who shopped without RAs. Haubl and his capacity in evaluating alternatives, which will allow them to
colleagues (Haubl and Murray 2006; Haubl andTrifts 2000) make better quality decisions. Moreover, RAs enable con-
found that the use of RAs led to a smaller number of alter- sumers to easily locate and focus on alternatives matching
natives seriously considered at the time the actual purchase their preferences, which may also result in increased decision
decision was made. However, there are some contradictory quality. It is therefore proposed that
reports concerning the effects of RA use on the size of consi-
deration set. Pereira (2001) observed that the use of RAs P2: RA use improves users’ decision quality.
significantly increased the set of alternatives the subjects
seriously considered for purchase in the first stage of the Pereira (2001) observed that query-based RAs improved
phased narrowing process. Pedersen (2000) found no signifi- consumers’ decision quality, measured both objectively by the
cant effect of RA use on the size of the consideration sets preference matching score of the chosen alternative and sub-
(based on the subjects’ self-reports). Swaminathan (2003) jectively by customers’ confidence in their decision. Similar
observed that the effect of RA use on the size of consideration results were also obtained by Dellaert and Haubl (2005).
set was moderated by product complexity. Haubl and his colleagues (Haubl and Murray 2006; Haubl and
Trifts 2000) found that RAs led to increased decision quality,
Concerning decision time, a few studies (Hostler et al. namely, a decrease in the proportion of subjects who pur-
2005; Pedersen 2000; Vijayasarathy and Jones 2001) noted chased non-dominated alternatives and a decrease in the pro-
that, compared to those who did not utilize RAs, RA users portion of subjects who switched to another alternative when
spent significantly less time searching for information and given an opportunity to do so. Olson and Widing (2002) also
com-pleting the shopping task. However, Olson and Widing observed that the use of RAs resulted in a lower proportion of
(2002) observed that consumers who used RAs had longer subjects who switched from their actual choice to the com-
actual decision time and perceived decision time. They puted best choice as well as a greater confidence in their
explained that the benefit of less information processing choices. Van der Heijden and Sorensen (2002) showed that the
time may be offset by the extra time required to enter use of RAs increased the number of non-dominated alter-
product attribute importance weights for the RAs. natives in the consideration set as well as consumers’ decision
confidence. Hostler et al. (2005) found that RAs increased
The Impact of RA Use on Decision Outcomes. Decision users’ objectively measured decision quality, but had no effect
quality refers to the objective or subjective quality of a con- on their decision confidence. The predicted impact for RAs on
sumer’s purchase decision. It is measured in various ways: decision quality, measured by subjects’ choices of non-
(1) whether a product chosen by a consumer is a non- dominated products, was not borne out in another study

Table 7. The Impact of RA Use on Decision Process and Decision Outcomes

The propositions in this table provide answers to research question 1.1.
Q1.1: How does RA use influence consumer decision making processes and outcomes?
Relationship Between Empirical Support P
Decision Effort P1
• Extent of product RA use reduced the total number of products subjects examined (Dellaert and • P1a
search Haubl 2005; Moore and Punj 2001); RA use reduced the number of products
about which detailed information are obtained (Haubl and Trifs 2000); RA use
led to smaller consideration sets (Haubl and Murray 2006; Haubl andTrifts
2000); RA use led to larger consideration sets (Pereira 2001); RA use had no
effect on self report consideration set size (Pedersen 2000); The effect of RA
RA Use
use on the size of consideration set size was moderated by product complexity
(Swaminathan 2003).
• Decision time RA users spent less time searching for information and completing the shopping • P1b
task (Hostler et al. 2005; Pedersen 2000; Vijayasarathy and Jones 2001); RA
users had longer actual and perceived decision time (Olson and Widing 2002).
No empirical study available

• Amount of user input • P1c
RA use improved consumers’ decision quality, in terms of preference matching
scores (Dellaert and Haubl 2005; Hostler et al. 2005; Pereira 2001), confidence
in decision (van der Heijden and Sorensen 2002; Olson and Widing 2002;
Pereira 2001), choice of non-dominated alternatives (Haubl and Murray 2006;
RA Use
Decision Quality Haubl and Trifts 2000; van der Heijden and Sorensen 2002), product switching P2
(Haubl and Murray 2006; Haubl and Trifts 2000; Olson and Widing 2002); RA
use had no impact on decision confidence (Hostler et al. 2005); RA use had no
impact on decision quality (Swaminathan 2003); RA use reduced decision
confidence (Vijayasarathy and Jones 2001).
(Swaminathan 2003). This was attributed to the lack of in support of the positive effects of RA use on decision
time pressure in the study. The longer time given to quality and decision effort is still inconclusive. Table 7
subjects in the control group (the group without RAs) to summarizes the propositions as well as the available
search for informa-tion might have resulted in better empirical support for these propositions.
decision quality even in the absence of RAs. Vijayasarathy
and Jones (2001) noted a negative relationship between the
use of decision aids and decision confidence, attributing RA Characteristics
this outcome to users’ lack of trust (which is discussed later
in the paper17) in the decision aids. All RAs are not created equal. As such, the effects of RA use
on consumer decision making are determined, at least in part,
Summary. The two propositions advanced above help answer by RA characteristics (i.e., RA type and characteristics asso-
research question (1.1): How does RA use influence con-sumer ciated with the input, process, and output design). In the
decision-making processes and outcomes? Using RAs during following subsections, guided by the theories of human infor-
online shopping is expected to improve consumers’ decision mation processing, particularly the effort-accuracy framework
quality while reducing their decision effort. While many and the constructed preferences perspective, we develop
studies have shown that RA use did result in improved propositions P3 through P7 (illustrated in Figure 2) con-
decision quality and decreased decision effort, there also exists cerning the effects of RA characteristics on consumers’
some counter evidence. Thus, the empirical evidence decision-making processes and outcomes.18
17
Factors contributing to users’ trust in RAs are discussed in the next
subsection. The potential interrelationship between trust and decision
quality is discussed in the subsection “Suggestions for Future Research.” 18
The condition for such effects is that the RAs are indeed used.

RA Type. The most common typology of RAs is based on Other researchers (Felix et al. 2001; Komiak and Benbasat
filtering methods: (1) content-filtering RAs, and (2) collab- 2006; Stolze and Nart 2004) have used the terms feature-
orative-filtering RAs (Adomavicius and Tuzhilin 2003; Ansari based RAs and needs-based RAs to refer to decision-
et al. 2000; Cosley et al. 2003; Massa and Bhattacharjee 2004; assisted RAs and expert-driven RAs, respectively. Stolze
Schein et al. 2002; Wang and Benbasat 2004a; West et al. and Nart (2004) have also advocated the use of hybrid RAs
1999; Zhang 2002). Content-filtering RAs generate which allow consumers to specify both desired product
recommendations based on consumers’ desired product attri- features and product-related needs.
butes. Examples of current commercial implementations of
content-filtering RAs include Active Buyers Guide and My The type of RAs used by consumers is expected to have an
Simon. Collaborative-filtering RAs, on the other hand, mimic effect on their decision-making outcomes and decision-making
“word-of-mouth” recommendations (Ansari et al. 2000) and processes. First, hybrid RAs combine both content-filtering
use the opinions of like-minded people to generate recom- and collaborative-filtering techniques, thus inte-grating
mendations (Ansari et al. 2000; Maes et al. 1999). Notable individual and community preferences. As such, they may lead
commercial implementations of collaborative-filtering RAs are to better decision quality than either pure content-filtering or
offered by Amazon, CD Now, and MovieLens. To take pure collaborative-filtering RAs. However, since hybrid RAs
advantage of both individuals’ desired item attributes and require users to both indicate their preferred product attribute
community preferences, many researchers (Balabanovic and level (as well as importance weights) and provide product
Soham 1997; Claypool et al. 1999; Good et al. 1999) have ratings, they may require higher user effort than do the other
advocated the construction of hybrid RAs that combine two types of RAs.
content-filtering and collaborative-filtering; Tango, an online
RA for newspaper articles, is an example.19 Additionally, research conducted by decision scientists has
shown that compensatory strategies are generally associated
RAs can also be classified in terms of decision strategies. with more thorough information search and accurate choices
Compensatory RAs allow tradeoffs between attributes, that is, than non-compensatory ones. Therefore, it is expected that the
desirable attributes can compensate for less desirable attri- use of compensatory RAs will improve decision making more
butes. All attributes simultaneously contribute to the compu- than non-compensatory RAs. However, since compen-satory
tation of a preference value. Weighted additive decision stra- RAs typically require users to provide more preference
tegy is the most widely used form of the compensatory model. information (e.g., attribute weights), they may increase users’
Non-compensatory RAs do not consider tradeoffs between decision effort (as indicated by the amount of user input).
attributes. Since aggregate utility scores are not typically
calculated, some attributes may not be considered at all. Also, Finally, an assumption made about RA use, which is not
some attributes may be considered before others. Non- always justified, is that the customers recognize their own
compensatory decision strategies do not form preference needs or at least have the ability to understand and answer the
scores for products as much as they narrow the set of products preference elicitation questions (Patrick 2002; Randall et al.
under consideration. Most of the currently available RAs, 2003). It is likely that customers may not possess the required
including the ones at My Simon, are non-compensatory RAs. knowledge about the product or its use to specify their pre-
ferences correctly. It is also not uncommon for customers to
Grenci and Todd (2002) differentiated between two types of
answer a different question than the one asked simply because
web-based RAs in terms of the amount of support provided by
they may not understand the question actually asked and thus
the RAs for consumer purchase. Decision-assisted RAs ask
answer the question that they think is being asked. Since
customers to specify what features they want in a product and
needs-based RAs provide support to consumers by asking for
then help customers narrow down the available choices and
their product related needs rather than their specifications of
recommend alternatives. Expert-driven RAs, on the other
product attributes (Felix et al. 2001; Grenci and Todd 2002;
hand, ask customers to provide information about themselves
Komiak and Benbasat 2006; Randall et al. 2003; Stolze and
and how they will use the product, and then recommend Nart 2004), they help consumers better recognize their needs
alternatives, from which the customers can make a choice. and answer the preference-elicitation questions, which should
result in better decision quality.
19 In sum, the type of RAs provided to the consumers to assist

Since the focus of the current paper is on design and adoption issues of
RAs, interested readers are referred to Adomavicius and Tuzhilin (2003), in their shopping tasks will affect consumers’ decision
Herlocker et al. (2004), and Zhang (2002) for more details on different
algorithms developed for each type of systems. quality and decision effort. It is therefore proposed that

P3: RA type influences users’ decision effort and decision Kramer (2007), lacking stable preferences, consumers may
quality. need to infer their preferences from their responses to pre-
ference elicitation tasks, use those self-interpreted preferences
P3a: Compared with pure content-filtering RAs or pure to evaluate the alternatives recommended by the RAs, and
collaborative-filtering RAs, hybrid RAs lead to make a choice decision based on the evaluations. Since an
better decision quality and higher decision effort implicit preference elicitation method makes it hard for
(as indicated by amount of user input). consumers to have insight into their constructed preferences,
they may make suboptimal choice decisions based on their
P3b: Compared with non-compensatory RAs, incorrectly inferred preferences. As such, we expect that the
compen-satory RAs lead to better decision means by which preferences are elicited will influence con-
quality and higher decision effort (as indicated sumers’ decision quality. Moreover, since an explicit pre-
by amount of user input). ference elicitation method requires users to indicate their
preferences with product ratings or answers to preference
P3c: Compared with feature-based RAs, needs-based elicitation questions, it demands more decision effort (e.g., in
RAs lead to better decision quality. terms of increased user input and decision time) from con-
sumers than does an implicit preference elicitation method. It
Schafer et al. (2002) observed that hybrid RAs generate more is therefore proposed that
confidence from the users compared to traditional collab-
orative-filtering ones. Fasolo et al. (2005) found that indi- P4: The preference elicitation method influences users’
viduals using compensatory RAs had more confidence in their decision quality and decision effort. The explicit pre-
product choices than did those using non-compensatory RAs. ference elicitation method leads to better decision
quality and higher decision effort (as indicated by
RA Input Characteristics. A central function of RAs is the amount of user input) than does the implicit
capturing of consumer preferences, which allows for the iden- preference elicitation method.
tification of products appropriate for a consumer’s interests.
The way a consumer’s preferences are gathered during the Aggarwal and Vaidyanathan (2003a) found that the pre-
input stage can significantly influence her online decision ferences inferred from profile building and the preferences
making. The input characteristics discussed in this section are explicitly stated by customers may not be similar. Kramer
preference elicitation method and included product attributes. (2007) observed that respondents were significantly more
likely to accept top-ranked recommendations (resulting in
Consistent with the notion of bounded rationality (Simon better decision quality) when their preferences had been
1955), as a result of their limited information processing elicited using a more transparent task (i.e., a self-explicated
capacity, individuals often lack the cognitive resources to approach in which users explicitly rate the desirability of
generate well-defined preferences. Instead of having pre- various levels of attributes as well as the relative
ferences that are revealed when making a decision, indi- importance of the attributes) as opposed to a less
viduals tend to construct their preferences on the spot, for transparent task (i.e., a full-profile conjoint analysis). The
example, when they must make a choice (Bettman et al. 1998; transparent (i.e., explicit) preference elicitation method
Payne et al. 1992). Since formation of consumer preferences is enabled users to infer their preferences from their responses
influenced by the context in which product choices are made to the measurement task and to use these preferences in
(Bellman et al. 2006; Lynch and Ariely 2000; Mandel and evaluating the RAs’ recom-mendations.
Johnson 1998; West et al. 1999), preference reversals often
When consumers depend on RAs for screening and evaluating
occur. Prior research (Nowlis and Simonson 1997; Tversky et
product alternatives, the RAs influence how consumers
al. 1988) has shown consistent preference reversals when
construct their preferences. The way a consumer’s pre-
preferences were elicited with different tasks (e.g., choice task
ferences are obtained during the input stage can significantly
versus rating task, choice task versus matching task). In the
influence their online shopping performance. Haubl and
context of RA use, consumer pre-ferences can either be
gathered implicitly (by building profiles from customer’s Murray (2003) note that real-world RAs20 are inevitably
purchase history or navigational pattern) or elicited explicitly selective “in the sense that they include only a subset of the
(by asking customers to provide product ratings/rankings or pertinent product attributes.” As such, “everything else being
answer preference elicitation questions). The means by which
preferences are elicited may affect what consumers do with the 20
Their study focuses on content-based RAs that elicit consumer
RAs’ recommendations. According to preferences explicitly.

equal, the inclusion of an attribute in an electronic recommen- filtering RAs are used) for the recommended alternatives.
dation agent renders this attribute more prominent in con- In line with the theory of constructed preferences, due to
sumers’ purchase decisions” (p. 75). Thus, when consumers human cognitive constraints, consumer preferences are
perform comparisons across products in the consideration set frequently influenced by the context in which particular
during the in -depth comparison stage (as illustrated in Table product evalua-tions and product choices are made. As
6), they are likely to consider the product attributes included in such, the utility scores or predicted ratings accompanying
the RAs’ preference-elicitation interface to be more important recommendations are likely to influence consumers’
than those not included. Consequently, pro-ducts that are product evaluations during both the initial screening stage
superior on the included attributes will be evaluated more and the in-depth comparison stage of the online shopping
favorably and will be more likely to be chosen by consumers decision-making process (see Table 6) as well as their final
than products that are superior on the excluded attributes. It is product choice. It is therefore proposed that
therefore proposed that
P6: Recommendation content influences users’ product
P5: Included product attributes21 influences users’ pre- evaluation and choice.
ference function and choice. Included product
attributes (in RA’s preference elicitation interface) are P6a: Recommendations provided by RAs influence
given more weight in the users’ preference function users’ choice to the extent that products recom-
and considered more important by the users than mended RAs are more likely to be chosen by
those not included. Product alternatives that are users.
superior on the included product attributes are more
likely to be chosen by users than are products superior P6b: The display of utility scores or predicted ratings
on the excluded product attributes. for recommended products influences users’
product evaluation and choice to the extent that
Haubl and Murray (2003) observed that the number of products with high utility scores or predicted
subjects who chose a product with the most attractive level ratings are evaluated more favorably and are
of the product attribute included in the RAs’ preference- more likely to be chosen by users.
elicitation interface was significantly larger than the
number of subjects who chose a product with the most Senecal (2003; see also Senecal and Nantel 2004), in investi-
attractive level of the product attribute excluded in the gating the influence of online relevant others (including other
RAs’ preference-elicitation interface, thereby confirming customers, human experts, and RAs), found that the presence
the hypothesized “inclusion effect.” Furthermore, they of online product recommendations significantly influenced
found that the inclusion effect persisted into subsequent subjects’ product choices. All subjects participating in the
choice tasks, even when no RA was present. study were asked to select one out of four product alterna-
tives. They were free to consult RAs (which would then
RA Output Characteristics. RA output characteristics dis- recommend an alternative out of the four) or not. Senecal
cussed in the section include recommendation content (e.g., observed that subjects who utilized the RAs were much more
specific recommendations generated by the RAs and the likely to pick the recommended alternative than those who did
utility scores or predicted ratings for recommended not utilize the RA. Similar effects have also been observed by
products) and recommendation format (e.g., the display Wang (2005). Focusing on an online collaborative-filtering
method for pre-senting recommendations and the number movie RA, Cosley et al. (2003) conjectured that showing the
of recommen-dations). predicted rating22 (the RA’s prediction of a user’s rating of a
movie, based on the user’s profile) for a recommended movie
Given that the primary function of RAs is to assist and advise
at the time the user rates it23 (at the output stage) might affect
consumers in selecting appropriate products that best fit their
a user’s opinion. They conducted an experiment where users
needs, it is expected that the recommendations presented by
were asked to re-rate a set of movies which they had pre-
the RAs will influence consumers’ product choices. In addi-
viously rated, after they were given an RA’s rating, which
tion to providing product recommendations to consumers,
some RAs also provide utility scores (when content-filtering
RAs are used) or predicted ratings (when collaborative- 22
RAs generally provide information about the items recommended. This
may include item descriptions, expert reviews, average user ratings, or
predicted personalized ratings for a given user.
21 23
This proposition applies only to one type of content-filtering RAs, in RAs (collaborative-filtering RAs in particular) often provide a way for
which individuals’ preferences for product attributes are explicitly elicited. users to rate an item when it is recommended.

was either higher, lower, or the same as the user’s initial tions in the recommendation list may degrade consumers’
rating of the movie. A comparison of the new ratings and choice quality by lowering the average quality of considered
the original ratings showed that users tended to adjust their alternatives and diverting the consumers’ attention from the
opinions to coincide more closely with the RA’s prediction, better options to the more mediocre ones (Diehl 2003).
whether the prediction was accurate or not, demonstrating
that users can be manipulated by such information. It is therefore proposed that
Recommendation format can also exert significant influence P7: Recommendation format influences users’ decision
on RA users’ decision processes and decision outcomes. First, processes and decision outcomes.
recommendations are displayed to the users either in the order
of how they satisfy users’ preferences or in a random order. P7a: Recommendation display method influences
The display method for presenting recommendations has an users’ decision strategies and decision quality to
effect on strategies used to evaluate products during the initial the extent that sorted recommendation lists result
screening stage of a user’s decision-making process. With a in greater user reliance on heuristic decision stra-
sorted recommendation list from RAs, there will be many tegies (when evaluating product alternatives) and
products that are close in their overall subjective utility (Diehl better decision quality.
et al. 2003) but which differ on several attributes (Shugan
1980). According to Shugan (1980), if the differ-ence in P7b: The number of recommendations influences
overall subjective utility between two products is small, users’ decision effort and decision quality to the
consumers must make many attribute comparisons between extent that presenting too many
the two products to determine the source(s) of this small recommendations in-creases users’ decision
difference, hence the choice is more difficult than when the effort (in terms of decision time and extent of
utility difference is large between the two products. As choice product search) and decreases decision quality.
difficulty increases, consumers tend to depart from the
normative, utility-based comparisons (i.e., evaluating the cur- Dellaert and Haubl (2005) found that a sorted list of person-
rently inspected product against the most attractive product, in alized product recommendations increased consumers’
terms of subjective utility, that they have inspected up to that tendency to engage in heuristic local utility comparison when
point) and instead base their choice on simplifying heuristics, evaluating alternatives. Diehl et al. (2003) have shown that,
such as attribute-based processing (rather than alternative- with a sorted list of recommendations (in the order of how
based processing) and local optimization (i.e., comparing the each is predicted to match users’ preferences), consumers had
utility between contiguously inspected products) (Haubl and better decision quality and paid lower prices. Aksoy and
Dellaert 2004; Payne et al. 1988). Consequently, consumers Bloom (2001) also observed that, compared to a randomly
may rely more on heuristic decision strategies when ordered list of options, a list of recommendations sorted based
distinguishing among the product alternatives in the sorted list on consumers’ preferences resulted in higher consumer
of recommendations and deciding on the ones to be included decision quality. As to the number of recommendations to be
in the consideration set. Recommendation display method can presented to RA users, Diehl (2003) found that recommending
also result in higher decision quality. Sorted recommendation more alternatives significantly increased the number of unique
lists contain many good choices of comparable quality (Diehl options searched, decreased the quality of the consideration
et al. 2003), with the most promising options at the beginning set, led to poor product choices, and reduced consumers’
of the list. Simply by choosing product alternatives close to the selectivity. Basartan (2001) constructed an RA in which
begin-ning of the sorted list, consumers can achieve better- response time and the number of alternatives displayed were
than-average decision quality. varied. She found that, when the RA provided too many
recommendations, it increased the users’ effort of evaluating
these alternative recommendations.
Second, the number of recommendations presented to the
consumers may affect their decision effort and decision Summary. The propositions presented in this section help
quality. Providing too many recommendations may prompt answer research question (1.2): How do the characteristics of
consumers to compare a larger number of alternatives, thus RAs influence consumer decision-making processes and
increasing their decision effort by increasing decision time and outcomes? Various types of RAs exert differential influence on
the size of the alternative sets. Moreover, in a sorted consumers’ decision quality and decision effort. RAs’
recommendation list, the most promising options are located at preference elicitation method (i.e., explicit or implicit) also
the beginning of the list. As such, considering more op- affects consumers’ decision quality and decision effort. In-

Table 8. The Impact of RA Characteristics on Decision Process and Decision Outcomes

The propositions in this table provide answers to research question 1.2:
Q1.2: How do the characteristics of RAs influence consumer decision making processes and outcomes?
RA Type
RA Type P3
• Content-filtering, • Hybrid RAs generated more confidence from the users than • P3a
collaborative filtering, did traditional collaborative-filtering ones (Schafer et al. 2002).
vs. hybrid
Decision Quality
• Compensatory vs. non- and Decision • Individuals using compensatory RAs had more confidence in • P3b
compensatory Effort their product choices than did those using non-compensatory
RAs (Fasolo et al. 2005).
• Feature-based vs. • No empirical study available. • P3c

needs-based
RA Input
Inferred and explicitly stated consumer preferences may not
Preference Elicitation Decision Quality converge (Aggarwal and Vaidyanathan 2003a); respondents
Method (implicit vs. and Decision were more likely to accept top-ranked recommendations when P4
explicit) Effort their preferences had been elicited using a more transparent task
as opposed to a less transparent one (Kramer 2007).
Preference Attributes included in an RA’s preference-elicitation interface and
Included product
Function and sorting algorithm were given more weight during the users’ P5
attributes
Choice product choice (Haubl and Murray 2003).
RA Output
Recommendation content P6
• Recommendations • Subjects who used the RA were much more likely to select the • P6a
recommended alternative than those who did not (Senecal
Product 2003; Wang 2005).
Evaluation and
• Utility Scores or Choice • The display of predicted rating for a recommended movie at • P6b
Predicted Ratings the time the user rates it affected a user’s opinions: users
adjusted their opinions to coincide more closely with the RA’s
prediction (Cosley et al. 2003).
Recommendation format P7
• Recommendation • Sorted recommendation list increased consumers’ tendency to • P7a
Display Method (sorted engage in local utility comparison when evaluating alternatives
vs. non-sorted) (Dellaert and Haubl 2005) and resulted in higher decision
Decision
quality (Aksoy and Bloom 2001; Diehl et al. 2003).
Strategies,
Decision Effort,
• Number of • Recommending more alternatives increased information
and Decision
Recommendations searched, decreased the quality of the consideration set, led • P7b
Quality
to poor product choices, and reduced consumers’ selectivity
(Diehl 2003); a shopbot that provided too many recommen-
dations increased the users’ cognitive effort and decreased
their preference for the shopbot (Basartan 2001).

cluded product attributes in RAs’ preference elicitation making process (whereby consumers reply on themselves to
interface, recommendations provided by the RAs, and the search for information, evaluate alternatives, and make pur-
utility scores (or predicted ratings) attached to RAs’ recom- chase decisions) and other-based decision-making process
mendations, influence consumers’ preference functions and (whereby consumers subcontract either part or all of their
product choices. Whereas a sorted recommendation list decision-making process). Whereas own-based decision-
results in greater consumer reliance on heuristic decision making processes occur when consumers have the capacity to
strategies when evaluating alternatives and better decision process information and perform a complex decision-making
quality, an excess of recommendations leads to increased process, other-based decision-making processes occur when
decision effort and decreased decision quality. Table 8 sum- consumers do not have the capacity to process information.
marizes the propositions concerning the impact of RA type, King and Balasubramanian (1994) found that product type had
as well as of RA characteristics associated with input and a significant impact on consumers’ reliance on a particu-lar
output design, on consumers’ decision processes and out- decision-making process: consumers evaluating a search
comes. Empirical support (when available) for these propo- product (e.g., a camera) were more likely to use own-based
sitions is also included. decision-making processes; in contrast, those evaluating an
experience product (e.g., a movie) tended to rely more on
other-based decision-making processes. Prior research has
Factors Related to Product, shown that consumers who employed other-based decision-
User, and User–RA Interaction making processes were likely to make purchasing decision in
keeping with salespersons’ recommendations (Formisano et al.
The impacts of RA use and RA characteristics on consumers’ 1982). Following this chain of reasoning, we conclude that
decision-making processes and outcomes are contingent upon consumers evaluating experience products are more likely to
factors related to product, user, and user–RA interaction as rely on salespersons’ assistance and adopt their recommen-
well as the interactions between these factors and RA charac- dations. Since the role of RAs in online shopping is similar to
teristics. The following sections discuss how the different that of the salespersons in a traditional shopping environ-ment,
factors related to product (i.e., product type and product it is expected that, compared to consumers who use RAs to
complexity), user (i.e., product expertise and perceived pro- shop for search products, those who shop for experi-ence
duct risks), and user–RA interaction (i.e., user–RA similarity) products aided by RAs are more likely to choose the products
moderate the effects of RA use on consumers’ decision making recommended by the RAs. It is therefore proposed that
processes and outcomes, as captured in P8 through P13 and
illustrated in Figure 2.
P8: Product type moderates the effects of RA use on users’
Product-Related Factors. Product-related factors such as choice. RA use influences the choice of users
product type and product complexity moderate the effects shopping for experience products to a greater extent
of RA use on consumers’ decision-making processes and than that of those shopping for search products.
out-comes.
Senecal (2003; see also Senecal and Nantel 2004) found that
In the context of online shopping, products can be categorized product type affected consumers’ propensities to follow RAs’
into search products and experience products. Search pro- product recommendations. Recommendations for experience
ducts (e.g., cameras, calculators, or books) are characterized products (wines) appeared significantly more influential than
by attributes (e.g., color, size, price, and components) that can recommendations for search products (calculators).
be assessed based on the values attached to their attributes,
without necessitating the need to experience them directly. Products can differ along another dimension, product com-
Experience products (e.g., clothing or cosmetics), on the other plexity, which is the extent to which the product is perceived
hand, are characterized by attributes (e.g., taste, softness and by the consumer as difficult to understand or use (Rogers
fit) that need to be experienced prior to purchase. 1995). Prior literature in marketing and IS has revealed
different schemes for characterizing products in terms of com-
Because of the inherent difficulty associated with the evalua- plexity. In marketing research, the complexity of a product is
tion of experience products prior to purchasing online, con- usually defined in terms of the number of attributes used to
sumers tend to feel uncertain as to whether the products would describe the product (Keller and Staelin 1987; Swaminathan
meet their expectations, which may lead to increased 2003), the number of alternatives for the product category
information search activities (Spiekermann 2001). Moreover, (Keller and Staelin 1987; Payne et al. 1993; Swaminathan
Olshavsky (1985) differentiated between own-based decision- 2003), or the number of steps involved in the use of the pro-

duct (Burnham et al. 2003). E-commerce literature defines there are more dimensions on which the alternatives are
product complexity along three dimensions: number of pro- different from one another, making it more difficult for users
duct attributes, variability of each product attribute, and to identify the best option and thus leading to greater search
interdependence of product attributes (Jahng et al. 2000). The even in the presence of the RA. On the other hand, when the
higher the number of attributes of a product, the higher the product is of low complexity (i.e., when the number of
level of variation of each attribute and the greater the degree of attributes is fewer), users have little difficulty finding the best
interdependence among product attributes, the greater the option with the assistance of the RA. However, this
product complexity. This paper adopts the definition of pro- unexpected finding may also be a result of Swaminathan’s
duct complexity by Jahng et al. (2000), but it adds an addi- manipulation of product complexity by varying the number of
tional criterion, namely, how the product attributes correlate attributes used to describe the product. Users may have prior
with one another. When product attributes are positively cor- perceptions of the complexity of a given product and therefore
related, an alternative that is favorable on one attribute tends to
may not be sensitive to such artificial manipulations.
also be favorable on other attributes. In contrast, when product
attributes are negatively correlated, a more attractive level of As noted by Haubl and Murray (2003), inter-attribute corre-
one attribute is associated with a less attractive level of lation moderates the impact of product attributes included (in
another. Products characterized by negative inter-attribute RAs’ preference elicitation interface) on consumers’ product
correlations are considered more complex (Fasolo et al. 2005),
choice. They argue that, for products characterized by nega-
since purchase decisions of such products require consumers
tive inter-attribute correlations, the relative importance at-
to make trade-offs among attributes.
tached to different attributes tends to be highly consequential
Previous research into information search has examined the with respect to the decision outcome: even very small differ-
impact of product complexity on consumer decision quality ences in relative attribute importance may affect which
and search behavior (Bettman et al. 1998; Keller and Staelin product is chosen from a set of alternatives. In contrast, for
1987; Payne et al. 1993). Keller and Staelin (1987) developed products characterized by positive inter-attribute correlation,
an analytical model, which showed that higher number of the relative importance attached to different attributes has
product attributes and alternatives resulted in decreased deci- much less of an impact on determining which product is
sion effectiveness. Payne et al. (1993) and Bettman et al. chosen. Therefore, although product attributes included in the
(1998) suggest that product complexity increases consumers’ RAs’ preference elicitation interface will be considered more
cognitive load, resulting in decision biases. As product com- important by consumers, such a change in the relative impor-
plexity increases, consumers often resort to heuristics to man- tance of product attributes will have a stronger impact on
age information overload, hence decision quality decreases. consumers’ product choice for products with negative inter-
Therefore, the benefits of using RAs, in terms of decision attribute correlations than for those with positive ones. It is
quality and search efforts, are likely to be greater when a pro- therefore proposed that
duct is complex. It is therefore proposed that
P10: Product complexity moderates the effect of included
P9: Product complexity moderates the effects of RA use on product attributes on users’ choice. The inclusion
users’ decision quality and decision effort. The use of effect is stronger for products with negative inter-
RAs for more complex products leads to greater attribute correlations (i.e., more complex products)
increase in decision quality and greater decrease in than for those with positive inter-attribute correlations
decision effort than for less complex products. (i.e., less complex products).
Empirical results from prior research, however, are not Haubl and Murray (2003) demonstrated that the preference-
supportive of this proposition. Fasolo and her colleagues construction effect of RAs depended on the inter-attribute
(2005) found that, in the presence of negatively related correlation structure across the set of available products.
attributes (an indicator of product complexity), consumers There existed a strong inclusion effect when there were
using RAs engaged in more information search, rated the nega-tive inter-attribute correlations, but not for positive
decisions to be more difficult, and were less confident in their inter-attribute correlations.
product choices. Swaminathan (2003) found that the RAs had
a more significant impact on search efforts when product User Related Factors. The effects of RA use on
complexity was relatively low, contrary to what he had consumers’ decision-making processes and outcomes are
hypothesized. Swaminathan explains that when the product is also moderated by user-related factors, including product
complex (i.e., when the number of attributes is greater), expertise and per-ceived product risks.

RAs do not have the same impact on decision outcomes across consumer evaluation and choice. He found that RAs had a
different types of users. Hoeffler and Ariely (1999) posit that more significant impact on decision quality when perceptions
consumers construct their preferences when they are new to a of product risk were high. Spiekermann (2001), however,
product category and eventually develop more stable observed that RA users addressed products with different risk
preferences with experience in a domain. Coupey et al. (1998) structures with different information search behaviors. While
also state that consumers who are highly familiar with a users viewed fewer alternatives and spent more time on each
product category are less subject to framing effects during alternative for products with high functional and financial
preference elicitation. As such, compared with individuals risks (i.e., cameras), they viewed more alternatives but spent
with low product expertise, RA users with high product less time on each alternative for products with high social-
expertise are likely to have more stable, better-defined pre- psychological risks (i.e., winter jackets).
ferences and thus are less likely to be affected by the RA’s
preference elicitation method. It is therefore proposed that Factors Related to User–RA Interaction. To serve as
“virtual advisers” for consumers, RAs must demonstrate
P11: Product expertise moderates the effect of preference similarity to their users, that is, they must internalize users’
elicitation method on users’ decision quality. Pre-ference product-related preferences and incorporate such
elicitation method has less effect on the decision quality preferences into their product screening and sorting
of users with high product expertise than on the decision process. RAs that generate and present recommendations
quality of those with low product expertise. not concordant with the consumers’ own needs are not
likely to enhance con-sumers’ decision quality. Moreover,
In his investigation of the effect of measurement task trans-
in accordance with the theory of interpersonal similarity
parency on preference construction and evaluations of
(Byrne and Griffitt 1969; Levin et al. 2002; Lichtenthal and
person-alized recommendations, Kramer (2007) observed
Tellefsen 1999; Zucker 1986), the similarities (actual
that, although respondents were in general more likely to
and/or perceived) between RAs and their users (in attribute
accept a top-ranked recommendation (resulting in better
importance weightings, decision strategies, goals, etc.) are
decision quality) when their preferences had been measured
expected to improve the predict-ability of the RAs’
using a more transparent task, the effect was modified by
behavior and focus users’ attention on more attractive
the product expertise of the users. Differences in accepting
a recom-mendation occurred only for those who did not product alternatives (Levin et al. 2002; Zucker 1986), thus
have product expertise. resulting in improved decision quality and reduced decision
effort in online shopping tasks. It is therefore proposed that
Several dimensions of product risks have been identified
and measured in consumer research, including financial, P13: User–RA similarity moderates the effects of RA use on
func-tional, social, and psychological (Spiekermann and users’ decision quality and decision effort. RA use leads
Paraschiv 2002). One or more of these sources may drive to greater increase in decision quality and greater de-
consumers’ overall perceptions of product risks, which are crease in decision effort when the RAs are similar to the
considered to be central to their product evaluations and users than when the RAs are not similar to the users.
product choice (Campbell and Goodstein 2001). Inasmuch
as information searches are used as part of risk-reduction Aksoy and Bloom (2001) examined the effect of actual
strategies (Dowling and Staelin 1994), RAs that are similarity of attribute weights and perceived similarity of
designed to facilitate con-sumers’ online product searches decision strategies between users and RAs on decision
by assisting them in screening and evaluating product quality and decision effort (as indicated by decision time).
alternatives are likely to play a more significant role in In their study, perceived similarity was measured by the
improving decision quality and reducing search efforts degree of similarity (1) between the proportionate weight
when product risks are high. It is therefore proposed that attached to an attribute by an RA and the proportionate
weight determined by consumers, and (2) between the
P12: Perceived product risks moderate the effects of RA use decision-making strategy used by an RA and the
on users’ decision quality and decision effort. When consumers’ own decision-making strategies. The
perceived product risks are high, RA use leads to greater researchers observed that consumers who were presented
improvements in decision quality and reduction in deci- with recommendations based on attribute weights similar to
sion effort than when perceived product risks are low. their own tended to make better decisions (e.g., they were
less likely to choose dominated alternatives) and spend less
Swaminathan (2003) conducted an investigation of the moder- time examining alternatives. However, no significant effect
ating role of perceived product risk on the impact of RAs on was found for similarities in decision-making strategies.

Table 9. The Impact of Factors Related to Product, User, and User–RA Interaction on Decision Process
and Decision Outcomes
Q1.3: How do other factors (i.e., factors related to user, product, and user–RA interaction) moderate the effects of RA use
and RA characteristics on consumer decision making processes and outcomes?
Relationship
Moderator Empirical Support P
Moderated
Product-Related Factors
Product Type
RA Use on Recommendations for experience products were more influential than those P9
(search vs.
Choice for search products (Senecal, 2003; Senecal and Nantel 2004).
experience)
RA Use on In the presence of negatively related attributes, consumers using decision
Product Decision Quality aids engaged in more information search and were less confident in product
P9
Complexity and Decision choices (Fasolo et al. 2005); RAs had a more significant impact on search
Effort efforts when product complexity was relatively low (Swaminathan 2003).
Included Product There existed a strong inclusion effect in a scenario characterized by negative
Product
Attributes on inter-attribute correlation, but not in the case of positive inter-attribute P10
Complexity
Choice correlation (Haubl and Murray 2003).
User-Related Factors
Although users were in general more likely to accept a top-ranked
Preference
recommendation when their preferences had been measured using a more
Product Elicitation
transparent task, the effect was modified by the product expertise of the P11
Expertise Method on
users: differences in accepting recommendation occurred only for those who
Decision Quality
did not have product expertise (Kramer 2007).
RAs caused a more significant impact on decision quality under conditions
RA Use on Deci-
Perceived involving perceptions of relatively high product risk (Swaminathan 2003); RA
sion Quality and P12
Product Risks users addressed products with different risk structure with different
Decision Effort
information search behavior (Spiekermann 2001).
User–RA Interaction
RA Use on Deci- User–RA similarity in attribute weights resulted in better decision quality and
User–RA less information search; No effect was found for user–RA similarity in decision P13
sion Quality and
Similarity making strategies (Aksoy and Bloom 2001).
Decision Effort
Summary. The propositions presented here provide answers to Users’ Evaluations of RAs
research question (1.3): How do other factors moderate the
effects of RA use and RA characteristics on consumer deci- The criteria users apply to evaluate RAs are based on their
sion-making processes and outcomes? The effects of RA use
general perceptions of the RAs, which are affected by RA use,
and RA characteristics on decision quality, product choice, and
RA characteristics (RA type as well as RA input, process, and
decision effort are stronger for complex products. RAs’ effects
output characteristics), RA provider credibility, and fac-tors
on influencing consumers’ product choice are stronger for
related to product, user, and user–RA interactions. All five
experience products. The higher the consumers’ product
expertise, the less likely their decision quality is influenced by theoretical perspectives are drawn upon to guide the
RAs. RAs are expected to exert the greatest effects on deci- development of hypotheses P14 through P28, which provide
sion quality and decision effort when the consumers’ answers to our second research question: How do RA use, RA
perceptions of product risks are high or when RAs are per- characteristics, and other factors influence users’ evaluations
ceived to be similar to the user. Table 9 summarizes the of RAs? The relationships investigated in this section are
theoretical propositions as well as their empirical support. depicted in Figure 3.

RA Use Therefore, users may consider that hybrid RAs require

more effort.
According to Fishbein and Ajzen (1975), an individual’s
descriptive beliefs about an object can be formed through Second, prior research has shown that, although consumers
direct experiences with such object. Prior IS research (e.g., do not enjoy exerting their own decision making effort,
Bajaj and Nidumolu 1998; Kim and Malhotra 2005a; they react positively to the effort exerted by others (Mohr
Limayem et al. 2003; Venkatesh et al. 2003) has also shown and Bitner 1995). For instance, Kahn and Baron (1995)
that IS use can serve as a basis for the formation (or update) of investi-gated the preferred choice rules by individuals and
user evaluations (i.e., usefulness, ease of use, trustworthiness, found that, although most participants chose the simple
and satisfaction) of the IS at a subsequent stage. In the context non- compensa-tory rules, they wanted their doctors or
of RA-assisted online shopping, the use of RAs is expected to financial officers to use compensatory rules (which require
influence consumers’ evaluations of the RAs. However, such more effort) when making decisions on their behalf. In an
an effect will be determined, at least in part, by RA RA-assisted shopping context, cognitive effort has shifted,
characteristics and factors related to product, user, and user– at least partially, from consumers to RAs. Since consumers
RA interaction. A general proposition such as “RA use will in such a context rely mostly on RAs’ efforts rather than on
influence users’ perceived usefulness of, perceived ease of use their own, we expect that they will evaluate compensatory
of, trust in, and satisfaction with an RA” does not provide RAs more favorably (i.e., more trustworthy, useful, and
much insight into our understanding of the effects of RA use satisfactory) than non-compensatory RAs. However, since
compensatory RAs require more user input (e.g., attribute
on users’ evaluations. In order to answer research question
weights) than non-compensatory ones, users may perceive
(2.1)—How does RA use influence users’ evaluations of RAs?
the former as more difficult to use.
—it is important to take into account RA characteristics and
other important moderating factors. In sum, RA type affects users’ trust in, perceived usefulness
of, perceived ease of use of, and satisfaction with RAs. It is
RA Characteristics therefore proposed that
Just as the effects of RA use on consumers’ decision-making

P14: RA type influences users’ evaluations of RAs.
processes and outcomes are influenced by the RA charac- P14a: Compared with pure content-filtering or pure
collaborative-filtering RAs, hybrid RAs lead to
teristics, their effects on users’ evaluations of RAs are also
greater trust, perceived usefulness, and satis-
determined partially by the type of RAs, as well as by the
faction but to lower perceived ease of use.
characteristics of the RAs associated with the input, process,
P14b: Compared with non -compensatory RAs,
and output design. In the following subsections, guided by the com-pensatory RAs lead to greater trust,
theories of trust formation, theories of interpersonal simi- perceived usefulness, and satisfaction but to
larity, and TAM, we develop a set of propositions, P14 through lower per-ceived ease of use.
P21 (illustrated in Figure 3) concerning the effects of RA
characteristics on users’ evaluations of RAs.24 Schafer et al. (2002) compared meta-recommender systems
that combine collaborative-filtering and content-filtering
The Impact of RA Type on Users’ Evaluations of RA. In tech-niques to traditional recommender systems (using only
addition to its effects on consumers’ decision making pro- collab-orative-filtering technique) and found that users
cesses and outcomes, the type of RA used is also expected to considered the former more helpful than the latter.
affect users’ evaluations of RAs. First, since hybrid RAs base
their recommendations both on individual users’ specifica- Users’ evaluations of certain RA types are also contingent on
tions for product attributes and on inputs from similar others user related factors. For instance, users’ product expertise will
(i.e., other consumers who are similar in tastes and prefer- influence their evaluation of collaborative-filtering versus
ences to the users), users may develop greater trust in such content-filtering and needs-based versus feature-based RAs.
RAs and consider them more useful than pure content-fil- The theoretical justifications, propositions and empirical
tering or collaborative-filtering RAs. However, hybrid RAs support (when available) are presented later in this section.
generally require more user input in the form of answers to
RA Input Characteristics. The means by which users’
preference elicitation questions and ratings on alternatives.
preferences are elicited, the ease for users to generate new
or additional recommendations, and the amount of control
users have when interacting with the RAs’ preference
24 elicitation interface influence users’ evaluations of the RAs.
Again, the condition for such effects is that the RAs are indeed used.

User-RA User Product

Provider
Interaction
• Product • Product Credibility
• User-RA expertise type
similarity
• User’s familiarity
with RA
• Confirmation/
disconfirmation
of expectations
P25, P26 P23 P22 P28

Outcomes of
P25 P23 P22
RA Use
RA Use
P23 User Evaluation of
RA
P27 P23 Trust
RA Characteristics
P24
RA Type
P14, P17, P20
• Content-filtering/
Perceived
collaborative-
filtering vs. hybrid Usefulness
• Compensatory vs.
P24
non-compensatory
• Feature-based P14, P17,
vs. needs-based Perceived
P20b, 20c,
P21 Ease of Use
Input P24
• Preference
elicitation method P14, P15,
ƒEase of generating P16, P21
Satisfaction
new/additional
recommendation P24
ƒ Use control
P14, P15, P16,
Process P17, P18, P19,
• Information about P20b, P20c,
search process P20d, P21
ƒ Explanation
ƒ Response time
Output
• Recommendation
content
• Recommendation
format
Figure 3. Effects of RA Use, RA Characteristics, and Other Factors on User Evaluation of RA

Given users’ reluctance to extend cognitive effort, RAs that was originally designed (Chu and Elam 1990; Silver 1988). As
employ explicit (implicit) preference elicitation method will be such, instead of trying to force changes in users’ decision-
considered more difficult (easier) to use, all other things being making processes, system designers should implement a
equal. Additionally, since individuals typically do not have minimally restrictive DSS and provide support for a variety of
stable preferences, they may need to modify their previously decision models that the decision maker might choose to
specified preferences while interacting with the RAs to employ (Silver 1990). A comprehensible, predictable, and
generate new or additional recommendations. As such, controllable system will give users the feeling of accom-
compared to RAs that require users to repeat the entire plishment (Shneiderman 1997) and assurance (DeLone and
preference-elicitation process when they desire new recom- McLean 2003). In the same vein, in the context of RA-assisted
mendations, RAs that make it convenient for users to adjust online shopping, allowing users to control their interactions
their preferences to generate new recommendations will be with RAs to meet their personal needs (e.g., allowing the users
considered easier to use. In line with DeLone and McLean’s to specify the type of information they are most interested in
(2003) IS success model, ease of use will enhance users’ and to personalize the interface) will increase their trust and
perception of RAs’ system quality, which in turn contributes to personal satisfaction (West et al. 1999) and reduce their
their satisfaction with the RAs. It is therefore proposed that perceived functional, financial, and socio-psychological risks
(Spiekermann 2001; Spiekermann and Paraschiv 2002).
P15: The preference elicitation method influences users’ Moreover, such control increases the degree of active
perceived ease of use of and satisfaction with the RAs. involvement of users in the decision task and thus creates an
Compared to an explicit preference elicitation method, illusion of control (see Davis and Kottemann 1994; Kottemann
an implicit preference elicitation method leads to greater and Davis 1994; Langer 1975), which, as dis-cussed
perceived ease of use of and satisfaction with the RAs. previously, can cause users to overestimate the advantage of
RAs that allow user control. This in turn results in the
P16: The ease of generating new or additional recom- perception of such RAs as more useful than those that put the
mendations influences users’ perceived ease of use of users in a more passive role. An example of user control of
and satisfaction with RAs. The easier it is for the interaction with RAs is the flexibility afforded to users to
users to generate new or additional recommendations, choose the length and depth of the interactions with RAs. It is
the greater their perceived ease of use of and therefore proposed that
satisfaction with the RAs.
P17: User control of interaction with RAs’ preference-
No study has investigated the effect of the preference elicita- elicita-tion interface influences users’ trust in,
tion method on users’ perceptions of how easy it is to use the satisfaction with, and perceived usefulness of the RAs.
RAs. Swearingen and Sinha (2001, 2002) observed that Increased user control leads to increased trust,
simplifying a process whereby a user can generate new or satisfaction, and perception of usefulness.
additional recommendations (i.e., not requiring users to repeat
an entire set of ratings or a question-answer process to obtain Pereira (2000) observed that increased user control over
new recommendations), improved estimations of the ease of interaction with RAs resulted in more positive affective reac-
use of RAs. Pereira (2000) found that giving the users the tions to RAs. The three ways of increasing the degree of user
capability to return to the preference specification stage at any control (i.e., giving the users the ability to return to the
time and restate their preferences significantly increased their preference specification stage at any time and restate their
positive responses to RAs. Bharati and Chaudhury (2004) also preferences, to skip responses to certain attribute specifica-
demonstrated the net positive effect of RAs’ system quality tions requested, and to express their degree of confidence in
(i.e., ease of use, convenience of access, and relia-bility) on the preference specifications for each attribute) significantly
users’ decision making satisfaction. increased users’ trust in and satisfaction with two content-
filtering RAs. McNee et al. (2003) contrasted system-con-
The attainment of the goal of a decision aid to improve users’ trolled RAs (i.e., the RAs decide which items the users can
decision-making processes may be undermined by system rate) with user-controlled ones (i.e., the users are allowed to
restrictiveness, defined as “the degree to which and the man- specify some or all of the items to be rated) and found that the
ner in which a DSS limits its users’ decision-making pro- user-controlled RAs generated higher user satisfaction than
cesses to a subset of all possible processes” (Silver 1990, p. system-controlled RAs. Users of user-controlled RAs also
52). Not only can users perceive physical limitations of the thought the RAs best understood their tastes and were most
system, they can also be induced to employ a much narrower loyal to the RAs. Komiak et al. (2005) found that control
range of decision processes than that for which the system process (i.e., a process whereby users have more control over

their interaction with the RAs) was one of the top contributors System response time, the time between the user's input and
to users’ trust in a virtual salesperson. Wang (2005) also the computer's response, has been widely recognized as one of
observed that RAs that were perceived by users as more the strongest stressors during human–computer interaction
restrictive (i.e., restricting users’ decision strategies to a (Thum et al. 1995). Assessments of the effects of response
greater extent) were considered as less trustworthy and useful. time have been conducted for personal computer use in a
variety of contexts. Long response time increases stress levels
RA Process Characteristics. Characteristics of RAs during the (Weiss et al. 1982), self-reports of annoyance (Schleifer and
recommendation generation process, such as the infor-mation Amick 1989), frustration, and impatience (Sears and Borella
about search process and system response time, will moderate 1997) of personal computer users. According to DeLone and
the effects of RA use on users’ evaluations of RAs.
McLean (2003), response time influences users’ perception of
King and Hill (1994) have noted that it is necessary to dis- system quality and thus their satisfaction with information
tinguish between consumers’ involvement with the decision systems. Prior research has shown that lengthy system
process and their experience with the outcome of the process. response times cause lower satisfaction among users (e.g.,
West et al. (1999) also suggest that, when measuring consu- Hoxmeier and DiCesare 2000; Shneiderman 1998). In the
mer satisfaction, it is important to consider satisfaction with context of RA-assisted shopping, we also expect longer
both the decision-making process and the final product choice. response times to negatively affect users’ satisfaction with
In the context of online shopping with the assistance of RAs, RAs. It is therefore proposed that
cognitive effort is partly shifted from consumers to the RAs.
While customers may be reluctant to exert their own efforts, P19: Response time influences users’ satisfaction with RAs.
they generally welcome efforts expended by others (Kahn and The longer the RAs’ response times, the lower the
Baron 1995; Mohr and Bitner 1995). Consumers use various users’ satisfaction with the RAs.
cues or indicators to assess the amount of effort saved by
decision aids (Mohr and Bitner 1995), such as indi-cations of Basartan (2001) constructed a simulated shopbot in which
the amount of information that has been searched by a response time was varied. She found that users’ preference for
decision aid (Bechwati and Xia 2003). For example, Microsoft the shopbot decreased when they had to wait a long time
Expedia informs users that the system is searching thousands before receiving recommendations. Swearingen and Sinha
of databases for the best airfare while the cus-tomers are
(2001, 2002), however, found that the time taken by users to
waiting for results. It is expected that customers who are
register and to receive recommendations from RAs did not
informed about the RA’s search progress (while waiting for
have a significant effect on users’ perceptions of the RA. The
recommendations) will perceive that the RAs have saved them
a higher amount of effort, and thus will be more impressed seemingly contradictory research findings regarding response
with the RAs’ empathy (i.e., care and attention). As predicted time may be explained by users’ cost–benefit assessments;
by DeLone and McLean (2003), empathetic RAs are when they perceive that the benefits of waiting (e.g., ob-
considered of higher service quality and thus result in higher taining quality recommendations) outweigh its costs, they will
user satisfaction with the RAs and with the RA-assisted not form negative evaluations of the RAs.
decision-making process. It is therefore proposed that
RA Output Characteristics. The output stage is where
RAs’ recommendations are presented to users. The content
P18: The provision of information about search progress, and the format of these recommendations can have
while users await results influences users’ satisfaction significant impact on users’ evaluations of RAs.
with RAs. Users who are informed about RAs’ search
progress (while waiting for recommendations) are Specifically, three aspects (i.e., the familiarity of the recom-
more satisfied with the RAs. mendations, the amount of information on recommended
products, and the explanations on how the recommendations
Bechwati and Xia (2003) observed that consumers’ satisfac-
are generated) are relevant to how recommendation content
tion with a decision process increased with the level of effort
they saved. They conducted two empirical studies on the use influences users’ evaluations of RAs. First, according to
of a job search RA, discovering that informing online shop- knowledge-based trust theorists (Luwicki and Bunker 1995)
pers about the progress of a search augmented the shoppers’ individuals develop trust over time as they accumulate knowl-
perceptions of the effort saved for them by an RA. Conse- edge relevant to trust, through their experiences with another
quently, their satisfaction with the decision-making process party (McKnight et al. 1998). Knowledge-based trust is
concerning purchasing the product also increased. grounded substantially in the predictability of the other party,

which develops as each party proceeds to know the other well P20a: Familiar recommendations increase users’
enough to be able to anticipate his or her behavior and thereby trust in the RAs.
to avoid surprises. Familiarity builds trust (Komiak 2003; P20b: The composition of the list of recommendations,
Komiak and Benbasat 2006). Therefore, we expect that as reflected by a balanced representation of
consumers will trust RAs that provide familiar recommen- familiar and unfamiliar (or new) product recom-
dations (i.e., product recommendations with which consumers mendations, influences users’ trust in, perceived
have had positive experience previously) more than those that usefulness of, and satisfaction with RAs.
provide unfamiliar or novel recommendations (i.e., product P20c: The provision of detailed information about
recommendations consumers have not experienced before). recommended products increases users’ trust
However, to be considered useful RAs that provide relevant in, perceived usefulness of, and satisfaction
information to consumers, the RAs must present them with with RAs.
unfamiliar alternatives that fit the consumers’ needs. One P20d: The provision of explanation on how the
possible solution suggested by Cooke et al. (2002) to over- recommendations are generated increases
come users’ negative reactions to RAs that provide unfamiliar users’ trust in and satisfaction with RAs.
recommendations involves embedding unfamiliar recommen-
dations among a set of recommendations that the users are Sinha and Swearingen (2001; Swearingen and Sinha 2001)
known to like (based on their purchase history or previous found that, although novel recommendations were generally
feedback). Recommendations of familiar products can serve as considered more useful than familiar recommendations,
a context within which unfamiliar recommendations are recommended products that were familiar to users or that had
evaluated, hence improving the attractiveness of the un- previously met the users’ approval played an important role in
familiar recommendations and the evaluation of the RAs. establishing users’ trust in RAs. Cooke et al. also showed that
unfamiliar recommendations lowered users’ evaluations of a
Second, detailed information about recommendations gener- simulated music CD RA.
ated by RAs (e.g., product descriptions in text or multimedia
format, expert reviews, or other customers’ evaluations) can Available empirical evidence also supports the positive effect
signal to the users that the RAs care about them, act in their of detailed product information on users’ evaluations of RAs.
interests, and behave in an honest and unbiased fashion, Sinha and Swearingen (2001; Swearingen and Sinha 2001)
thereby contributing to users’ assessments of the RAs’ found that users’ trust in RAs increased when the RAs
benevolence and integrity. Additionally, RAs that provide provided detailed product information. Expert reviews and
detailed information can educate users about the product other consumers’ ratings, for example, were very helpful in
category in general and the recommended alternatives in consumers’ decision-making processes. Cooke et al. demon-
particular, thus contributing to the users’ perception of the strated that a technique RAs could use to increase the
RAs’ usefulness. Detailed information also promotes users’ attractiveness of unfamiliar recommendations was to provide
perception of RAs’ information quality, thereby enhancing users with additional information about a new product. Bharati
their satisfaction with the RAs. and Chaudhury (2004) also observed a net positive effect of
factors related to information quality (i.e., relevance, accuracy,
Third, research on explanation facilities in knowledge-based completeness, and timeliness) on users’ decision-making
systems (KBS) has demonstrated that explanations can help satisfaction.
alleviate information asymmetry (which occurs when the
Investigating the effects of different types of explanations on
trustee has more or better information than the trustor) and
users’ trust in content-based RAs for digital cameras, Wang
make a KBS more transparent to its users, thus contributing to
and Benbasat (2004a) found that the explanations of RAs’
the users’ trust in the KBS (Gregor and Benbasat 1999). In the
reasoning logic (how explanations) strengthened users’ trusting
same vein, the provision of explanations on how the RAs’
beliefs in the RAs’ competence and benevolence. Likewise,
recommendations reflect users’ preferences and requirements Herlocker et al. (2000) showed that the addition of explanation
will increase users’ trust in the RAs. facilities increased the acceptance of MovieLens, an online
collaborative-filtering movie RA. Most participants in their
It is therefore proposed that study valued the RA’s explanations and wanted them to be
included in MovieLens. Sinha and Swearingen, in a series of
P20: Recommendation content influences users’ studies involving RAs in different domains (Sinha and
evaluations of RAs. Swearingen 2001, 2002; Swearingen and Sinha 2001,

2002), also confirmed that explanations can enhance per- Factors Related to Product,
ceptions of the transparency of RAs’ inner workings, which User, User–RA Interaction
in turn result in increased user trust in the RAs. Similar
effects have also been observed by Wang (2005). The effects of RA use and RA characteristics on users’
How users evaluate RAs is also influenced by recommen- evaluations of RAs are contingent upon such user and
dation format, as reflected by the navigation and layout of product related factors as the type of products, users’
the recommendation presentation interface. Given human expertise with the products or product categories, RA-user
reluc-tance to extend unnecessary cognitive effort, similarity, and user’s familiarity with RAs. The following
consumers generally negatively evaluate an RA that is sections discuss the moderating effects of factors related to
inconvenient to use. Although providing detailed products, users, and user–RA interactions, as captured in
information about recom-mended alternatives is desirable, propositions P22 through P27 and illustrated in Figure 3.
its benefits will not be fully realized unless the navigational
paths to product information and the layout of the product Product-Related Factors. User evaluations of RAs are not
information are clear. It is therefore proposed that likely to be consistent across different product types. In the
context of online shopping, products can be categorized into
P21: Recommendation format influences users’ perceived search products and experience products. On the one hand,
usefulness of, perceived ease of use of, and satisfaction currently available RAs (with text and static images) allow
with the RAs. RAs with a clear navigational path and search products (but not experience products) to be ade-
layout are considered more useful, easier to use, and quately assessed. This gives users opportunities to appraise the
more satisfactory than those without. usefulness of RAs. As such, users may perceive RAs for
search products more useful than RAs for experience products.
In their investigation of RAs in different domains, Sinha and On the other hand, as discussed earlier, since the assessment of
Swearingen (2001; Swearingen and Sinha 2002) found that experience products is more complex than the evaluation of
interface features, such as navigation and layout, were most search products, consumers shopping for experi-ence products
significant when they presented excess obstacles to users. For are more likely to employ other-based (rather than own-based)
instance, users were generally dissatisfied when too many decision-making processes. As a result, compared with
clicks were required to obtain detailed information about consumers shopping for search products, those shopping for
recommended items, or when only a few recommendations experience products may perceive RAs to be more trustworthy
were displayed on each screen. Bharati and Chaudhury, and thus be more inclined to follow the RAs’
however, failed to find a significant relationship between recommendations. It is therefore proposed that
navigational efficiency and user satisfaction.
P22: Product type moderates the effects of RA use on users’
Summary. The propositions presented here help answer trust in and perceived usefulness of RAs. Users have
research question (2.3): How do characteristics of RAs higher trust in RAs for experience products and higher
influence users’ evaluations of RAs? Users’ trust in, per-ceived perceived usefulness of RAs for search products.
usefulness and ease of use of, and satisfaction with an RA are
influenced by RA type as well as by RA charac-teristics Aggarwal and Vaidyanathan (2003b) examined the
associated with input, process, and output designs. Various perceived effectiveness (measured by user perceptions of
types of RAs result in different user perceptions of RAs. For the quality of RAs’ recommendations, satisfaction with the
instance, whereas hybrid or compensatory RAs are perceived recommenda-tions, and intention to acquiesce to a
as more difficult to use, they generally lead to higher trust and recommendation) of two types of RAs (content-based RAs
are considered more useful and satisfactory. Adequate user and collaborative-filtering RAs). They found that users
control, familiar recommendations, detailed information about perceived RAs to be more effective for search goods than
recommended products, and explanations on system logic for experience goods. Senecal (2003; see also Senecal and
enhance users’ trust in RAs and their percep-tion of the Nantel 2004) observed that consumers’ were more likely to
usefulness of RAs. Implicit preference elicitation methods, follow RAs’ recommenda-tions for experience products
ease of generating new recommendations, clear navigational (wines) than for search products (calculators).
paths, and neat layout will increase perceived ease of use of
the RAs. Moreover, information about search progress and User-Related Factors. Research by Nah and Benbasat (2004)
short response time will increase users’ satisfaction with the regarding knowledge-based systems revealed that expert and
RAs. Table 10 summarizes the rela-tionships investigated in novice users exhibited different levels of criticality and
this section. involvement in their area of expertise. Not only were experts

Table 10: The Impact of RA Characteristics on Users’ Evaluations of RA

Q2.2: How do characteristics of RAs influence users’ evaluations of RAs?
RA Type
RA Type P14
• Hybrid vs. Content- Trust, Users considered meta-recommender systems that combine collab- • P14a
Filtering (or Perceived orative-filtering and content-filtering more helpful than traditional
collaborative- Usefulness, recommender systems (using only collaborative-filtering technique)
filtering) Perceived Ease (Schafer et al. 2002).
of Use, and
• Compensatory vs. Satisfaction No empirical study available. • P14b
Non-compensatory
RA Input
Perceived Ease
Preference Elicitation
of Use and No empirical study available. P15
Method
Satisfaction
Ease of generating new or additional recommendations improved
estimations of the ease of use of an RA (Swearingen and Sinha 2001,
Perceived Ease 2002); giving users the capability to return to the preference
Ease of Generating
of Use and specification stage at any time and restate their preferences in-
New/Additional P16
Satisfaction creased their positive responses to RAs (Pereira 2000); an RA’s
Recommendations
system quality (i.e., ease of use, convenience of access, and
reliability) had positive effect on users’ decision making satisfaction
(Bharati and Chaudhury 2004).
Increased user control over interaction with RA resulted in increased
users’ trust in and satisfaction with two content-filtering RAs (Pereira
2000); user-controlled RAs generated higher user satisfaction than
Trust,
did system-controlled RAs. Users of user-controlled RAs also thought
Satisfaction,
User Control the RA better understood their tastes and were more loyal to the P17
and Perceived
system (McNee et al. 2003); control process was one of the top
Usefulness
contributors to users’ trust in a virtual salesperson (Komaik et al.
2005); RAs that were perceived by users as more restrictive were
considered less trustworthy and useful by them (Wang 2005).
RA Process
Informing online shoppers about the progress of a search augmented
Information about shoppers’ perceptions of the effort saved for them by an RA and,
Satisfaction P18
Search Progress consequently, increased their satisfaction with the decision-making
process before purchasing the product (Bechwati and Xia 2003).
Long wait time before receiving recommendations decreased users’
preference for shopbots (Basartan 2001); the time taken by users to
Response time Satisfaction register and to receive recommendations from RAs did not have a P19
significant effect on users’ perceptions of the RA (Swearingen and
Sinha 2001, 2002).

Table 10: The Impact of RA Characteristics on Users’ Evaluations of RAs (Continued)

RA Output
Recommendation P20
content
• Familiar • Familiar recommendations played an important role in establishing • P20a
recommendations user trust in an RA (Sinha and Swearingen 2001; Swearingen and
Sinha 2001); unfamiliar recommendations lowered users’ evalua-
tions of a simulated music CD RA (Cooke et al. 2002).
• Composition of the • No empirical study available • P20b

list of recommen-
dations
• Detailed informa- • User trust in an RA increased when the RA provides detailed • P20c
tion about Trust, product information (Sinha and Swearingen 2001; Swearingen and
recommended Perceived Sinha 2001); a technique RAs can use to increase the attractive-
products Usefulness, and ness of unfamiliar recommendations was to provide users with
Satisfaction additional information about a new product (Cooke et al. 2002); an
RA’s information quality (i.e., relevance, accuracy, completeness,
and timeliness) had a significant effect on users’ decision making
satisfaction (Bharati and Chaudhury 2004).
• Explanation • Explanations of an RA’s reasoning logic strengthened users’ • P20d

trusting beliefs in the RA’s competence and benevolence (Wang
and Benbasat 2004a); the addition of explanation facilities
increased the acceptance of MovieLens, an online collaborative-
filtering movie RA (Herlocker et al. 2000); explanations enhanced
perceptions of the transparency of an RA’s inner workings,
resulting in increased user trust in the RA (Sinha and Swearingen
2001, 2002; Swearingen and Sinha 2001, 2002; Wang 2005).
Interface features such as navigation and layout were most significant
when they presented excess obstacles to users. Users were
Recommendation Perceived
generally dissatisfied when too many clicks were required to obtain
format Usefulness,
detailed information about recommended items, or when only a few
• Navigation and Perceived Ease recommendations were displayed on each screen (Sinha and P21
layout of Use, and
Swearingen 2001; Swearingen and Sinha 2002); there was no
Satisfaction
significant relationship between navigational efficiency and user
satisfaction (Bharati and Chaudhury 2004).
less likely to be persuaded by a knowledge-based system than evaluating alternatives, individuals with high product exper-
were novices, experts also found such a system less useful tise may perceive RAs as constraining and hampering their
than did novices. In the context of RAs, the extent to which knowledge-based search, feeling that the RAs make them do
consumers have expertise in a particular product affects their something other than what they are capable of doing (Kamis
evaluation of RAs for that product (King and Hill 1994). and Davern 2004). As such, users with low product expertise
While RAs may be a necessity for consumers with low are expected to evaluate RAs more favorably than are users
product expertise who require assistance in selecting and with high product expertise. It is therefore proposed that

P23: Product expertise moderates the effects of RA use on knowledge component that translates users’ needs to attribute
users’ evaluations of RAs (i.e., trust, perceived useful- specifications, which are in turn translated to recommenda-
ness, perceived ease of use, satisfaction). The higher tions. Users with high product expertise may consider needs-
the product expertise of the users, the less favorable based RAs less transparent and thus less trustworthy than
the users’ evaluations of RAs. feature-based RAs that directly translate users’ specifications
to recommendations. As such, we expect consumers with high
Van Slyke and Collins (1996) noted that users’ knowledge product expertise to favor feature-based RAs and those with
(including both domain knowledge and technical knowledge) lower product expertise to desire needs-based RAs.
is a key component in the trust-building process between the
users and the RAs. Consumers with little product knowledge Additionally, while experienced consumers tend to base
are more likely to develop trust when a salesperson recog- their product choice on attribute information, individuals
nizes their concerns, listens to their needs, and takes on the with low product expertise have been found to seek more
role of a consultant. Similarly, less knowledgeable consumers summary information (Brucks 1985), given their lack of
generally exhibit greater preferences for using web-based ability to process attribute information as efficiently. As
advisors than more knowledgeable consumers. Urban et al. such, collab-orative-filtering RAs, which do not require
(1999) compared subjects’ relative preferences for two web- users to specify product attribute preferences, are likely to
sites selling pick-up trucks online: in their experiment, one appeal more to consumers with lower product expertise
website was equipped with an RA, the other was not. Although than to those with higher product expertise, who may
the overall preferences for the two sites appeared to be about perceive the absence of attribute information in the
equal, consumers participating in the study who were not very collaborative-filtering RAs as preventing them from using
knowledgeable about trucks expressed stronger preferences for their knowledge in evaluating alternatives (Pereira 2000).
an RA-enabled website, whereas those who were experts
demonstrated stronger preferences for the web-site that lacked In sum, users’ product expertise affects their evaluation of
an RA. Thus, in the pick-up truck study, RA advice appeared different types of RAs. It is therefore proposed that
to be more valuable to consumers with a lower level of
product knowledge. Similarly, Spiekermann (2001), in an P24: Product expertise moderates the effects of RA type on
experiment that involved a three-dimensional users’ evaluations of RAs (i.e., trust, perceived useful-
anthropomorphic RA, observed that highly knowledgeable ness, perceived ease of use, satisfaction). The higher the
subjects were generally less satisfied with the RA and there- product expertise of the users, the more (less) favorable
fore less reliant on it for choosing products than less-knowl- the users’ evaluations of feature-based (needs-based)
edgeable subjects. Finally, examining the effects of product RAs. The higher the product expertise of the users, the
category knowledge on users’ perceptions of online shopping more (less) favorable the users’ evaluations of content-
tools, Kamis and Davern (2004) observed that product filtering (collaborative-filtering) RAs.
category knowledge was negatively related to perceived ease
of use and perceived usefulness of the decision tools. Focusing on users with low product knowledge, Komiak and
Benbasat (2006) found that needs-based RAs led to users’
Product expertise is also expected to influence users’ percep- beliefs that the RAs fully understood their true needs and took
tions of different types of RAs. For instance, while experi- their needs as the RAs’ own preferences, which in turn
enced consumers can process attribute information efficiently, resulted in higher trust regarding the RAs’ competence,
individuals with low product expertise will likely find such benevolence, and integrity. Felix et al. (2001) further ob-
tasks more difficult (Pereira 2000). Attribute-oriented mes- served that users considered advice from needs-based RAs
sages are found to be less informative to novices. Given the better suited for product novices than that from feature-based
difficulty for individuals with low product expertise to RAs. Stolze and Nart (2004) also found that users preferred
enumerate the product attributes considered and gauge RAs equipped with both needs-based and feature-based
attribute importance, it is expected that they will find feature- questions to those equipped with feature-based questions only.
based RAs uninformative but show positive affective reac- Pereira (2000) investigated the interaction effects between RA
tions (i.e., trust, perceived usefulness and ease of use, type (content-filtering versus collaborative-filtering) and users’
satisfaction) to needs-based RAs. In contrast, for consumers product class knowledge. He found that users with high
with high product expertise, needs-based RAs may hamper product class knowledge had more positive affective reactions
their ability to specify exact attribute-related requirements and (trust and satisfaction) to the content-filtering RAs than the
will thus be considered less useful and more difficult to use collaborative-filtering ones. The reverse was true for users
than feature-based RAs. Moreover, needs-based RAs have a with low product class knowledge.

Factors Related to User–RA Interaction. The similarity significance vested in certain attributes by RAs and the
between RAs and their users, user’s familiarity with RAs, significance that would be given to those attributes by users
as well as the confirmation/disconfirmation of users’ influenced user perceptions of the utility of the recom-
expecta-tions will also moderate the effects of RA use on mendations generated by the RA.
users’ evaluations of the RAs.
Individuals develop trust over time as they accumulate knowl-
In line with the theory of interpersonal similarity, individuals edge relevant to trust through their experiences with each other
will be attracted to other individuals who exhibit similar (Luwicki and Bunker 1995; McKnight et al. 1998). Familiarity
characteristics (Byrne and Griffitt 1969). They tend to iden- is a necessary condition for developing knowl-edge-based
tify with and have positive attitudes toward similar others (in trust. Therefore, users may judge the trustworthi-ness of RAs
terms of decision-making strategies, attitudes, tastes, goals, or on the basis of their behavioral experiences. Repeated use of
preferences). McKnight et al. (1998) describe unit grouping as RAs, or RA training, not only familiarizes the user with the
one type of cognitive categorization processes individuals use workings of the RAs but also enables them to assess the
to develop trusting beliefs in new relationships. Unit grouping performance and consistency of the RAs. Sinha and
means “to put the other person in the same category as Swearingen (2002) suggest that one way for a user to decide
oneself” (p. 480). They argue that individuals who are grouped whether to trust recommendations is to examine the success of
together tend to form more positive trusting beliefs about each prior suggestions from the RAs, as evidence of the RAs’
other, because they tend to share common goals and values.
credibility. It is therefore proposed that
Studies conducted by Brewer and Silver (1978) and Zucker et
al. (1996) have likewise provided evidence that unit grouping P26: User’s familiarity with RAs moderates the effects of
quickly leads to highly positive trusting beliefs. In the context
RA use on trust in the RAs. Increased familiarity with
of RAs, similarities between users and RAs can promote a
RAs leads to increased trust in the RAs.
sense of group membership, and thus enhance users’
perceptions of attractiveness and trustworthiness of RAs.
Komiak and Benbasat (2006) observed that users’
Moreover, insomuch as the main utility of the RAs lies in their
familiarity with the workings of RAs (e.g., the way to
capability to provide recommendations that match users’
specify their pre-ferences to the RAs, to access the
preferences, similarities between users and RAs may result in
recommendations that better fit users’ needs and thus
explanations, and to review information on recommended
contribute to more optimistic perceptions of the usefulness of items) allowed them to develop trust-relevant knowledge
the RAs. It is therefore proposed that and to assess the consistency of the RAs’ actions. Their
empirical study demonstrated that familiarity increased
P25 User–RA similarity moderates the effects of RA use on users’ trust in the RAs’ benevolence and integrity, but it did
users’ trust in, satisfaction with, and perceived useful- not influence their trust in the RAs’ competency.
ness of RAs. The more the RAs are perceived to be
similar to their users, the more they are considered to According to the confirmation-disconfirmation paradigm,
be trustworthy, satisfactory, and useful. consumers form satisfaction based on their confirmation level
and the expectation on which that confirmation was based
In their study of consumer acceptance of online movie RAs’ (Bhattacherjee 2001). Confirmation occurs when perceived
advice, Gershoff et al. (2003) observed that when assessing performance meets the expectation. Positive (negative) dis-
how informative RAs were, consumers paid greater attention confirmation occurs when perceived performance exceeds
to past instances when they agreed with the RAs’ opinions and (falls below) the expectation. Satisfaction is achieved when
ratings (an indication of similarity in tastes or pre-ferences). expectations are fulfilled (i.e., confirmed). Negative discon-
Higher rates of agreement led to greater confidence in and firmation of expectations will result in dissatisfaction, where-
greater likelihood of accepting the RAs’ advice. More-over, in as positive disconfirmation will result in enhanced satisfaction
addition to the overall agreement rate, consumers paid special (Selnes 1998).
attention to the intensity of the agreements (i.e., agreement on
extreme, highly positive or negative, past opinions). Hess et al. Applying confirmation-disconfirmation theory to RA context,
(2005) observed that personality similarity between the users we expect that the confirmation or the positive disconfirma-
and the RAs contributed to in-creased involvement with the tion of users’ expectations about RAs’ functionalities and per-
decision aid, which in turn resulted in increased user formance will enhance users’ satisfaction with RAs, whereas
satisfaction with the RAs. Aksoy and Bloom (2001) also the negative disconfirmation of their expectations will lead to
demonstrated that similarities in the dissatisfaction. It is therefore proposed that

P27: The confirmation/disconfirmation of expectations Shaffer 1991; Stamm and Dube 1994; Verplanken 1991).
about RAs moderates the effects of RA use on users’ Sources that consumers attribute with high credibility appear
satisfac-tion with the RAs. Confirmation or positive to influence consumer attitudes more significantly (Goldberg
disconfir-mation of users’ expectations about RAs and Hartwick 1990; Stamm and Dube 1994), and their efforts
contributes positively to users’ satisfaction with the to persuade consumers are more effective (Lirtzman and Shuv-
RAs. In con-trast, negative disconfirmation of users’ Ami 1986; Mondak 1990). In the context of this paper, source
expectations about RAs contributes negatively to credibility refers to the credibility of RA providers, determined
users’ satisfaction with the RAs. by the type and the reputation of the RA pro-viders, both of
which influence users’ trusting beliefs in RAs’ competence,
No empirical study has directly investigated the effects of benevolence, and integrity, as captured in proposition P28 and
expectation (dis)confirmation on satisfaction with RAs. illustrated in Figure 3.
However, several IS researchers have called attention to the
importance of managing consumer expectations in the design According to Doney and Cannon (1997), trust can develop
of RAs (Komiak and Benbasat 2006; Wang and Benbasat through a transference process, in which “trust can be trans-
2004b; West et al. 1999), suggesting that consumers might ferred from one trusted ‘proof source’ to another person or
lose faith in and stop using an RA when it provides recom- group with which the trustor has little or no direct experience”
mendations that do not meet with their expectations. Sinha and (p. 37). For example, a new salesperson representing a repu-
Swearingen (2001; Swearingen and Sinha 2002) found that, table company would benefit from buyers’ previous positive
although users generally trusted RAs that provide familiar experiences with the company. A website that features an RA
recommendations, they were often disappointed with RAs that is referred to in this paper as the provider of the RA. The type
provided too many familiar recommendations, because such and the reputation of the RA providers may affect users’
RAs failed to help them broaden their horizons. Komiak and trusting beliefs in the RAs’ competence, benevolence, and
Benbasat (2004) suggest that RAs should use needs-based integrity, because the user may transfer trust, or distrust, from
preference-elicitation questions as a way of managing users’ the providers to the RAs provided at those websites. West et al.
expectations. Protocol analyses by Komiak et al. (2005) and (1999) call attention to the fact that characteristics of a website
Wang and Benbasat (2004b) revealed that expectation provide important cues for building trust in an online shopping
disconfirmation was an important factor contri-buting to advisor. Urban et al. (1999) also state that trust in a specific
distrust in RAs. website is the first stage toward development of trust in an
expert advisor; trust cannot be vested in an expert advisor until
Summary. The propositions presented here provide the ans- trust has been established toward the Internet and the website
wer to research question (2.3): How do other factors (factors that provides the advisor.
related to user, product, and user–RA interaction) moderate
the effects of RA use and RA characteristics on users’ RAs are embedded either in the websites of online vendors
evaluations of RAs? RAs for search (experience) products are (e.g., Amazon.com) or in third party websites (e.g., Price-
considered more useful (trustworthy). Users with more pro- line.com). Prior research (see Senecal 2003; Senecal and
duct expertise tend to have less favorable perceptions of RAs Nantel 2004) has shown that consumers tend to discredit
in general. However, the higher the product expertise of the recommendations from endorsers if they suspect that the latter
users, the more favorable their evaluations of feature-based have non-product related motivations to recommend a parti-
and content-filtering RAs. In addition, the greater the RAs are cular product (e.g., overstocking of that product). Therefore, as
perceived to be similar to their users and the greater the users’ endorsers of RAs’ recommendations, independent third party
familiarity with the RAs, the higher the users’ trust in the RAs. websites may be perceived by consumers as less biased and
Finally, whereas confirmation (or positive disconfir-mation) of more credible than vendor websites.
users’ expectations about RAs will enhance their satisfaction
with the RAs, negative disconfirmation of their expectations Moreover, McKnight et al. (1998) describe reputation catego-
will hamper their satisfaction with RAs. Table 11 summarizes rization, the assignment of attributes to another person based
the relationships investigated in this section. on second-hand information about the person, as one type of
process individuals use to develop trusting beliefs in a new
relationship. Individuals with good reputations are catego-
Provider Credibility rized as trustworthy individuals who are competent, benevo-
lent, honest, and predictable. Individuals will quickly develop
The effect of source credibility has been extensively inves- trusting beliefs about a person with a good reputation, even
tigated by researchers studying communications (Smith and without firsthand knowledge of them. Applying this reputa-

Table 11. The Impact of Factors Related to User, Product, and User–RA Interaction on Users’
Evaluations of RA
Q2.3: How do other factors (i.e., factors related to user, product, and user–RA interaction) moderate the effects of RA
use and RA characteristics on users’ evaluations of RAs?
Moderated
Moderator Relationship Empirical Support P
Product-Related Factors
RA Use on Users perceived RAs to be more effective for search goods than for
Product Type
Trust and experience goods (Aggarwal and Vaidyanathan 2003b); consumers’ were
(search vs. P22
Perceived more likely to follow RA’s recommendations for experience products than
experience)
Usefulness for search products (Senecal 2003; Senecal and Nantel 2004).
User-Related Factors
Less knowledgeable consumers expressed stronger preferences for an
RA Use on RA-enabled website, whereas those who were experts evinced stronger
Trust, Perceived preferences for the website that lacked an RA (Urban et al. 1999); highly
Product Usefulness, knowledgeable subjects were generally less satisfied with the RA and
P23
Expertise Perceived Ease therefore less reliant on it for choosing products than less-knowledgeable
of Use, and subjects (Spiekermann 2001); product category knowledge was nega-
Satisfaction tively related to perceived ease of use and perceived usefulness of the
decision tools (Kamis and Davern 2004).
For users with low product knowledge, a needs-based RA resulted in
higher trust (Komiak and Benbasat 2006); users considered advice from
RA type on needs-based RAs better suited for product novices than that from feature-
Trust, Perceived based RAs (Felix et al. 2001); users preferred RAs equipped with both
Product Usefulness, needs-based and feature-based questions to those equipped with
P24
Expertise Perceived Ease feature-based questions only (Stolze and Nart 2004); users with high
of Use, and product class knowledge had more positive affective reactions (trust and
Satisfaction satisfaction) to the content-filtering RAs than the collaborative-filtering
ones. The reverse was true for users with low product class knowledge
(Pereira 2000).
User–RA Interaction
When assessing how informative an RA was, consumers paid greater
attention to past instances when they had agreed with the RA’s opinions
and ratings. Higher rates of agreement led to greater confidence in and
RA Use on greater likelihood of accepting an RA’s advice (Gershoff et al. 2003);
User–RA Trust, Perceived personality similarity between the user and the decision aid contributed to
P25
Similarity Usefulness, and increased involvement with the decision aid, which in turn resulted in
Satisfaction increased user satisfaction with the decision aid (Hess et al. 2005);
user–RA similarity in attribute weighting had a significant impact on user
perceptions of the utility of the recommendations generated by the RA
(Aksoy and Bloom 2001).
The user’s familiarity with the workings of an RA increased trust in an
User’s Familiarity RA Use on
RA’s benevolence and integrity, but not its competency (Komiak and P26
with RAs Trust
Benbasat 2006).
Confirmation/
RA Use on
Disconfirmation No empirical study available. P27
Satisfaction
of Expectations
Table 12. The Impact of RA Use and Provider Credibility on Users’ Evaluations of RA
Q2.4: How does provider credibility influence users’ evaluations of RAs?
Provider Credibility The type of a website (i.e., seller, commercially linked third-party, or independent
(Provider Type and Trust third-party) that provides an RA did not affect the perceived trustworthiness of the P28
Provider Reputation) RA (Senecal 2003; Senecal and Nantel 2004).
tion categorization in the context of RAs, we posit that users provided by vendors’ websites. Additionally, they will have
will consider reputable providers more trustworthy than those higher trust in RAs provided by reputable websites than
that are unknown or had a bad reputation, and subsequently those provided by unknown or non-reputable websites.
they will transfer that trust to the RAs featured by the
providers. For instance, a consumer may trust the RA at
Amazon.com more than a similar RA at an unknown website. Relationships among Other Variables
In sum, the type and reputation of RA providers determine This paper focuses on investigating the effects of RA user, RA
their credibility, which in turn influences users’ trust in the characteristics, and other factors (i.e., factors related to user,
RAs. It is therefore proposed that product, user–RA interaction, and provider credibility) on two
groups of outcomes of RA use: (1) consumer deci-sion-making
P28: Provider credibility, determined by the type of RA processes and outcomes and (2) users’ evalua-tions of RAs
pro-viders and the reputation of RA providers, (i.e., perceived usefulness, perceived ease of use, trust, and
influences users’ trust in RAs. RAs provided by satisfaction). The relationships between the two groups of
independent third party websites are considered more outcome variables, among the variables in each group, as well
trustworthy than those provided by vendors’ websites. as between the outcome variables and RA reuse intention and
RAs provided by reputable websites are considered reuse behavior, albeit important, are beyond the scope of this
more trustworthy than those provided by websites that paper. Although empirical investiga-tions of these
are unknown or non-reputable. relationships can be found in general IS litera-ture, there is
little discussion in specific RA literature, which is the focus of
Senecal (2003; see also Senecal and Nantel 2004) identified this review. Therefore, no proposition is stated for these
three types of websites: seller, commercially linked third- relationships in this section. However, for compa-tibility with
party, and independent third-party. She hypothesized that the prior IS work, such relationships are highlighted by the dashed
type of the website that provides an RA will affect the per- lines in Figure 1 and briefly explained below.
ceived trustworthiness of the RA: the RA at an independent
third-party website will be perceived as the most trustworthy, Prior research (e.g., Davis and Kottemann 1994; Kottemann
while the RA at a seller’s website will be perceived as the least and Davis 1994) reveals that active involvement of users in
trustworthy. However, her experimental data did not support interacting with a decision support system may create an
this hypothesis. We believe that this failure may be a result of “illusion of control” (defined as a person’s expectation of
the limited number of alternatives (only four) avail-able in her success on a task that is inappropriately higher than objective
study for each product, as well as the relatively low value of circumstances warrant—Langer 1975) causing users to over-
the products (calculators and wine) involved in the estimate its effectiveness, resulting in an effort–confidence
experiment. No other reported study has directly ex-amined link and a related mismatch between actual performance (e.g.,
the effect of provider credibility on trust in RAs. objective decision quality) and performance beliefs (e.g.,
confidence belief and usefulness belief). Therefore, the rela-
Summary. Proposition P28 provides answers to research tionships between decision effort and decision quality as well
question (2.4): How does provider credibility influence users’ as between decision quality and perceived RA usefulness will
evaluations of RAs? Users’ trust in RAs is proposed to be be influenced by users’ illusion of control.
influenced by the type and reputation of RA providers, as
summarized in Table 12. Users will have higher trust in RAs There is also ample empirical evidence in the IS literature
provided by independent third party websites than those supporting the causal link from subjective evaluations (such

as trust, perceived usefulness, perceived ease of use, and the paper.25 The answers should be of interest to academic
satisfaction) to adoption intention and adoption behavior researchers, designers of RAs, and current or potential pro-
(Davis 1989; Gefen et al. 2003; Taylor and Todd 1995; viders of RAs. Next, we present the contributions of this
Venkatesh 2000). Several researchers have developed inte- study and identify areas for future research.
grated models of TAM, trust, and satisfaction. For instance,
integrating trust, risk, and TAM, Pavlou (2003) and Gefen et
al. (2003) hypothesized about and empirically confirmed the Contributions to Research
links from trust to perceived usefulness and adoption inten-
tion. Wixom and Todd (2005) integrated technology accep- Prior research on RAs has focused mostly on developing and
tance and satisfaction literature and empirically demonstrated evaluating different underlying algorithms that generate
that system satisfaction and information satisfaction contri- recommendations. This paper has attempted to identify other
bute to perceived ease of use and perceived usefulness, important aspects of RAs, namely RA use, RA characteristics,
respectively. Wang and Benbasat (2005) have extended the provider credibility, and factors related to product, user, and
integrated trust–TAM model (Gefen et al. 2003) to online RA user–RA interaction, which exert influence on users’ decision
adoption and proven the causal link from trust to adoption making processes and outcomes as well as their evaluations of
intention. Abstracting from all of this literature, we expect that RAs. One of our objectives was to go beyond generalized
trust, perceived usefulness, perceived ease of use, and models such as TAM to identify the RA-specific features, such
satisfaction will positively contribute to users’ intentions to as RA input, process, and output design characteristics that
use RAs in the future and, subsequently, their future use of influence users’ beliefs and evaluations, including usefulness
RAs. In addition, we expect to observe the following rela- and ease-of-use concerning RA use.
tionships: satisfaction  perceived ease of use; perceived ease
of use  trust; perceived ease of use, trust, and satis-faction Using the conceptual model illustrated in Figure 1 as a starting
 perceived usefulness. point, we derived 28 propositions concerning the out-comes of
RA use from five main theoretical perspectives, including
Only a few studies have directly investigated the relationship theories of human information processing, the theory of
between perceptions of RAs and intention to adopt the RAs. interpersonal similarity, the theories of trust formation, TAM,
Wang and Benbasat (2005) found that consumers’ initial trust and the theories of satisfaction. We have also justified the
not only directly influenced their intention to adopt RAs but it propositions with existing empirical work in RAs (when
also exerted an indirect effect on such intention by enhancing available). Tables 7 through Table 12 provide summaries of
perceptions of the usefulness of the RAs. Users’ perceptions of the propositions that have been presented and empirical
the ease of use of RAs, however, did not have a significant evidence (if available) associated with each of them. As
impact on their adoption intention. Komiak and Benbasat delineated in Table 13, the majority of the propositions are
(2006) distinguished between two levels of RA use intentions: supported fully (including 13 propositions and four sub-
intention to use RAs as decision aids (i.e., to let RAs narrow propositions) or partially (including three propositions and
down product choices) and intention to use RAs as delegated four sub-propositions) by available empirical evidence.
agents (i.e., to let RAs make decisions on behalf of However, there also exist three propositions and two sub-
consumers). The results of their experimental study of a digital propositions that lack conclusive empirical support, in addition
camera RA demonstrated that both cognitive trust and to two propositions that were not bolstered by available
emotional trust were significant predictors of consumer inten- empirical studies, which signals the existence of contingency
tion to use RAs as decision aids. Additionally, emotional trust factors not yet uncovered and/or problems with experimental
fully mediated the impact of cognitive trust on the inten-tion to designs and RA implementations. As such, further theorizing
use RAs as delegated agents. or more rigorous experimental designs are needed to
investigate the proposed relationships. Moreover, there are two
propositions and four sub-propositions that are yet to be
empirically tested. Table 13 highlights the discre-pancy
Discussion and Concluding Comments between what we know and what we need to know,
pinpointing venues of future research to close this breach, as
discussed further in the “Suggestions for Future Research.”
In this paper, we have presented a set of theory-based propo- 25

sitions concerning the outcomes of RA use and RA adoption A summary of the propositions presented in this paper as well as the
research questions to which the propositions provide answers can be found
intentions in e-commerce settings. The propositions provide in Appendix C.
answers to the two research questions that initially motivated

Table 13. Summary of Empirical Support for Propositions

Propositions Sub-Propositions
Fully Supported P5, P7, P8, P10, P11, P16, P17, P18, P22, P23, P24, P25, P26 P6a, P20a, P20c, P20d
Partially Supported P4, P12, P13 P3a, P3b, P6b, P14a
Inconclusively Supported P2, P19, P21 P1a, P1b,
Not Supported P9, P28
No Empirical Study P27, P15 P1c, P3c, P14b, P20b
Available
Note: The shaded areas are those in which promising future studies can be conducted.
To keep our conceptual model manageable, we have not e-commerce websites, following from our improved under-
intended for it to encompass all of the constructs discussed in standing of such phenomena. It must be noted, however,
prior RA studies. Although an examination of the empirical that such prescriptions are dependent on the empirical
studies reviewed in this paper reveals a few constructs (e.g., validation of the appropriate propositions.
users’ cost–benefit consideration, multimedia vividness) out-
side of the research framework that we have proposed, they This review has identified two sets of factors affecting con-
either do not appear to influence outcomes of RA use or their sumers’ decision-making processes and outcomes and
effects are still not well-understood. For instance, it is pos- users’ evaluations of e-commerce RAs: (1) factors that are
sible that cost–benefit considerations associated with RA use under the control of RA designers or providers, and (2)
might influence users’ perceptions of and interactions with those that are not.
RAs. Spiekermann (2001) hypothesized that perceived costs
and benefits of searching for product information may affect The following factors can be controlled to a certain extent
users’ interactions with RAs. However, experimental data has by developers of RAs or websites providing RAs, hence
not supported her hypothesis. Neither costs nor benefits were they are suggested to be implemented (or, in the case of RA
found to influence users’ willingness to interact with RAs. A charac-teristics, be incorporated into the design of RAs) to
few other studies (Komiak and Benbasat 2006; Wang and improve users’ shopping decision-making and enhance
Benbasat 2004a) have included effort–quality preference as their positive evaluation of the RAs.
covariates in their analyses, but none of these constructs were
found to be significant. Similarly, Hess et al. (2005) hypothe- • Insomuch as the use of RAs generally results in increased
sized about the positive impact of multimedia vividness on decision quality (P2) and reduced decision effort (P1),
user involvement with RAs, only to be surprised with a signi- RAs should be implemented in e-commerce websites to
ficant hard-to-explain negative effect. assist consumers with shopping decision making.
Based on the fact that (1) the conceptual model presented in • Explicit preference elicitation methods are considered
Figure 1 integrates most of the constructs, as well as the more transparent by users and lead to higher decision
inter-relationships among the constructs, identified in quality (P4). However, implicit preference elicitation
previous research in RAs, and (2) the model and methods demand less decision effort (P4) and are con-
propositions derived from five different theoretical sidered easier to use and more satisfactory (P15). It is
perspectives not only sum-marize prior empirical effort but suggested that explicit and implicit preference
also provide directions for future research (as illustrated in elicitation methods be integrated so as to balance
Table 13 and discussed further below), we conclude that the quality with effort.
conceptual model and the theoretical propositions provide
an adequate framework to account for phenomena relating • The provision of explanations augments perceptions of
to the outcomes of RA use in e-commerce. the transparency of RAs’ reasoning logic and increases
users’ trust in and satisfaction with the RAs (P20d). As
such, explanations should be provided for how the RAs
Contributions to Practice derive their recommendations.
In this section, prescriptive guidelines are suggested to prac- • Since different types of RAs (1) result in different
titioners concerning the design and implementation of RAs in deci-sion-making processes and outcomes as well as in

different user evaluations (P3 and P14) and (2) appeal to • Similarity between RAs and their users simplifies and
users of different product expertise (P24), it is suggested reduces product search, improves decision quality, and
that multiple RAs be provided (if cost permits) to give increases trust in, perceived usefulness of, and satisfac-
different users the flexibility of choosing the desired RAs tion with the RAs (P13 and P25). RAs can be made
to assist in their shopping tasks. “similar” by the use of needs-based questions that elicit
users’ product preferences and their choices of the
• Inasmuch as familiar recommendations increase users’ decision strategies the users prefer (e.g., elimination by
trust, RAs should present unfamiliar recommendations in aspect), when such information is available. RAs can also
the context of familiar ones (as indicated by sub-propo- be designed to assume personalities (e.g., extra-version or
sitions P20a and P20b). To be able to provide familiar introversion) similar to those of their users, as illustrated
recommendations, RAs must track and apply users’ shop- by Hess et al. (2005).
ping history, feedback, and Internet navigation patterns. A
• The confirmation/disconfirmation of users’ expectations
possible way to facilitate the provision of familiar
about RAs influences their satisfaction with the RAs
recommendations to new users is to generate recom-
(P27). West et al. (1999) suggest that one way to man-age
mendations known to be very popular.
consumer expectations regarding RAs’ performance is to
communicate the RAs’ limitations and requirements (e.g.,
• The provision of detailed information about RAs’ recom-
the kind and amount of user input required) to customers
mendations increases users’ trust in, perceived usefulness
before they begin using the RAs, so that they can set
of, and satisfaction with the RAs (as indicated by sub-
realistic expectations for the agents. For instance,
proposition P20c). The information can include product
MovieCritic, a movie RA, educates users about the
descriptions, expert reviews, and other consumers’ eval-
importance of providing input to the RA, and it provides
uations. This information must also be easily accessible
movie recommendations only after users have rated at
by the users. Thus, clear navigational paths and clear least 12 films and answered a battery of questions. It
layout (P21) are very important design considerations. manages users’ expectations by informing them that the
Furthermore, RAs should provide their recommendations quality of its product recommendations is related to the
in the format of sorted lists (P7a) but avoid presenting too quality and quantity of input users provide to the system.
many product recommendations, particularly on a single Spiekermann and Paraschiv (2002) argue that, when
screen (P7b). conceiving a system for commercial purposes, it is
important to consider the specific expectations of poten-
• RA designers should allow users to control their inter- tial buyers regarding an interface’s functionalities. They
actions with the RAs (P17) and provide capabilities for propose a “user-centric” approach to RA design (i.e., an
them to generate new or additional recommendations approach to RA design that emphasizes the users’ points
easily (P16). Additionally, the RAs should keep their of view) in order to motivate user interaction with RAs.
response times to a minimum (P19). When the users are
waiting to receive recommendations, RAs should display • The type and reputation of RA provider can increase
information about their search progress to demonstrate users’ trust in the RA (P28). RA providers are advised
their effort to the users (as indicated by proposition P18). to implement such mechanisms as trust-assuring
arguments (Kim and Benbasat 2003) and third party
• RAs can have greater influence on product choice, seals to signal their competence, benevolence, and
decision quality, and/or decision effort for complex integrity to RA users.
products and experience products (P8, P9, and P22);
thus online retailers are advised to provide RAs for • Since RAs can serve as “double agents,” the product
such pro-ducts, although the implementation of RAs attributes included in the RA’s preference-elicitation
for experi-ence products may require extra investment interface (P5), the recommendations generated by the
in advanced multimedia presentation technologies. RAs (P6a), and the utility scores or predicted ratings for
Since current Internet technology allows search recommended alternatives (P6b) may exert significant
products to be ade-quately assessed prior to purchase, impact on consumers’ decision processes and decision
the RAs for such products are generally considered outcomes. However, retailers should avoid manipulating
more useful than those for experience products (P22). these factors in their own interest and misleading con-
Therefore, retailers of search products are advised to sumers intentionally. It is much easier to destroy than to
incorporate RAs in their online stores. establish consumer trust and confidence.

On the other hand, although the following user-related required to test the propositions related to user–RA
factors also have significant impact on the outcomes of RA interaction factors such as users’ familiarity with RA and
use and RA adoption intention, they are comparatively the confirma-tion/disconfirmation of users’ expectations.
difficult to control a priori:
As noted above, part of our conceptual model has already
• Users’ familiarity with the operations of RAs can been validated by prior RA research. However, in many
increase their trust in the RAs (P26). cases, even fully validated propositions have received only
limited support (e.g., P6a, P7), demonstrating a lack of
• Users’ product expertise affects their perceptions knowledge accumulation in this area. As such, additional
(P23). Novices consider RAs more useful and testing of these propositions is suggested. Moreover, further
trustworthy than do experts. validation should be conducted in areas where no empirical
investigations have been undertaken or where unexpected
• Users’ perceptions of risk attendant to particular or inconclusive results have been obtained in prior
products affect their shopping performance and their endeavors, as demonstrated in Table 13.
perceptions of RAs. RAs have the greatest impact on
decision quality and information search under
conditions of high product risk (P13). Developing the Conceptual Model
Our review suggests important areas for further theoretical

development. First, this paper has investigated only a limited
Suggestions for Future Research number of user-related factors. Future studies should investi-
gate such additional factors as control propensity (i.e., the
Testing the Conceptual Model extent to which an individual is naturally inclined to control
other parties in general), trust propensity, effort and quality
Our conceptual model presented Figure 1 is a causal model preferences, cost–benefit analysis, and susceptibility to inter-
and its propositions should be best tested as such. The impersonal influence. For instance, users’ effort and quality
pacts of RA use, RA characteristics, and other contingency preference may determine their preference for different types
factors on users’ decision-making processes and outcomes, as of RAs: whereas individuals who favor better decision quality
well as on their evaluations of RAs, would preferably be tested may prefer to use compensatory, hybrid, or manual–
utilizing the laboratory experiment method to facilitate the ephemeral RAs, those who desire less effort may choose to
manipulation of independent variables (in particular, different employ non-compensatory, pure collaborative-filtering or
RA characteristics) and the control of extraneous factors. In content-filtering, or automatic–permanent RAs.
fact, most of the empirical studies reported in this paper have
used the laboratory experimental method. Field experiments In addition, this review focuses on two major outcomes of RA
are also possible when partnerships with e-commerce RA use—consumer decision making and users’ evaluations of RAs
providers can be fostered. —in a parallel fashion, without hypothesizing about the
interrelationships between the two groups of dependent vari-
Since the complexity of the conceptual model makes it ables or among the variables in each group. Further theori-
infeasible to validate the model as a whole, it is suggested that zation is needed to explore the between-group relationships, as
the higher-level conceptual model (Figure 1), and even the two well as those among the variables in the consumer decision
lower-level models (Figures 2 and 3), be broken into smaller making group, given that the relationships among the four user
and more manageable parts, allowing different RA evaluation variables (i.e., perceived usefulness, perceived ease
characteristics as well as user- and product-related factors to of use, trust, and satisfaction) have already been exten-sively
be tested one small group at a time. For example, proposi-tions studied in prior literature. In addition to the relation-ships
P17, P19, P20, P22, P25, and P28 can be tested as a group to highlighted with dashed lines in Figure 1 and discussed in the
validate the proposed effects of three RA charac-teristics section “Relationships among Other Variables,” that is, the
associated with input, process, and output, respec-tively (i.e., relationships between decision effort and decision quality, as
user control, response time, and recommendation content), one well as between decision quality and perceived RA useful-
product-related factor (i.e., product type), one user–RA ness, other pair-wise relationships (e.g., trust and decision
interaction factor (i.e., user–RA similarity), and provider effort, satisfaction and decision quality, trust and decision
credibility on users’ evaluations of RAs. Whereas studies on quality) can also be explored. For instance, when users have
one period of use are appropriate for validating most of the trusting beliefs in RAs’ competence, benevolence, and
propositions, a longitudinal approach may be integrity, they may engage in less product search (an indicator

of reduced decision effort), which, according to Diehl (2003), of examining why customers choose to discontinue services to
will enhance their decision quality in an ordered environment. which they subscribe. Venkatesh and Brown (2001) have
They may also have greater confidence in their decisions, an identified a set of critical barriers (e.g., rapid change, high
indicator of decision quality. Conversely, increased decision cost, and lack of knowledge) to home PC adoption. Cenfetelli
quality and reduced decision effort during trial use of RAs and Benbasat have developed an integrated model of usage
may contribute to users’ positive perceptions of the RAs. inhibitors, which include beliefs about information, systems,
Moreover, there may exist strong correlations between and services (Cenfetelli 2004; Cenfetelli and Benbasat 2003).
decision quality and satisfaction, as demonstrated by Parikh Despite their apparent advantages, RAs are currently adopted
and Fazlollahi (2002). by only about 10 percent of shoppers, according to a study by
Montgomery et al. (2004), which lists lack of awareness, lack
The current conceptual model can be expanded not only with of benefits, lack of information, slow response time, and poor
variables discussed in prior RA studies (e.g., users’ cost– interface design as the reasons why people do not use RAs.
benefit consideration, multimedia vividness), but also with Previous literature about RAs has focused mostly on factors
those suggested by pervious DSS research in general. For that contribute to better decision-making quality, trust, and
instance, task is identified by Eierman et al. (1995) as an adoption of RAs. Future research effort should be extended to
important factor affecting user behavior and performance; it is uncovering more factors hampering users’ positive evaluation
defined “as the set of functions that a working person, unit, of RAs and inhibiting their use intentions.
organization is expected to fulfill or accomplish. It is the job
that is to be done using the decision support system” (p. 5).
The construct is not included in our conceptual model since Exploring New RA Design Issues
the “task” is fixed in this paper (i.e., to purchase a product
with/without an RA individually). However, this construct Only RA characteristics discussed in prior conceptual or
may become relevant when investigating RA characteristics empirical work are included in our model to keep its com-
needed for online shopping tasks of different complexity and plexity manageable. Future research may explore other RA
structuredness (e.g., individual shopping versus group design issues. First, with the emergence of new charac-teristics
shopping, or shopping for oneself versus shopping for related to RAs’ input, process, and output design, the impact of
someone else). An additional construct worth further
such characteristics on users’ decision-making processes and
investigation is perceived risk. The concept of perceived risk is
outcomes, as well as on their evaluation of RAs should be
founded on a large body of literature developed in marketing
investigated. For instance, Jiang and Benbasat (2005) have
since the 1960s (Bauer 1960; Cunningham 1967; Dowling and
advocated the provision of virtual product experience (VPE) to
Staelin 1994; Jacoby and Kaplan 1974). It is defined as “an
simulate direct experience. Enabled by visual control (which
assessment consumers make of the consequences of making a
enables online consumers to manipulate product images, for
purchase mistake as well as of the probability of such a
example, to move, rotate, and magnify a product’s image so as
mistake to occur” (Spiekermann and Paraschiv 2002, p. 265).
Spiekermann and Paraschiv have identified two different types to view it from different angles and distances) and functional
of perceived risks in an online environment: product risks control (which enables consumers to sample the different
(which consist of functional, financial, socio-psychological, functions of products) technologies, VPE has been shown to
increase consumers’ product under-standing, brand attitude,
and delivery risks26) and privacy risks. Whereas perceived
purchase intention, and affect, as well as to decrease their
product risk is part of our conceptual model, the impact of
perceived risks (Griffith and Chen 2004; Jiang and Benbasat
perceived privacy risk on the two groups of outcome variables
2005; Li et al. 2003; Suh and Lee 2005). Jiang et al. (2005)
should be explored in the future.
have designed a “multimedia-based interactive advisor” that
Finally, many IS researchers have emphasized the importance integrates visual control and func-tional control with product
of studying inhibitors to use. For instance, Parthasarathy and recommendation technologies. It will be interesting to explore
Bhattacherjee (1998) have called attention to the importance whether such RAs can improve users’ understanding of
product features, promote their trust in the RAs, and enhance
their shopping enjoyment. Qiu (2005) proposes the
26 investigation of the potentials of en-hancing users’ social
Functional risk refers to uncertainty that a product might not perform as
expected; financial risk refers to uncertainty that a product might not be worth experience from interacting with RAs, with the help of
the financial price; socio-psychological risk refers to uncertainty that a poor emerging multimedia technologies, such as animated face and
product choice might harm a consumer’s ego or may result in embar-rassment speech output. Specifically, he aims to explore how RAs’
before his or her friends and family; and delivery risk refers to the uncertainty
that products might not arrive on time or in perfect condition. multimedia and anthropomorphic inter-faces shape and affect
users’ perceptions of social presence,

their trusting beliefs toward the RAs, and perceived advantage to e-commerce leaders. By providing product
enjoyment from interacting with the agents. recommendations based on consumers’ preferences, RAs
have the potential to support and improve the quality of the
Additionally, Brusilovsky and Tasso (2004) advocate the decisions consumers make when searching for and
abandonment of the typical “one-size-fits-all” approach by selecting products online as well as to reduce the
providing better mechanisms for gathering users’ information information overload facing consumers and the complexity
needs. Current RAs provide the same set of preference- of online searches. Despite such apparent advantages, RAs
elicitation questions to all users, which results usually in an are currently used by only 10 percent of online shoppers
extended list of questions intended to cover all of the (Montgomery et al. 2004). Based on a comprehensive
important aspects of products. However, not all aspects of a review of empirical RA research conducted in multiple
given product conveyed by the questions are likely to be disciplines, this paper organizes the knowledge about the
important to all users; sets of questions of importance to users effective design and development of RAs and provides
of different gender, expertise, or goals are usually not uni- advice to IS practitioners on how to improve RA design, in
form. For instance, according to a CNET report on the gender addition to identifying those areas in which research is
gap in digital camera purchases,27 men and women are needed to advance our understanding of RA use and impact
different in their buying motivations and the features they
desire. It is important to investigate whether or not RAs that
are context-sensitive, customizing preference-elicitation ques- Acknowledgments
tions according to users’ backgrounds, expertise, and current
goals, are more likely to be adopted for use. We would like to thank the Natural Science and Engineering
Research Council of Canada (NSERC) and the Social Science and
Furthermore, previous studies on RAs have treated them as Humanities Research Council of Canada (SSHRC) for their
stand-alone systems, independent of other functionalities pro- support of this research. We are grateful to the senior editor,
vided by their hosting websites. Inasmuch as the ultimate associate editor, and three anonymous reviewers whose comments
have improved this paper considerably.
purpose of RAs is to facilitate consumers’ online shopping,
and considering that RAs are just one of the technologies
employed by websites to either provide more information to References28
customers or to try to convert browsers into buyers, it makes
sense to study RA effectiveness as part of an overall inter- Adomavicius, G., and Tuzhilin, A. “Recommendation Tech-
active online system. For instance, a “live help” function with nologies: Survey of Current Methods and Possible
instant chatting and co-browsing capabilities (Qiu and Ben- Extensions,” MISRC Working Paper 03-29, MIS Research
basat 2005; Zhu 2004) can be incorporated into RAs to enable Center, University of Minnesota, Minneapolis, MN, 2003.
customer service representatives to help consumers answer *Aggarwal, P., and Vaidyanathan, R. “Eliciting Online Customers’
difficult preference-eliciting questions, to “push” pages Preferences: Conjoint vs Self-Explicated Attribute-Level Mea-
surements,” Journal of Marketing Management (19:1-2),
containing recommended products or detailed information
2003a, pp. 157-177.
about those products to the customers (e.g., expert reviews or *Aggarwal, P., and Vaidyanathan, R. “The Perceived Effectiveness of
other customers’ testimonies), or to direct customers through Virtual Shopping Agents for Search vs. Experience Goods,”
the entire shopping process. Such RAs will have the potential Advances in Consumer Research (30:1), 2003b, pp. 347-348.
to assist customers in all six stages of online shopping *Aksoy, L., and Bloom, P. N. “Impact of Ordered Alternative
processes (needs identification, product brokering, merchant Lists on Decision Quality: The Role of Perceived Similarity,”
brokering, negotiation, purchase and delivery, product paper presented at the Society for Consumer Research Winter
services, and product evaluation), as identified by Maes et al. Conference, Scottsdale, AZ, February 15-17, 2001.
(1999), whereas currently available RAs focus primarily on Ansari, A., Essegaier, S., and Kohli, R. “Internet
Recommendation Systems,” Journal of Marketing Research
product brokering and merchant brokering. We believe that
(37:3), 2000, pp. 363-375.
this is an important area for future studies. Ariely, D., Lynch, J. G., and Aparicio, M. “Learning by Collab-
orative and Individual-Based Recommendation Agents,”
As indicated by the recent EBay acquisition of Shopping.com
Journal of Consumer Psychology (14:1-2), 2004, pp. 81-95.
(a successful comparison shopping website), recommendation
technologies are considered to be a valuable competitive
28
The empirical studies reviewed in this paper are marked with an asterisk
in the list of references. The papers also appear in alphabetical order by
27http://www.cnet.com/4520-7913_1-5140928-1.html.
authors’ names in Appendix A.

Bajaj, A., and Nidumolu, S. R. “A Feedback Model to Understand Brusilovsky, P., and Tasso, C. “Preface to Special Issue on User
Information System Usage,” Information and Management Modeling for Web Information Retrieval,” User Modeling and
(33:4), 1998, pp. 213-224. User-Adapted Interaction (14:2-3), 2004, pp. 147-157.
Balabanovic, M., and Soham, Y. “Fab: Content-Based, Collabora- Burke, R. R. “Technology and the Customer Interface: What Con-
tive Recommendation,” Communications of the ACM (40:3), sumers Want in the Physical and Virtual Store,” Journal of the
1997, pp. 66-72. Academy of Marketing Science (30:4), 2002, pp. 411-432.
*Basartan, Y. “Amazon Versus the Shopbot: An Experiment About Burnham, T. A., Frels, J. K., and Mahajan, V. “Consumer
How to Improve the Shopbots,” Unpublished Ph.D. Summer Switching Costs: A Typology, Antecedents, and
Paper, Carnegie Mellon University, Pittsburgh, PA, 2001. Consequences,” Academy of Marketing Science (31:2), 2003,
Bauer, R. A. “Consumer Behavior as Risk Taking,” in Dynamic pp. 109-126.
Marketing for a Changing World, R. Hancock (ed.), American Byrne, D., and Griffitt, W. “Similarity and Awareness of
Marketing Association, Chicago, 1960, pp. 389-398. Similarity of Personality Characteristics as Determinants of
*Bechwati, N. N., and Xia, L. “Do Computers Sweat? The Impact Attraction,” Journal of Experimental Research in Personality
of Perceived Effort of Online Decision Aids on Consumers’ (3:3), 1969, pp. 179-186.
Satisfaction with the Decision Process,” Journal of Consumer Campbell, M. C., and Goodstein, R. C. “The Moderating Effect of
Psychology (13:1-2), 2003, pp. 139-148. Perceived Risk on Consumers’ Evaluations of Product Incon-
Bellman, S., Johnson, E. J., Lohse, G. L., and Mandel, N. gruity: Preference for the Norm,” Journal of Consumer
“Designing Marketplaces of the Artificial with Consumers in Research (28:3), 2001, pp. 439-449.
Mind: Four Approaches to Understanding Consumer Behavior Cenfetelli, R. T. “An Empirical Study of the Inhibitors of Tech-
in Electronic Environments,” Journal of Interactive Marketing nology Usage,” in Proceedings of the 25th International
(20:1), 2006, pp. 21-33. Conference of Information Systems, R. Agarwal, L. Kirsch, and
Benbasat, I., DeSanctis, G., and Nault, B. R. “Empirical Research J. I. DeGross (eds.), Washington, DC, December 2004, pp.
in Managerial Support Systems: A Review and Assessment,” in 157-170.
Recent Developments in Decision Support Systems, A. B. Cenfetelli, R. T., and Benbasat, I. “Frustrated Incorporated: An
Whinston (ed.), Springer-Verlag, Berlin, 1991, pp. 383-437. Exploration of the Inhibitors of IT-Mediated Customer
Benbasat, I., and Todd, P. “The Effects of Decision Support and
Service,” in Proceedings of the 9th Americas Conference on
Task Contingencies on Model Formulation: A Cognitive Per-
Information Systems, J. Ross and D. Galletta (eds.), Tampa,
spective,” Decision Support Systems (17:4), 1996, pp. 241-252.
FL, 2003, pp. 281-288.
Benbasat, I., and Todd, P. “The Use of Information in Decision
Chiasson, T., Hawkey, K., McAllister, M., and Slonim, J. “An
Making: An Experimental Investigation of the Impact of Com-
Architecture in Support of Universal Access to Electronic
puter-Based Decision Aids,” MIS Quarterly (16:3), 1992, pp.
Commerce,” Information and Software Technology (44:5),
373-393.
2002, pp. 279-289.
Benbasat, I., and Zmud, R. W. “The Identity Crisis Within the IS
Discipline: Defining and Communicating the Discipline’s Core
Chu, P.-C., and Elam, J. J. “Induced System Restrictiveness: An
Properties,” MIS Quarterly (27:2), 2003, 183-194.
Experimental Demonstration,” IEEE Transaction on Systems,
Bettman, J. R., Luce, M. F., and Payne, J. W. “Constructive Con-
sumer Choice Processes,” Journal of Consumer Research (25), Man, and Cybernetics (20:1), 1990, pp. 195-201.
December 1998, pp. 187-217. Claypool, M., Gokhale, A., Miranda, T., Murnikov, P., Netes, D.,
*Bharati, P., and Chaudhury, A. “An Empirical Investigation of and Sartin, M. “Combining Content-Based and Collaborative
Decision-Making Satisfaction in Web-Based Decision Support Filters in an Online Newspaper,” paper presented at the ACM
Systems,” Decision Support Systems (37:2), 2004, pp. 187-197. SIGIR Workshop on Recommender Systems—Implementation
Bhattacherjee, A. “Understanding Information Systems Continu- and Evaluation, Berkeley, CA, August 19, 1999.
ance: An Expectation-Confirmation Model,” MIS Quarterly *Cooke, A. D. J., Sujan, H., Sujan, M., and Weitz, B. A.
(25:3), 2001, pp. 351-370. “Marketing the Unfamiliar: The Role of Context and Item-
Breese, J. S., Heckerman, D., and Kadie, C. “Empirical Analysis Specific Information in Electronic Agent Recommendations,”
of Predicrtive Algorithms for Collaborative Filtering,” in Pro- Journal of Marketing Research (39:4), 2002, pp. 488-497.
ceedings of the 14th Conference on Uncertainty in Artificial *Cosley, D., Lam, S. K., Albert, I., Konstan, J., and Riedl, J. “Is
Intelligence, G. F. Cooper and S. Moral (eds.), Morgan Kauf- Seeing Believing? How Recommender Systems Influence
mann, San Francisco, April 1998, pp. 43-52. Users’ Opinions,” in Proceedings of the SIGCHI Conference
Brewer, M. B., and Silver, M. “Ingroup Bias as a Function of Task on Human Factors in Computing Systems, Fort Lauderdale,
Characteristics,” European Journal of Social Psychology(8:3), FL, April 2003, pp.585-592.
1978, pp. 393-400. Coupey, E., Irwin, J. R., and Payne, J. W. “Product Category
Brown, D. L., and Jones, D. R. “factors That Influence Reliance Familiarity and Preference Construction,” Journal of
on Decision Aids: A Model and an Experiment,” Journal of Consumer Research (24), March 1998, pp. 459-468.
Infor-mation Systems (12:2), 1998, pp. 75-94.
Brucks, M. “The Effects of Product Class Knowledge on Informa- Cunningham, M. “The Major Dimensions of Perceived Risk,” in Risk
tion Search Behavior,” Journal of Consumer Research (12:1), Taking and Information Handling in Consumer Behavior, D. Cox
1985, pp. 1-16. (ed.), Harvard University Press, Boston, 1967, pp. 82-108.
Davis, F. D. “Perceived Usefulness, Perceived Ease of Use, and
User Acceptance of Information Technology,” MIS Quarterly
(13:3), September 1989, pp. 319-340.
Davis, F. D., Bagozzi, R. P., and Warshaw, P. R. “User Acceptance of Fitzsimons, G. J., and Lehmann, D. R. “Reactance to Recommen-
Computer-Technology: A Comparison of Two Theoretical dations: When Unsolicited Advice Yields Contrary
Models,” Management Science (35:8), 1989, pp. 982-1003. Responses,” Marketing Science (23:1), 2004, pp. 82-94.
Davis, F. D., and Kottemann, J. E. “User Perceptions of Decision Formisano, R. A., Olshavsky, R. W., and Tapp, S. “Choice
Support Effectiveness: Two Production Planning Strategy in a Difficult Task Environment,” Journal of
Experiments,” Decision Sciences (25:1), 1994, pp. 57-78. Consumer Research (8:4), 1982, pp. 474-479.
*Dellaert, B. G. C., and Haubl, G. “Consumer Product Search Gefen, D., Karahanna, E., and Straub, D. W. “Trust and TAM in
with Personalized Recommendations,” unpublished working Online Shopping: An Integrated Model,” MIS Quarterly (27:1),
paper, Department of Marketing, Business Economics and 2003, pp. 51-90.
Law, University of Alberta, Edmonton, AB, 2005. *Gershoff, A. D., Mukherjee, A., and Mukhopadhyay, A. “Con-
DeLone, W., and McLean, E. “The DeLone and McLean Model of sumer Acceptance of Online Agent Advice: Extremity and
Information System Success: A Ten-Year Update,” Journal of Positivity Effects,” Journal of Consumer Psychology (13:1-2),
Management Information Systems (19:4), 2003, pp. 9-30. 2003, pp. 161-170.
DeLone, W., and McLean, E. “Information Systems Success: The Goldberg, M. E., and Hartwick, J. “The Effects of Advertiser
Quest for the Dependent Variable,” Information Systems Repu-tation and Extremity of Advertising Claim on
Research (3:1), 1992, pp. 60-95. Advertising Effec-tiveness,” Journal of Consumer Research
Detlor, B., and Arsenault, C. “Web Information Seeking and (17:2), 1990, pp. 172-179.
Retrieval in Digital Library Contexts: Towards an Intelligent Good, N., Schafer, J. B., Konstan, J. A., Borchers, A., Sarwar, B.
Agent Solution,” Online Information Review (26:6), 2002, pp. M., Herlocker, J. L., and Riedl, J. “Combining Collaborative
Filtering with Personal Agents for Better Recommendations,”
404-412.
in Proceedings of the 16th National Conference on Artificial
*Diehl, K. “Tempted by the Cheap and Easy: Searching Too Intelligence, Orlando, FL, July 1999, pp. 439-446.
Much in Ordered Environments,” unpublished working paper, Gregor, S., and Benbasat, I. “Explanations from Intelligent Sys-
Marketing Department, Moore School of Business, University tems: Theoretical Foundations and Implications for Practice,”
MIS Quarterly (23:4), 1999, pp. 497-530.
of South Carolina, Columbia, SC, 2003.
Grenci, R. T., and Todd, P. A. “Solutions-Driven Marketing,”
*Diehl, K., Kornish, L. J., and Lynch, J. G. “Smart Agents: When
Communications of the ACM (45:2), 2002, pp. 64-71.
Lower Search Costs for Quality Information Increase Price
Griffith, D. A., and Chen, Q. “The Influence of Virtual Direct
Sensitivity,” Journal of Consumer Research (30:1), 2003, pp.
Experience (VDE) on On-line Ad Message Effectiveness,”
56-71.
Journal of Advertising (33:1), 2004, pp. 55-68.
Doney, P. M., and Cannon, J. P. “An Examination of the Nature of
Hanani, U., Shapira, B. and Shoval, P. “Information Filtering:
Trust in Buyer-Seller Relationships,” Journal of Marketing
Overview of Issues, Research and Systems,” User Modeling
(61:2), 1997, pp. 35-51.
and User-Adapted Interaction (11:3), 2001, pp. 203-259.
Dowling, G. R., and Staelin, R. “A Model of Perceived Risk and
Haubl, G., and Dellaert, B. G. C. “A Behavioral Theory of Con-
Intended Risk-Handling Activity,” Journal of Consumer
sumer Product Search,” unpublished working paper,
Research (21:1), 1994, pp. 119-134. Department of Marketing, Business Economics and Law,
Edwards, W., and Fasolo, B. “Decision Technology,” Annual University of Alberta, Edmonton, AB, 2004.
Review of Psychology (52:1), 2001, pp. 581-606. *Haubl, G., and Murray, K. B. “Double Agent: Assessing the Role
Eierman, M. A., Niederman, F., and Adams, C. “DSS Theory: A of Electronic Product-Recommendation Systems,” Sloan Man-
Model of Constructs and Relationships,” Decision Support agement Review (47:3), 2006, pp. 8-12.
Sys-tems (14:1), 1995, pp. 1-26. *Haubl, G., and Murray, K. B. “Preference Construction and
Farhoomand, A. F., and Drury, D. H. “Managerial Information Persis-tence in Digital Marketplaces: The Role of Electronic
Overload,” Communications of the ACM (45:10), 2002, pp. Recom-mendation Agents,” Journal of Consumer Psychology
127-131. (13:1-2), 2003, pp. 75-91.
*Haubl, G., and Trifts, V. “Consumer Decision Making in Online
*Fasolo, B., McClelland, G. H., and Lange, K. A. “The Effect of Shopping Environments: The Effects of Interactive Decision
Site Design and Interattribute Correlations on Interactive Web- Aids,” Marketing Science (19:1), 2000, pp. 4-21.
Based Decisions,” in Online Consumer Psychology: Under- *Herlocker, J., Konstan, J. A., and Riedl, J. “Explaining Collab-
standing and Influencing Behavior in the Virtual World, C. P. orative Filtering Recommendations,” in Proceedings of the
Haughvedt, K. Machleit, and R. Yalch (eds.), Lawrence 2000 ACM Conference on Computer Supported Cooperative
Erlbaum Associates, Mahwah, NJ, 2005, pp. 325-344. Work Philadelphia, PA, December 2000, pp. 241-250.
*Felix, D., Niederberger, C., Steiger, P., and Stolze, M. “Feature- Herlocker, J., Konstan, J. A., Terveen, L. G., and Riedl, J. “Eval-
Oriented vs. Needs-Oriented Product Access for Non-expert uating Collaborative Filtering Recommender Systems. ACM
Online Shoppers,” in Towards the E-Society: E-Commerce, E- Transactions on Information Systems (22:1), 2004, pp. 5-53.
Business, and E-Government, B. Schmid, K. Stanoevska- *Hess, T. J., Fuller, M. A., and Mathew, J. “Involvement and
Slabeva, and V. TschammerZürich (eds.), Springer, New York, Decision-Making Performance with a Decision Aid: The Influ-
2001, pp. 399-406. ence of Social Multimedia, Gender, and Playfulness,” Journal
Fishbein, M., and Ajzen, I. Belief, Attitude, Intention and of Management Information Systems (22:3), 2005, pp. 15-54.
Behavior: An Introduction to Theory and Research, Addison-
Wesley, Reading, MA, 1975.

Hoeffler, S., and Ariely, D. “Constructing Stable Preferences: A King, M. F., and Hill, D. J. “Electronic Decision Aids: Integration
Look into Dimensions of Experience and Their Impact on of a Consumer Perspective,” Journal of Consumer Policy
Preference Stability,” Journal of Consumer Psychology (8:2), (17:2), 1994, pp. 181-206.
1999, pp. 113-139. Komiak, S. The Impact of Internalization and Familiarity on
*Hostler, R. E., Yoon, V. Y., and Guimaraes, T. “Assessing the Trust and Adoption of Recommendation Agents, unpublished
Impact of Internet Agent on End Users’ Performance,” Doctoral Dissertation, University of British Columbia,
Decision Support Systems (41:1), 2005, pp. 313-325.. Vancouver, BC, Canada, 2003.
Hoxmeier, J. A., and DiCesare, C. “System Response Time and *Komiak, S., and Benbasat, I. “The Effects of Personalization and
User Satisfaction: An Experimental Study of Browser-Based Familiarity on Trust and Adoption of Recommendation
Applications,” in Proceedings of the 6th Americas Conference Agents,” MIS Quarterly (30:4), 2006, pp. XXX-XXX.
on Information Systems, H. M. Chung (ed.), Long Beach, CA, Komiak, S., and Benbasat, I. “Understanding Customer Trust in
2000, pp. 140-145. Agent-Mediated Electronic Commerce, Web-Mediated Elec-
Jacoby, J. “Consumer Behavior: A Quadrennium.” Annual Review tronic Commerce, and Traditional Commerce,” Information
of Psychology (49:1), 1998, pp. 319-344. Technology and Management (5:1-2), 2004, pp. 181-207.
Jacoby, J., and Kaplan, L. B. “Components of Perceived Risk: A *Komiak, S., Wang, W., and Benbasat, I. “Comparing Customer
Cross-Validation,” Journal of Applied Psychology and Trust in Virtual Salespersons with Customer Trust in Human
Marketing (59:3), 1974, pp. 287-291. Salespersons,” in Proceedings of the 38th Annual Hawaii
Jahng, J., Jain, H., and Ramamurthy, K. “Empirical Investigation into International Conference on System Sciences, Big Island, HI,
Impact of Electronic Commerce Systems Richness on User January 2005, p. 175a.
Behavior: The Case of a Complex Product,” in Proceedings of the Kottemann, J. E., and Davis, F. D. “Computer-Assisted Decision
6th Americas Conference on Information Systems, H. M. Chung Making: Performance, Beliefs, and the Illusion of Control,”
(ed.), Long Beach, CA, 2000, pp. 1393-1397. Organizational Behavior and Human Decision Processes
Jiang, Z., and Benbasat, I. “Virtual Product Experience: Effects of (57:1), 1994, pp. 26-37.
Visual and Functional Control of Products on Perceived *Kramer, T. “The Effect of Measurement Task Transparency on
Diagnos-ticity and Flow in Electronic Shopping,” Journal of Preference Construction and Evaluations of Personalized Recom-
Management Information Systems (21:3), 2005, pp. 111-148. mendations,” Journal of Marketing Research, 2007, forthcoming.
Jiang, Z., Wang, W., and Benbasat, I. “Multimedia-Based Inter-active Langer, E. J. “The Illusion of Control,” Journal of Personality
Advising Technology for Online Consumer Decision Support,” and Social Psychology (32:2), 1975, pp. 311-328.
Communications of the ACM (48:9), 2005, pp. 92-98. Levin, D. Z., Cross, R., and Abrams, L. C. “Why Should I Trust You?
Kahn, B. E., and Baron, J. “An Exploratory Study of Choice Predictors of Interpersonal Trust in a Knowledge Transfer
Rules Favored for High-Stakes Decisions,” Journal of Context,” paper presented at the Academy of Management Annual
Consumer Psychology (4:4), 1995, pp. 305-328. Meeting, Denver, CO, August 9-14, 2002.
*Kamis, A., and Davern, M. J. “Personalizing to Product Category Li, H., Daugherty, T., and Biocca, F. “The Role of Virtual Experi-
Knowledge: Exploring the Mediating Effect of Shopping Tools ence in Consumer Learning,” Journal of Consumer
on Decision Confidence,” in Proceedings of the 37th Annual Psychology (13:4), 2003, pp. 395-407.
Hawaii International Conference on System Sciences, Big Lichtenthal, J. D., and Tellefsen, T. “Toward a Theory of Buyer–
Island, HI, January 2004, p. 70202a. Seller Similarity,” Paper No. 22-1999, Institute for the Study of
Keller, K. L., and Staelin, R. “Effects of Quality and Quantity of Business Markets, The Pennsylvania State University,
Information on Decision Effectiveness,” Journal of Consumer University Park, PA, 1999.
Research (14:2), 1987, pp. 200-214.
Kim, D. J., and Benbasat, I. “The Effects of Trust-Assuring Argu- Limayem, M., Cheung, C. M. K., and Chan, G. W. W. “Explaining
ments on Consumer Trust in Internet Stores,” in Proceedings Information Systems Adoption and Post-adoption: Toward an
of the 24th International Conference on Information Systems, S. Integrative Model,” in Proceedings of the 24th International
T. March, A. Massey, and J. I. DeGross (eds.), Seattle, WA, Conference on Information Systems, S. T. March, A. Massey, and
2003, pp. 767-773. J. I. DeGross (eds.), Seattle, WA, 2003, pp. 720-731.
Kim, S. S., and Malhotra, N. K. “A Longitudinal Model of Lirtzman, S. I., and Shuv-Ami, A. “Credibility of Sources of
Continued IS Use: An Integrative View of Four Mechanisms Com-munication on Products’ Safety Hazards,” Psychological
Underlying Postadoption Phenomena,” Management Science Reports (58:3), 1986, pp. 707-718.
(51:5), 2005a, pp. 741-755. Luwicki, R. J., and Bunker, B. B. “Trust in Relationships: A
Model of Trust Development and Decline,” in Conflict,
Kim, S. S., and Malhotra, N. K. “Predicting System Usage from Cooperation and Justice, B. B. Bunker and J. Z. Rubin (eds.),
Intention and Past Use: Scale Issues in the Predictors,” Jossey-Bass, San Francisco, 1995, pp. 133-173.
Decision Sciences (36:1), 2005b, pp. 187-196. Lynch, J. G., and Ariely, D. “Wine Online: Search Costs Affect
King, M. F., and Balasubramanian, S. K. “The Effects of Competition on Price, Quality, and Distribution,” Marketing
Expertise, End Goal, and Product Type on Adoption of Science (19:1), 2000, pp. 83-103.
Preference Forma-tion Strategy,” Journal of the Academy of Maes, P. “Agents that Reduce Work and Information Overload,”
Marketing Science (22:2), 1994, pp. 146-159. Communications of the ACM (37:7), 1994, pp. 31-40.

Maes, P., Guttman, R. H., and Moukas, A. G. “Agents that Buy and *Olson, E. L., and Widing, R. E. “Are Interactive Decision Aids
Sell,” Communications of the ACM (42:3), 1999, pp. 81-91. Better than Passive Decision Aids? A Comparison with Impli-
Mak, B., and Lyytinen, K. “A Model to Assess the Behavioral Impacts cations for Information Providers on the Internet,” Journal of
of Consultative Knowledge Based Systems,” Information Interactive Marketing (16:2), 2002, pp. 22-33.
Processing & Management (33:4), 1997, pp. 539-550. Olson, J., and Dover, P. A. “Disconfirmation of Consumer Expec-
Mallach, E. G. Decision Support and Data Warehouse Systems: tations through Trial,” Journal of Applied Psychology (64:2),
Irwin McGraw-Hill, Boston, 2000. 1979, pp. 179-189.
Mandel, N., and Johnson, E. “Constructing Preferences Online: Parikh, M. A., and Fazlollahi, B. “An Empirical Examination of
Can Web Pages Change What You Want?,” unpublished the User Satisfaction/System Effectiveness Interrelation,” in
working paper, Marketing Department, The Wharton School, Pro-ceedings of the 8th Americas Conference on Information
University of Pennsylvania, Philadelphia, PA,1998. Systems, R. Ramsower and J. Windsor (eds.), Dallas, TX, 2002,
Massa, P., and Bhattacharjee, B. “Using Trust in Recommender pp. 208-215.
Systems: An Experimental Analysis,” paper presented at the Parthasarathy, M. B., and Bhattacherjee, A. “Understanding Post-
Second International Conference on Trust Management, Adoption Behavior in the Context of Online Services,”
Oxford, United Kingdom, March 29-April 1, 2004. Information Systems Research (9:4), 1998, pp. 362-379.
McKinney, V., Yoon, K., and Zahedi, F. M. “The Measurement of Patrick, A. S. “Building Trustworthy Software Agents,” IEEE
Web-Customer Satisfaction: An Expectation and Disconfirma- Internet Computing (6:6), 2002, pp. 46-53.
tion Approach,” Information Systems Research (13:3), 2002, Patton, P. “Buy Here, and We’ll Tell You What You Like,” The
pp. 296-315. New York Times on the Web, September 22, 1999.
McKnight, D. H., Cummings, L. L., and Chervany, N. L. “Initial Pavlou, P. A. “Consumer Acceptance of Electronic Commerce:
Trust Formation in New Organizational Relationships,” Integrating Trust and Risk with the Technology Acceptance
Academy of Management Review (23:3), 1998, pp. pp. 473. Model,” International Journal of Electronic Commerce (7:3),
*McNee, S., Lam, S. K., Konstan, J. A., and Riedl, J. “Interfaces 2003, pp. 101-134.
for Eliciting New User Preferences in Recommender Systems,” Payne, J. W. “Contingent Decision Behavior,” Psychological
in Proceedings of the 9th International Conference on User Bulletin (92:2), 1982, pp. 382-402.
Modeling (User Modeling 2003), Johnston, PA, June 2003, pp. Payne, J. W., Bettman, J. R., and Johnson, E. The Adaptive Deci-
178-187. sion Maker, Cambridge University Press, New York, 1993.
Mohr, L. A., and Bitner, M. J. “The Role of Employee Effort in Payne, J. W., Bettman, J. R., and Johnson, E. “Adaptive Strategy
Satisfaction with Service Transactions,” Journal of Business Selection in Decision Making,” Journal of Experimental
Research (32:3), 1995, pp. 239-252. Psych-ology: Learning, Memory, and Cognition (14:3), 1988,
Mondak, J. J. “Perceived Legitimacy of Supreme Court pp. 534-552.
Decisions: Three Functions of Source Credibility,” Political Payne, J. W., Bettman, J. R., and Johnson, E. “Behavioral
Behavior (2:4), 1990, pp. 363-384. Decision Research: A Constructive Processing Perspective,”
Montaner, M., Lopez, B., and De La Rosa, J. L. “A Taxonomy of Annual Review of Psychology (43:1), 1992, pp. 87-131.
Recommender Agents on the Internet,” Artificial Intelligence *Pedersen, P. “Behavioral Effects of Using Software Agents for
Review (19:4), 2003, pp. 285-330. Product and Merchant Brokering: An Experimental Study of
Montgomery, A. L., Hosanagar, K., Krishnan, R., and Clay, K. B. Consumer Decision Making,” International Journal of
“Designing a Better Shopbot,” Management Science (50:2), Electronic Commerce (5:1), 2000, pp. 125-141.
2004, pp. 189-206. *Pereira, R. E. “Influence of Query-Based Decision Aids on
*Moore, R., and Punj, G. “An Investigation of Agent Assisted Consumer Decision Making in Electronic Commerce,” Infor-
Con-sumer Information Search: Are Consumers Better Off?,” mation Resources Management Journal (14:1), 2001, pp. 31-
Advances in Consumer Research (28:1), 2001, p. 128. 48.
Nah, F. F., and Benbasat, I. “Knowledge-Based Support in a *Pereira, R. E. “Optimizing Human-Computer Interaction for the
Group Decision Making Context: An Expert-Novice Electronic Commerce Environment,” Journal of Electronic
Comparison,” Journal of the Association for Information Commerce Research (1:1), 2000, pp. 23-44.
Systems (5:3), 2004, pp. 125-150. Pornpitakpan, C. “The Persuasiveness of Source Credibility: A
Nowlis, S. M., and Simonson, I. “Attribute-Task Compatibility as Critical Review of Five Decades’ Evidence,” Journal of
a Determinant of Consumer Preference Reversals,” Journal of Applied Social Psychology (34:2), 2004, pp. 243-281.
Marketing Research (34), May 1997, pp. 205-218. Punj, G., and Rapp, A. “Influence of Electronic Decision Aids on
O’Keefe, R. M., and McEachern, T. “Web-Based Customer Consumer Shopping in Online Stores,” paper presented at the
Decision Support Systems,” Communications of the ACM Home Oriented Informatics and Telematics 2003 (HOIT 2003)
(41:3), 1998, pp. 71-78. Conference, Irvine, CA, April 6-8, 2003.
Qiu, L. “Designing Social Interactions with Animated Face and
Olshavsky, R. W. “Perceived Quality in Consumer Decision- Speech Output for Product Recommendation Agents in Elec-
Making: An Integrated Theoretical Perspective,” in Perceived tronic Commerce,” unpublished paper (Doctoral Dissertation
Quality: How Consumers View Stores and Merchandise, J. C. proposal), University of British Columbia, Vancouver, 2005.
Olson (ed.), D. C. Heath and Company, Lexington, MA, 1985, Qiu, L., and Benbasat, I. “An Investigation into the Effects of
pp. 3-29. Text-to-Speech Voice and 3D Avatars on the Perception of

Presence and Flow of Live Help in Electronic Commerce,” Silver, M. S. “User Perceptions of Decision Support System
ACM Transactions on Computer-Human Interaction (12:4), Restrictiveness: An Experiment,” Journal of Management
2005, pp. 329-355. Information Systems (5:1), 1988, pp. 51-65.
Randall, T., Christian, T., and Ulrich, K. T. “User Design of Cus- Simon, H. A. “A Behavioral Model of Rational Choice,” Quarterly
tomized Products,” unpublished seminar paper, Operations and Journal of Economics (69), February 1955, pp. 99-118.
Information Management Department. The Wharton School, *Sinha, R., and Swearingen, K. “Comparing Recommendations
University of Pennsylvania, Philadelphia, Philadelphia, PA, Made by Online Systems and Friends,” in Proceedings of the
2003. 2nd DELOS Network of Excellence Workshop on
Rogers, E. M. Diffusion of Innovations (4th ed.), The Free Press, Personalisation and Recommender Systems in Digital
New York, 1995. Libraries, Dublin, Ireland, June 18-20, 2001.
Sarwar, B. M., Karypis, G., Konstan, J. A., and Riedl, J. “Analysis of *Sinha, R., and Swearingen, K. “The Role of Transparency in
Recommendation Algorithms for E-Commerce,” in Pro-ceedings Recommender Systems,” in CHI 2002 Extended Abstracts on
of the 2nd ACM Conference on Electronic Commerce, Minneapolis, Human Factors in Computing Systems, Minneapolis, MN,
MN, October 17-20, 2000, pp. 158-167. April 20-25, 2002, pp. 830-831.
Schafer, J. B., Konstan, J. A., and Riedl, J. “E-Commerce Recom-
mendation Applications,” Data Mining and Knowledge Dis- Smith, S., and Shaffer, D. R. “Celerity and Cajolery: Rapid
covery (5:1-2), 2001, pp. 115-153. Speech May Promote or Inhibit Persuasion Through its Impact
Schafer, J. B., Konstan, J. A., and Riedl, J. “Meta- on Message Elaboration,” Personality and Social Psychology
Recommendation Systems: User-Controlled Integration of Bulletin (7:6), 1991, pp. 663-669.
*Spiekermann, S. Online Information Search with Electronic
Diverse Recommen-dations,” paper presented at the 11 th
Agents: Drivers, Impediments, and Privacy Issues, unpublished
International Conference on Information and Knowledge
doctoral dissertation, Humboldt University Berlin, Berlin,
Management, McLean, VA, November 2002.
2001.
*Schein, A. I., Popescul, A., Ungar, L. H., and Pennock, D. M.
Spiekermann, S., and Paraschiv, C. “Motivating Human-Agent
“Methods and Metrics for Cold-Start Recommendations,” in
Interaction: Transferring Insights from Behavioral Marketing
Proceedings of the 25th Annual International ACM Special to Interface Design,” Electronic Commerce Research (2:3),
Interest Group on Information Retrieval (SIGIR) Conference 2002, pp. 255-285.
on Research and Development in Information Retrieval,
Tampere, Finland, August 2002, pp. 253-257. Stamm, K., and Dube, R. “The Relationship of Attitudinal Com-
Schleifer, L. M., and Amick, B. C. “System Repines Time and ponents to Trust in Media,” Communications Research (2:1),
Method of Pay: Stress Effects in Computer-Based Tasks,” 1994, pp. 105-123.
International Journal of Human-Computer Interaction (1:1), Stolze, M., and Nart, F. “Well-Integrated Needs-Oriented Recom-
1989, pp. 23-39. mender Components regarded as Helpful,” in CHI 2004
Sears, A., and Borella, M. S. “WANDS: Tools for Designing and Extended Abstracts on Human Factors in Computing Systems,
Testing Distributed Documents,” Technical Report No. #97-01, Vienna, Austria, April 24-29, 2004, p. 1571.
School of Computer Science, DePaul University, 1997. Suh, K. S., and Lee, Y. E. “Effects of Virtual Reality on Consumer
Selnes, F. “Antecedents and Consequences of Trust and Learning: An Empirical Investigation in Web-Based Electronic
Satisfaction in Buyer-Seller Relationships,” European Journal Commerce,” MIS Quarterly (29:4), 2005, pp. 673-697.
of Marketing (32:3-4), 1998, pp. 305-322. *Swaminathan, V. “The Impact of Recommendation Agents on
*Senecal, S. Essays on the Influence of Online Relevant Others Consumer Evaluation and Choice: The Moderating Role of Cate-
on Consumers’ Online Product Choices, unpublished doctoral gory Risk, Product Complexity, and Consumer Knowledge,”
dissertation, HEC Montreal, University of Montreal, 2003. Journal of Consumer Psychology (13:1-2), 2003, pp. 93-101.
*Senecal, S., and Nantel, J. “The Influence of Online Product *Swearingen, K., and Sinha, R. “Beyond Algorithms: An HCI
Recommendations on Consumers’ Online Choices,” Journal of Perspective on Recommender Systems,” paper presented at the
Retailing (80:2), 2004, pp. 159-169. ACM SIGIR Workshop on Recommender Systems, New
Shneiderman, B. Designing the User Interface: Strategies for Orleans, LA, September 13, 2001.
Effective Human-Computer Interaction (3rd ed.). Addison- *Swearingen, K., and Sinha, R. “Interaction Design for Recom-
Wesley, Reading, MA, 1998. mender Systems,” paper presented at the International Con-
Shneiderman, B. “Direct Manipulation for Comprehensible, Pre- ference on Designing Interactive Systems (DIS 2002), London,
dictable and Controllable User Interfaces,” in Proceedings of June 25-28, 2002.
the 2nd International Conference on Intelligent User Interfaces, Taylor, S., and Todd, P. “Understanding Information Technology
Orlando, FL, January 6-9, 1997, pp. 33-39. Usage: A Test of Competing Models,” Information Systems
Shugan, S. M. “The Cost of Thinking,” Journal of Consumer Research (6:2), 1995, pp. 144-176.
Research (7:2), 1980, pp. 99-111. Thum, M., Boucstein, W., Kuhman, W., and Ray, W. J. “Stan-dardized
Silver, M. S. “Decision Support Systems: Directed and Non- Task Strain and System Response-Times in Human-Computer
directed Change,” Information Systems Research (1:1), 1990, Interaction,” Ergonomics (38:7), 1995, pp. 1342-1351.
pp. 47-70. Todd, P., and Benbasat, I. “The Influence of Decision Aids on
Choice Strategies: An Experimental Analysis of the Role of
Cognitive Effort,” Organizational Behavior and Human
Decision Processes (60:1), 1994, pp. 36-74.
Tversky, A., Sattath, S., and Slovic, P. “Contingent Weighting in West, P. M., Ariely, D., Bellman, S., Bradlow, E., Huber, J.,
Judgment and Choice,” Psychological Review (95:3), 1988, pp. Johnson, E., Kahn, B., Little, J., and Schkade, D. “Agents to
371-384. the Rescue?,” Marketing Letters (10:3), 1999, pp. 285-300.
*Urban, G. L., Sultan, F., and Qualls, W. “Design and Evaluation Wixom, B. H., and Todd, P. “A Theoretical Integration of User
of a Trust Based Advisor on the Internet,” unpublished Satisfaction and Technology Acceptance,” Information Systems
Working Paper 123, Sloan School of Management, Research (16:1), 2005, pp. 85-102.
Massachusetts Institute of Technology, 1999. Zachary, W. “A Cognitively Based Functional Taxonomy of Deci-
*van der Heijden, H. V. D., and Sorensen, J. B. “The Mobile sion Support Techniques,” Human-Computer Interaction (2:1),
Decision Maker: Mobile Decision Aids, Task Complexity, and 1986, pp. 25-63.
Decision Effectiveness,” unpublished working paper 2002-25, Zhang, Y. Literature Review for Online Communities and Recom-
Copenhagen Business School, Frederiksberg, Denmark, 2002. mender Systems: State of the Art Paper, unpublished doctoral
Van Slyke, C., and Collins, R. “Trust Between Users and Their dissertation, University of Minnesota, Minneapolis, 2002.
Intelligent Agents,” paper presented at the Annual Meeting of the Zhu, L. B. I. Let’s Shop Online Together: An Investigation of
Decision Sciences Institute, Orlando, FL, November 24-26, 1996. Online Collaborative Shopping Support, unpublished Master’s
Venkatesh, V. “Determinants of Perceived Ease of Use: thesis, University of British Columbia, Vancouver, 2004.
Integrating Control, Intrinsic Motivation, and Emotion into the Zucker, L. G. “Production of Trust: Institutional Sources of
Technology Acceptance Model,” Information Systems Economic Structure, 1840-1920,” in Research in
Research (11:4), 2000, pp. 342-365. Organizational Behavior, B. M. Staw and L. L. Cummings
Venkatesh, V., and Brown, S. A. “A Longitudinal Investigation of (eds.), JAI Press, Greenwich, CT, 1986, pp. 53-111.
Personal Computers in Homes: Adoption Determinants and Zucker, L. G., Darby, M. R., Brewer, M. B., and Peng, Y.
Emerging Challenges,” MIS Quarterly (25:1), 2001, pp. 71-102. “Collaboration Structure and Information Dilemmas in Bio-
Venkatesh, V., Morris, M., Davis, G. B., and Davis, F. D. “User technology: Organizational Boundaries as Trust Production,”
Acceptance of Information Technology: Toward a Unified in Trust in Organizations: Frontiers of Theory and Research,
View,” MIS Quarterly (27:3), 2003, pp. 425-478. R. M. Kramer and T. R. Tyler (eds.), Sage Publications,
Verplanken, B. “Persuasive Communication of Risk Information: Thousand Oaks, CA, 1996, pp. 90-113.
A Test of Cue Versus Message Processing Effects in a Field
Experiment,” Personality and Social Psychology Bulletin (7:2),
1991, pp. 188-193. About the Authors
*Vijayasarathy, L. R., and Jones, J. M. “Do Internet Shopping
Aids Make a Difference? An Empirical Investigation,” Bo Xiao is a Ph.D. student in Management Information Systems
Electronic Markets (11:1), 2001, pp. 75-83. at the Sauder School of Business, University of British Columbia,
*Wang, W. Design of Trustworthy Online Recommendation Vancouver, Canada. Her research focuses on intelligent decision
Agents: Explanation Facilities and Decision Strategy Support, support systems, with special interest in electronic commerce
unpub-lished Doctoral Dissertation, University of British product recommendation agents.
Columbia, Vancouver, 2005.
*Wang, W., and Benbasat, I. “Impact of Explanations on Trust in Izak Benbasat, Fellow-Royal Society of Canada, is CANADA
Online Recommendation Agents,” Working Paper No. 02-MIS- Research Chair in Information Technology Management at the
002, MIS Division, Sauder School of Business, University of Sauder School of Business, University of British Columbia,
British Columbia, Vancouver, 2004a. Vancouver, Canada. He currently serves on the editorial boards of
*Wang, W., and Benbasat, I. “Trust and Distrust Building Journal of the Association for Information Systems and Journal of
Processes in Online Recommendation Agents: A Process- Management Information Systems. He was editor-in-chief of
Tracing Analy-sis,” unpublished Working Paper 04-MIS-003, Infor-mation Systems Research, editor of the Information Systems
MIS Division, Sauder School of Business, University of British and Decision Support Systems Department of Management
Columbia, Vancouver, 2004b. Science, and a senior editor of MIS Quarterly. The general theme
*Wang, W., and Benbasat, I. “Trust In and Adoption of Online of his research is improving the communication between
Recommendation Agents,” Journal of the AIS (6:3), 2005, pp. information technology (IT), management, and IT users. This
72-100. communication exits at different levels of the organization and has
Weiss, S. M., Boggs, G., Lehto, M., Hodja, S., and Martin, D. J. different facets, for example, interaction between individual
“Computer System Response Time and Psychophysiological managers and intelligent decision support systems, design of
Stress,” in Proceedings of the Human Factors Society 26 th interfaces for customers shopping at web stores, and dialogues
Annual Meeting, Seattle, WA, October 1982, pp. 698-702. between IT professionals and line managers.

Appendix A
Summary of Empirical Studies
Aggarwal and Vaidyanathan (2003a)

Context • Computer aided experiment with 109 student subjects
Independent Variables • Product type (search/experience)
• Recommendation development process (rule-based filtering vs. collaborative filtering)
Dependent VariablesPerceived agent effectiveness:
• Perceived quality of recommendations
• Satisfaction with the recommendations
• Intent to follow-up on a recommendation
Results • Agents were more effective for search goods than for experience goods.
• Rule-based filtering process was perceived to be more effective than collaborative-based filtering process.
Aggarwal and Vaidyanathan (2003b)

Context • A study with 42 subjects
• Two preference elicitation tasks for refrigerators
Independent Variables • Preference elicitation methods (conjoint based inference vs. self-explicated ratings)
Dependent Variables • Convergence of ratings
Results • Inferred preferences varied significantly from stated inferences.
• Implications: the two methods are not equivalent and using one method over the other may result in a recom-
mendation that does not match the preferences of the consumer.
Aksoy and Bloom (2001)

Theory Theories of human decision-making
Context • RA embedded in a simulated fictitious online shopping site
• Lab experiment with 172 subjects
• A search and choice task for a cellular phone
Independent Variables • Degree of similarity between attribute weights used by the agent to generate the listing and the consumers’ own
weights
• Degree of similarity between the decision strategy used by the agent and the consumers’ own decision strategies
Dependent Variables • Perceived utility
• Cognitive cost
• Amount of information searched
• Decision quality
Results • Attribute weight similarity was important in increasing perceived utility of using ordered listings, decreasing the
cognitive cost and amount of information searched, and improving decision quality.
• Subject’s perceptions of correspondences between the decision strategy used by an RA and the subject’s preferred
decision strategy moderated the effect of decision strategy similarity manipulations on the dependent variables.
Basartan (2001)
Theory Analytical model of consumer utility by Montgomery et al. (2001)
Context • Simulated shopbots
• Experiment with 190 student subjects
• The task was to complete an exercise to estimate utility models, shop at several shopbots, and evaluate preferences
for shopping at the presented shopbot or Amazon
Independent Variables • Number of alternatives displayed
• Response time
Dependent Variables • User preference
Results • Shopper preference for shopbots declined with too many alternatives and long waiting times.

Bechwati and Xia (2003)

Theory • Effort-accuracy model
• Equity theory
• Expectation-performance paradigm
Context • Electronic aid for job search
• Two studies with 180 and 52 student subjects, respectively
• In Study 1, subjects completed a paper and pencil task, responding to the questionnaire based on imagination of
a scenario described in a booklet
• In Study 2, subjects performed a simulated job search with a simple web-based decision aid
Independent Variables Study 1:
• Type of aid (no aid, human aid, electronic aid)
Study 2:
• Information about search progress
• Customization of results
Dependent Variables Study 1:
• Perception of own effort
• Perception of aid’s effort
• Perception of effort saving
Study 2:
• Satisfaction with the process
• Perception of effort saving
Results • Study 1: Consumers believed that electronic decision aids saved them an equal level of effort. Consumers,
however, perceived electronic aids as exerting less effort than human aids.
• Study 2: Online shoppers’ satisfaction with decision process was positively associated with their perception of
the effort saved for them by electronic aids. Moreover, informing shoppers about search progress led to a
higher level of perceived saved effort and, consequently, satisfaction.
Bharati and Chaudhury (2004)

Theory IS success model
Context • A survey with 210 subjects
• Different web-based decision support systems
Independent Variables • System quality
• Information quality
• Information presentation
Dependent Variables • Decision making satisfaction
" Decision confidence
" Decision effectiveness
Results • System quality and information quality were directly and positively correlated with decision-making satisfaction.
• The relative weight of information quality was higher than system quality.
• Presentation was not directly and positively correlated with decision-making satisfaction.
Cooke, Sujan, Sujan, and Weitz (2002)

Theory Importance of context
Context • Three studies with 179, 118, and 65 student subjects respectively
• Simulated music CD shopping agents
• Subjects were asked to rate the agent as well as to indicate their likelihood of buying each of the unfamiliar
CDs Independent Variables • Number of recommended items
• Familiarity of the recommended items
• Preference (well-liked or moderately liked)
• Context
" recommendations
" Simultaneous condition
" Sequential condition

• Item-specific information
" With music clips
" Without music clips
• Similarity of information provided for the familiar and unfamiliar recommendations
" Endorsed by the same reviewer
" Endorsed by different reviewers
Dependent Variables • Agent competence
• Agent usefulness
• Likelihood of buying the unfamiliar CDs
Results • Unfamiliar recommendations lowered agent evaluations.
• Additional recommendations of familiar products served as a context within which unfamiliar
recommendations were evaluated.
• When the presentation of the recommendations made unfamiliar and familiar products appear similar,
evaluative assimilation resulted.
• When additional information about the unfamiliar products was given, consumers distinguished them from the
familiar products, producing evaluative contrast.
Cosley, Lam, Albert, Konstan, and Riedl (2003)

Theory Literature on conformity and persuasive computing
Context • Three experiments with 536 users
• Movie-Lens movie recommender system
• Users were asked to rate a set of movies
Independent Variables • Types of rating scales
• Whether or not predictions are shown when users rate movies
Dependent Variables • Accuracy of CF predictions
• Users’ ratings
Results • Users preferred finer-grained scales; however, granularity was not the only factor.
• Users rated fairly consistently across rating scales.
• The display of predicted ratings on unrated movies led users to rate in the direction of the prediction.
• Users could detect systems that manipulated predictions.
Dellaert and Haubl (2005)

Theory • Normative search theory
• Research in behavioral decision making
Context • Lab experiment with 455 subjects
• RAs for compact stereo systems and home rentals
• A preference-elicitation task followed by a product choice task
Independent Variables • The availability of personalized product recommendations
• Expected increase in utility that the consumer derives from looking at the next alternative in the
recommendation list
• The standard error in the prediction of the expected utility of the next alternative in the list
• Utility difference between current and most preferred prior alternative
• Attribute-based difference between current and most preferred prior alternative
• Utility difference between current alternative and the one inspected just prior to it
Dependent Variables • The probability of continuing to search
• Currently most preferred alternative
Results • Significant positive effect of the expected difference in utility between the next alternative in the personalized list
and the currently most preferred alternative on the probability of continuing to search.
• Significant positive effect of the utility difference between the currently inspected product and the best previously
encountered one on the choice of the currently most preferred alternative.
• Personalized product recommendations increased consumers’ tendency to rely on local utility comparisons in
deciding which alternative was their currently most preferred one.
• Personalized product recommendations reduced consumers’ tendency to rely on a comparison of the utility of the
current alternative and that of the best previously inspected one in determining the currently most preferred
alternative.

Diehl (2003)
Theory Research on consumer decision making
Context • Three lab experiments with 51, 47, and 100 subjects, respectively
• RAs for greeting cards and MP3 players
• All three experiments are based on principal-agent tasks. Subjects were asked to choose an alternative that would
be liked by a target customer
Independent Variables Experiment 1
• Search cost
• Type of recipient
Experiment 2
• Number of recommendations presented
• Type of recipient
Experiment 3
• Search costs
• Accuracy (high accuracy goal vs. low accuracy goal)
Dependent Variables Experiment 1
• Amount of search
• Quality of consideration set
• Quality of the chosen card
• Selectivity
Experiment 2
• Selectivity
Experiment 3
• Size of consideration set
Results Experiment 1
• Lower search costs significantly increased the number of unique options searched, decreased the quality of the
consideration set, led to worse choices, and reduced selectivity.
Experiment 2
• Recommending more cards significantly increased the number of unique options searched, decreased the quality
of the consideration set, led to worse choices, and reduced selectivity.
Experiment 3
• The negative effects of lower search costs were heightened if consumers had a greater motivation to be accurate.
Diehl, Kornish, and Lynch (2003)

Theory Research on search costs and price sensitivity
Context • Three lab experiments with 64, 43, and36 subjects, respectively
• RA for greeting cards
• Task was to choose greeting cards for two specifically described recipients
• Type of search agent (ordered vs. random)
• Assortment size
• Order of recipient
Experiment 2
• Sequence of search
• Order of recipient
• Trial (1 vs. 2)
Experiment 3
• Relative importance of price in the reward function
• Type of search agent
• Order of search


• Price of the chosen card
Experiment 2
Experiment 3
Results • With a good ordering agent, consumers had better decision quality and paid lower price when the underlying
assortment was larger.
• Consumers paid lower prices when recommendations were ordered than when they were not.
• Repeated use of the ordered search agent decreased prices more.
• Ordered search agent led to higher/lower prices being chosen when price was relatively less/more important.
Fasolo, McClelland, and Lange (2005)

Theory Effort-accuracy framework
Context • Web-based experiment with 60 subjects
• RA for digital cameras
• Task was to make four distinct choices, using four different decision sites
Independent Variables • Interattribute correlations
• Site design (Compensatory vs. noncompensatory)
Dependent Variables • Clicks
" Total clicks
" Option clicks
" Attribute clicks
• Choice quality
• Satisfaction with the choices made
• Confidence in the choices made
• Difficulty of the choice tasks
• Ease of use of the decision site
• Satisfaction with the decision site used
Results • Site design influenced users’ choice behavior
" Compensatory site: more option clicks
" Noncompensatory site: more attribute clicks
• Interattribute correlation also affected choice behavior
" More total clicks when the correlation was negative than when it was positive
• The interaction between interattribute correlation and site design
" Compensatory site (Negative correlation: more option clicks)
" Noncompensatory site (Negative correlation: more attribute clicks)
• Site design and interattribute correlation affected users’ psychological perceptions
" More positive perceptions were associated with the compensatory site than with the noncompensatory site
" More positive perceptions were choices characterized by positive rather than negative interattribute correlations
" The real difference between the two designs emerged when attributes were negatively related
" The compensatory site enabled more choices that were of high quality
Felix, Niederberger, Steiger, and Stolze (2001)

Context • Digital camera RA
• The task was to engage in two RA-assisted shopping sessions for digital cameras (one with the assistance of a
feature-based RA and the other with a needs-based RA)
Independent Variables • Type of RA (feature-based vs. needs-based)
• Order (which type of RA is used first)
Dependent Variables • Preference for the RA
• Perceived suitability of the RA’s advice to product novices

Xiao & Benbasat/E-Commerce Product Recommendation
Agents
Results • Most subjects recommended needs-based RAs for novices.

• No significant difference was found for user preferences between the two types of RAs.
• Self-reported preferences did not match the observations of the experimenter—some novices considered
themselves experts.
Gershoff and Mukhopadhyay (2003)

Theory • Goal-based emotion
• Social categorization
• Anchoring and adjustment
• Correspondence judgments
• Extremity effect (negativity and positivity effects)
Context • Agent: an internet-based movie critic (movie rating service)
• Two lab experiments with 85 and 43 student subjects, respectively
• The task was for the subjects to examine their own movie ratings with those provided by the online movie critic
and then evaluate the likelihood of accepting the agent’s advice
Independent Variables Study 1:
• Overall agreement (high/low)
• Extreme agreement (high/low)
Study 2:
• Extreme agreement (positive/negative)
• Advice valence (positive/negative)
Dependent Variables Study 1:
• Likelihood of accepting the agent’s advice
Study 2:
• Acceptance of agent advice
• Confidence in agent advice
• Perceived similarity of attributes likes and dislikes
Results • Study 1: In addition to the overall agreement rate, consumers paid special attention to extreme opinion agreement
when assessing agent diagnosticity (i.e., extremity effect).
• Study 2: Positive extreme agreement was more influential than negative extreme agreement when advice valence
was positive, but the converse did not hold when advice valence was negative (i.e., positivity effect).
Haubl and Murray (2003)

Theory Constructive preferences
Context • Online attribute-based RA
• Lab experiment with 347 student subjects
• Three tasks: one agent assisted shopping task for backpacking tents and two paired choice tasks
Independent Variables • Attribute inclusion in RA calibration interface
• Inter-attribute correlation (negative, positive)
• Perceived rationale for attribute inclusion (strong, neutral, weak)
Dependent Variables • Relative importance of the attributes (included in RA calibration) to the user when making decisions
• Amount of information searched
" Total amount of time spent searching
" Number of alternatives for which a detailed description is viewed
Results • The inclusion of an attribute in an RA rendered this attribute more important when consumers made product
choices.
• This preference-construction effect was moderated by the inter-attribute correlation and the perceived rationale
for the selective inclusion of attributes.
• This type of preference-construction effect persisted into subsequent choice tasks where no electronic decision aid
was present, and the extent of such persistence was greater if a stronger rationale for the selective inclusion of
attributes in an RA had been provided during an earlier shopping trip.
Haubl and Murray (2006)

Theory • Effort/accuracy trade-off
• Constructive preferences

Context • RA for notebooks

• Eight repeated shopping trips
Independent Variables • RA (with/without)
Dependent Variables • Decision effort
" Number of alternatives subjects looked at before making product choice across eight shopping trips
" Subjective utility score of the chosen product
Results • The presence of personalized product recommendations reduced search effort.
• Using an RA dramatically increased subjects’ decision quality.
Haubl and Trifts (2000)

Theory • Effort/accuracy trade-off
• Strengths and weaknesses of human decision makers in information processing
• Two stage process to decision making
Context • Custom built product-brokering RA
• Lab experiment with 249 student subjects
• The task was to shop for a product in each of two categories—backpacking tents and compact stereo systems
Independent Variales • RA (with/without)
• Comparison matrix (with/without)
• Product category
• Product category order
Dependent Variables • Amount of product information searched
• Size and quality of consideration set
" Whether non-dominated alternatives were selected
" Product switching
" Confidence in purchase decisions
Results • The use of RAs reduced search effort for product information, decreased the size but increased the quality of
consideration sets, and improved the quality of purchase decisions.
• The use of a comparison matrix (CM) led to a decrease in the size but an increase in the quality of consideration
sets, and it tended to have a favorable effect on objective decision quality.
• Product category and order did not have a moderating effect on the relationship between RA/CM and the
dependent variables.
Herlocker, Konstan, and Riedl (2000)

Theory Theory of explanation
Context • MovieLens online movie RA
• Two studies with 78 and 210 subjects, respectively
Independent Variables • Different techniques of providing explanation
• Explanation facilities (with, without)
Dependent Variables • How likely the users would go and see the movie.
• User acceptance of the collaborative filtering (CF) RA
• Filtering performance of user
Results • What models and techniques are effective in supporting explanation in an automated collaborative filtering (ACF)
system? (Study 1)
Big winners: histograms of neighbors’ ratings, past performance, similarity to other items in the user’s profile, and
favorite actor or actress.
• Can explanation facilities increase the acceptance of automated collaborative filtering systems? (Study 2)
Most users valued the explanations and would like to see them added to their ACF system.
• Can explanation facilities increase the filtering performance of ACF system users? (Study 2)
Unable to prove or disprove our hypothesis.
Users performed filtering based on many different channels of input.

Hess, Fuller, and Mathew (2005)

• Computer-based decision aid for apartments
• An apartment selection task
Independent Variables • Multimedia vividness (text only, text and voice, animation)
• Personality similarity (between user and decision aid)
• Gender
• Computer playfulness
Dependent Variables • Involvement with decision aid
• Decision making outcomes
" Satisfaction
" Understanding
" Decision time
" Use of decision aid features
" Decision quality
Results • Computer playfulness increased user involvement.
• Women were more involved with the decision aid than men.
• When the personalities of the user and the decision aid were more similar (lower difference scores), users were
more involved with the decision aid.
• The vividness of the multimedia did significantly affect involvement, but not in the hypothesized direction. The
addition of animation appeared to reduce user involvement with the decision aid.
• Involvement positively affected user satisfaction and understanding with the decision aid.
• Users that were more involved with the decision aid also spent more time using the decision aid.
• Involvement did not significantly affect the number of decision aid features used or decision quality/accuracy.
Hostler, Yoon, and Guimaraes (2005)

• Shopbot for DVD
• DVD shopping task
Independent Variables • Use of shopbot
Dependent Variables • End-user performance
" Time spent
" Decision quality
" Confidence in the decision
" Cognitive effort
Results • Using the shopbot saved users time and increased their decision quality. There were no significant differences in
either the subjects’ decision confidence or their perception of the mental effort required to perform this particular
online shopping task.
Kamis and Davern (2004)

Theory • Research in product category knowledge
• TAM
• Decision tools implementing different decision strategies for printers and computers
• Shopping tasks for printers and computers
Independent Variables • Product category knowledge
Dependent Variables • Perceived usefulness
• Perceived ease of use
• Decision confidence
Results • Product category knowledge was negatively related to perceived ease of use and perceived usefulness.
• Perceived usefulness was positively related to decision confidence.

Komiak and Benbasat (2006)

Theory • Identification-based trust
• Unit-grouping-based trust
• Cognitive-emotional trust
Context • Lab experiment with 100 student subjects
• Two constraint-satisfaction online product brokering RAs
• Task was to shop for notebook computers, desktop computers, and/or digital cameras using an RA
Independent Variables • Perceived internalization
• Familiarity
Control variables:
• Control propensity
• Trust propensity
• Product experience
• Preference for decision quality vs. effort saving
Dependent Variables • Cognitive trust:
" Competence
" Benevolence
" Integrity
• Emotional trust
• Intention to use an RA as a decision aid
• Intention to use an RA as a delegated agent
Results • Internalization increased cognitive trust and emotional trust in an RA.
• Familiarity increased cognitive trust in benevolence and integrity.
• Cognitive trust and emotional trust predicted intentions to use an RA as a decision aid.
• Emotional trust fully mediated the impact of cognitive trust on the intention to use an RA as a delegated agent .
• No significant differences between the groups in terms of four control variables.
Komiak, Wang, and Benbasat (2005)

Theory Trust formation
Context • An exploratory study with 44 subjects
• Virtual salesperson at RadioShack website
• Task was to compare the services of virtual salesperson and human salesperson
Independent Variables • Type of salesperson (virtual vs. human)
Dependent Variables • Processes for trust and distrust formation
Results • Similar to trust in a human salesperson, trust in a virtual salesperson contained trust in competence, benevolence,
and integrity.
• The formation processes of trust in virtual salespersons, trust in human salespersons, distrust in virtual salespersons,
and distrust in human salespersons were different.
Kramer (2007)
Theory Constructed preferences
Context • Three lab experiments with 102, 123, and 164 subjects, respectively
• RA for digital cameras and PDA
• Tasks for all three experiments consisted of a preference measurement and a recommendation evaluation part
• Task transparency
" Full-profile conjoint analysis
" Self-explicated approach
" Product expertise
Experiment 2
• Task transparency
• Offer timing (immediate vs. delayed)
• Recommendation content (higher-price/quality vs. lower-price/quality)
Experiment 3
• task transparency
• reminder of measured responses (absent vs. present)


• Recommendation acceptance (the acceptance by users of the top-ranked digital camera)
Experiment 2
• Recommendation acceptance
• Likelihood of buying any one of the recommended PDAs
• Difficulty of choosing a PDA from the list of recommendations
Experiment 3
• Recommendation acceptance
• Respondents were significantly more likely to accept a personalized recommendation when their preferences had
been measured using a more transparent task, i.e., self-explicated approach.
• Difference in recommendation acceptance occurred only for novices (i.e., those who did not own a digital camera).
Experiment 2
• Respondents were significantly more likely to choose the recommended PDA following the more transparent
measurement task.
• The effect was moderated by offer timing and recommendation content.
Experiment 3
• Significant transparency × reminder interaction.
McNee, Lam, Guetzlaff, Konstan, and Riedl (2003)

Context • Online experiment with 223 users
• MovieLens: RA for movies
• Task was to use MovieLens to perform movie selections
Independent Variables • Presence of confidence display
• Experience with MovieLens (New vs. experienced users)
• Training vs. no training
Dependent Variables • User satisfaction
• Perceived value of the confidence display
Results • Adding a confidence display to an RA increased user satisfaction.
• Adding a confidence display to an RA altered users’ behavior when engaging tasks with varying amounts of risks.
• New and experienced users reacted differently to the addition of a confidence display.
• Training had a profound impact on user satisfaction in a recommender system: providing training to new users
increased user satisfaction over just adding the confidence display to the system; providing training to experienced
users increased their usage of the confidence system, but decreased their overall satisfaction with the recommender.
McNee, Lam, Konstan, and Riedl (2003)

• MovieLens: RA for movies
• Preference elicitation task
Independent Variables • Three types of RAs
" System-controlled
" User-controlled
" Mix-initiative: provides users with a choice of the system
Dependent Variables • User satisfaction
• Quality of user models
Results • User-controlled interface had the best model quality.
• User-controlled interface had the highest user satisfaction.
" Users of user-controlled interface thought the system best understood their tastes.
" They also had the highest retention rate.
• There was often a tradeoff between giving users control and increasing their effort.
" Users of user-controlled interface spent nearly twice as long as the other two groups in the signup process.
However, they did not feel that they were spending a long time in the signup process, because asking them
to recall titles created focus and engagement with the system.

Moore and Punj (2001)

Context • Lab experiment
• RA for apartment
• An apartment search task
Independent Variables • Environment (web vs. traditional print)
• Time pressure
• Number of alternatives
Dependent Variables • Amount of search
• Satisfaction with search
Results • Amount of information search was higher in the traditional environment.
• Amount of search was not shown to affect satisfaction.
Olson and Widing (2002)

Context • Two lab experiments with 59 and 70 subjects, respectively
• Four different interactive decision aids for word processing programs
• Choice task for word processing programs
Independent Variables • Type of decision aids
" Alphabetical
" Discordant
" Equal weight
" Interactive linear weighted
Dependent Variables • Subjective reactions
" Decision accuracy
" Confusion experienced
" Frustration experienced
" Confidence in choice
" Format satisfaction
• Time
" Perceived decision time
" Decision time
" Relative accuracy
" Discrete accuracy
" Switching
" Selection of dominated alternative
Results • Interactive decision aid led to higher relative accuracy and discrete accuracy than did discordant one.
• Interactive decision aid led to less switching than alphabetical and discordant ones.
• Interactive decision aid led to higher perceived accuracy, confidence, satisfaction and less frustration,
confusion, perceived decision time than did alphabetical and discordant ones.
• Interactive decision aid performed as well as the equal weighted one on both subjective and objective measures.
Pedersen (2000)
Theory Decision making model based on Solomon’s four-stage buying process
• A shopbot for selecting financial service providers
• Task consisted of choosing a financial service provider for a mortgage, a savings account, and a current account
for monthly average outstanding salary after tax
Independent Variables • Access to shopbot service
Dependent Variables • Problem attention
• Problem complexity
• Search time
• Information sources
• Amount of information
• Internet search satisfaction

Xiao & Benbasat/E-Commerce Product Recommendation
Agents
• Consideration set size

• Attributes evaluated
• Attention to quantitative attributes
• Quantitative criteria used in choice
• Single quantitative criteria
• Combined quantitative criterion
Results • At the problem recognition stage, no differences in consumer buying behavior between shopbot users and nonusers
in terms of focused attention to choice problem and understanding of the complexity of choice problem.
• At the information search stage, there was strong evidence of differences in buying behavior. Shopbot users
" Spent less time searching for information.
" Reported a significantly larger number of information sources.
" Collected a greater amount of information.
" Were generally more satisfied with using the Internet as an information search medium.
• At the judgment stage, no differences in consumer buying behavior between shopbot users and nonusers in terms
of consideration set size, attributes evaluated, and attention to quantitative attributes.
• At choice stage, there was only partial and weak support for the hypotheses of differences.
Pereira (2000)
Theory Research in information filtering strategies
Context • Two lab experiments with 160 and 80 subjects, respectively
• A smart agent for cars
• Car choice tasks
• Product knowledge
• Agent search strategy (elimination by aspect, weighted average, profile building, simple hypertext)
Experiment 2
• Product knowledge
• WAD vs. EBA
(the software application for the EBA and WAD search strategies was modified to increase the degree of control
provided to the user and to increase the amount of information provided to the user)
Dependent Variables • Satisfaction with decision process
• Trust in the agent’s recommendations
• Confidence in the decision
• Propensity to purchase
• Perceived cost savings
• Cognitive decision effort
• Subjects with high product class knowledge had more positive affective reactions to the agent in the WAD and
EBA conditions than in the profile building condition.
• Subjects with low product class knowledge had more positive affective reactions to the agent in the profile building
condition than in the EBA and WAD conditions.
Experiment 2
• Subjects responded more positively to the previously less preferred strategy, thus weakening the interaction effect.
Pereira (2001)
• A query-based decision aid for cars
• A choice task for cars
Independent Variables • Access to the query-based decision aid
Dependent Variables • Confidence in the decision
• Satisfaction with the decision process
• Number of stages in the phased narrowing of the consideration set

Mediating variables
• Cognitive decision effort
• size of the consideration set
• similarity among the alternatives in the consideration set
• perceived cost savings
Results • QBDA had a significant impact on all the mediating and dependent variables (e.g. satisfaction, confidence, and
accuracy).
Schafer, Konstan, and Riedl (2002)

Context • Three web-based experiments with 50, 32, and 60 subjects, respectively
• A hybrid RA for movies: MetaLens
• Movie selection tasks
Independent Variables Experiment 1 and 2
• Formats for displaying recommendations (Default, All, Custom, Automatic)
Experiment 3
• Type of RA (meta-recommender vs. traditional recommenders)
Dependent Variables Experiment 1 and 2
• Confidence in movie selection
• Helpfulness of recommendations
• Reliance on previous knowledge in making selections
Experiment 3
• Confidence in movie selection
• Reliance on previous knowledge in making selections
Results Experiment 1 and 2
• The All format that provided the most information was considered most helpful.
• As the amount of recommendation data increased, users found the All format less helpful and began to prefer the
Custom format.
Experiment 3
• A meta-recommender system was considered more helpful and generated more confidence from the users than
traditional recommender systems.
Senecal (2003)
Senecal and Nantel (2004)
Theory Information influence
Human decision-making process
Discounting principle of attribution theory
Context • Online experiment with 488 subjects
• Custom-built RA embedded in three different types of websites
• Three different sampling frames: consumers, people interested in e-commerce, students
• Shop for two products: calculators, and wine
rd rd
Independent Variables • Type of website (seller, 3 party commercially linked to sellers, 3 party not commercially linked to sellers)
• Type of recommendation source (No recommendation source, human experts, other consumers, recommender
system
• Degree of recommendation uniformity
• Product type (search, experience)
• Perceived risks (determined by product type, product class knowledge, product class familiarity)
• Recommendation source credibility (determined by type of website and type of recommendation source)
• Susceptibility to interpersonal influence
Dependent Variables • Consultation or non-consultation of the product recommendation
• Influence of the recommendation source on consumers’ online product selections
• Product choice confidence
• Reason for product choice
Results • RAs were found to be the most influential recommendation source even if they were perceived as possessing less
expertise than human experts and as being less trustworthy than other consumers.

• The type of website on which recommendation sources were used did not affect their perceived trustworthiness
as recommendation sources and did not influence consumers’ propensity to consult or follow the product
recom-mendations.
• The type of product did not influence subjects’ propensity to consult a product recommendation; but for
subjects who did consult a product recommendation, the product type had an influence on their propensity to
follow a product recommendation. Recommendations for experience products were significantly more
influential than for search products.
• Online product recommendations significantly influenced subjects’ online product choices.
• Subjects who consulted and followed a product recommendation showed less confidence in their product
choices than subjects who did not consult the product recommendation and those who did consult the
recommendation but did not follow it.
• Subjects who were influenced by product recommendations tended to lessen the recommendation influence by
attributing their choice to various product attributes.
Sinha and Swearingen (2001)

Context • Three-book RAs and three-movie RAs
• A user study with 19 student subjects
• The task was to test either three-book or three-movie systems as well as to evaluate recommendations made by
three friends
Independent Variables • Source of recommendation (friend vs. online RA)
• Item domain (books vs. movies)
• RA characteristics
Dependent Variables • Quality of recommendation (good recommendations, useful recommendations, trust-generating recommendations)
• Overall satisfaction with recommendations and with online RA (usefulness and ease of use)
Results Results from both quantitative and qualitative analysis.
• Users found friends’ recommendations better and more useful; however, they also found items recommended by
online RAs useful; recommended items were often new and unexpected.
• Recommended items that had been previously liked by users played a unique role in establishing the credibility
of the RA.
• Increase in number of ratings (i.e., amount of user input) did not negatively affect ease of use.
• Users preferred continuous scale to binary choice scale to provide initial ratings.
• Detailed information about recommended items correlated positively with both the usefulness and ease of use
of RAs.
• Users liked easy ways to generate new recommendation sets.
• Interface (navigation and layout) mattered, mostly when it got in the way.
Sinha and Swearingen (2002)

Theory Literature on explanation (expert systems, search engines)
Context • A user study of 5 music RAs
• 12 student subjects
• The task was to rate 10 recommendations provided by the RAs
Independent Variables • Perceived transparency
Dependent Variables • Liking of recommendations
• Confidence in recommendations
Results • In general, users liked and felt more confident in recommendations perceived as transparent.
Spiekermann (2001)
Theory Literature on information search
Context • A 3-D anthropomorphic advisor agent – Luci
• Two products: compact cameras and winter jackets
• An experiment with 206 subjects and a survey with 39 subjects
• The task for the experiment was to shop either for a winter jacket or for a compact camera at an online store
supported by Luci
• The task for the survey was to judge the 112 agent questions employed by Luci (56 questions per product)

Independent Variables Experiment

• Product nature (search vs. experience)
• Purchase involvement
• Stage in the buying process
• Product class knowledge
• Perceived risk (functional, financial, social-psychological, delivery)
• Privacy concerns
• Flow
• Time cost of search
• Benefit of interaction
• Perceived uncertainty
Survey
• Perceived importance of information request
• Perceived legitimacy of information request
• Perceived difficulty of information request
Dependent Variables Experiment
• Interaction with agent
• Manual information search
Both variables are measured on two dimensions: time and the number of page requests
Survey
• Private consumer information cost
Results Experiment
• Consumers preferred to manually control the search process if they perceived more risk..
• Product involvement positively affected both agent interaction and manual searches.
• Product class knowledge negatively influenced the perception of risk.
• More perceived product knowledge led to less interaction with an agent. However, the reduced level of interaction
with an agent may be attributable to the failure of Luci to satisfy the needs of highly knowledgeable customers.
• Higher perceived time cost led to less manual information searches.
• Expressed privacy concerns led to reduced levels of interaction with the agent.
• Unfavorable privacy settings induce a feeling of discomfort among some users which then led to less
interaction with agent.
• Customers associated different types of purchase risk with the products they sought.
• Higher levels of uncertainty in product judgment led to more manual searches and less interaction with an
agent. Survey
• Perceived legitimacy and difficulty of information request affected private consumer information cost.
• The legitimacy of agent questions was relatively stronger when driven by their perceived importance in the
purchase context.
• Users accepted personal questions as long as they relate to the product context.
Stolze and Nart (2004)

Context • A prototype RA for digital cameras
• Comparative user tests with 8 subjects
Independent Variables • Full system (needs-oriented plus feature-oriented) vs. Baseline system (feature-oriented only)
Dependent Variables • User preferences
Results • Users preferred the full system to the baseline system.
Swaminathan (2003)
Theory Literature on information overload
Context • Web-based attribute-based RA
• Computer-based experiment with 100 subjects
• RA-assisted shopping task for backpacking tent and power toothbrush
• Subjects were given 2 weeks to complete the task
Independent Variables • RA (yes/no)
Moderating variables:
• Product complexity
• Category risk
• Category knowledge
• Category order

Dependent Variables • Amount of searched

" The proportion of available alternatives for which detailed product information is viewed
" Whether or not non-dominated alternative is purchased
Results • RA had a greater impact on decision quality under conditions of high category risk.
• RA had a greater impact on reducing the amount of search when the number of attributes used to describe a
product was fewer.
Swearingen and Sinha (2001)

Context • Three book RAs and three movie RAs
• User study with 19 student subjects
• The task was to test either three-book or three-movie systems as well as to evaluate recommendations made by
three friends
Independent Variables • Source of recommendation (friend vs. online RA)
• Item domain (books vs. movies)
• RA characteristics
Dependent Variables • Quality of recommendation (good recommendations, useful recommendations, trust-generating recommendations)
• Overall satisfaction with recommendations and with online RA (usefulness and ease of use)
• Time measures: time spent registering and receiving recommendations from the system
Results • From a user’s perspective, an effective RA inspired trust in the system; had system logic that was at least somewhat
transparent; pointed users towards new, not-yet-experienced items; provided details about recommended items,
including pictures and community ratings; and, finally, provided ways to refine recommendations by including or
excluding particular genres.
• Users expressed willingness to provide more input to the system in return for more effective recommendations.
Swearingen and Sinha (2002)

Context • 11 online RAs
• Two user studies with a total of 32 student subjects
• The task was to interact with several RAs, provide input to the system, and evaluate 10 recommendations from
each system
Independent Variables System as a whole
• System transparency
• Familiar recommendations
Input:
• Type of rating process (open-ended, ratings on Likert Scale, binary liking, hybrid rating process)
• Amount of input (number of items to rate)
• Genre filtering (whether RA offers filter-like controls over genres)
Output:
• Ease of getting more recommendations
• Information about recommended items (basic item information, expert and community ratings, item sample)
• Navigational path to detailed information
Dependent Variables • Liking
• Action towards item
• Usefulness
• Ease of use
• Preferred RA
• Trustworthiness
Results • From a user’s perspective, an effective RA inspired trust in the system; had system logic that was at least somewhat
transparent; pointed users towards new, not-yet-experienced items; provided ease of obtaining more recom-
mendations; provided details about recommended items, including pictures and community ratings; and, finally,
provided ways to refine recommendations by including or excluding particular genres. Users expressed willingness
to provide more input to the system in return for more effective recommendations.
• Trust was affected by several aspects of the users’ interactions with the systems, in addition to the accuracy of the
recommendations themselves: transparency of system logic, familiarity of the items recommended, and the process
for receiving recommendations.

Urban, Sultan, and Qualls (1999)

Theory Literature on trust
Context • A prototype virtual personal advisor for pickup truck purchasing
• A use study with 250 respondents (all of whom had purchased a truck in the last 18 months)
• The task was to evaluate the virtual personal advisor in terms of trust, quality of recommendations and willingness
to use and pay for the service
Independent Variables Types of systems:
• A virtual advisor-based Internet site – Truck Town
• An information intensive site with no advisor
• Traditional auto dealer system
Product knowledge
Dependent Variables • Trust
• Quality of recommendations
• Willingness to use and pay for the service
• Overall preference
Results • The proposed virtual advisor-based site could build trust; most customers would consider buying a vehicle at this
site and would be willing to pay for the service.
• A combination of the advisor site and an information intensive site together dominated the existing auto dealer
system in terms of customer acceptance.
• The preference for the two Internet sites was equal, but analysis indicates segmentation in the site preferences.
Consumers who were not very knowledgeable about trucks, who visited more dealers, and who were younger and
more frequent Internet users had the highest preference for the virtual personal advisor.
van der Heijden and Sorensen (2002)

Theory • Two-stage consumer decision process
Context • An experiment at an artificial camera store with 86 subjects
• A mobile decision aid for digital cameras
• A camera selection task
Independent Variables • Task complexity
• Availability of the decision aid
Dependent Variables • Consideration set quality
Results • The availability of the decision aid
" increased the number of non-dominated alternatives in the users’ consideration set.
" increased users’ confidence in the final decision when the task complexity was high.
• There were no correlations between the three decision effectiveness measures
" users held opinions about accuracy which were decidedly unrelated to actual consideration set quality.
Vijayasarathy and Jones (2001)

Context • A lab experiment with 124 subjects
• RA at MySimon.com
• Purchase simulation task for televisions
Independent Variables • Access to RA
Dependent Variables Perceptions
• Convenience
• Ease of use
• Enjoyment
• Speed
• Helpfulness of comparisons
• Confidence in product selected
• Confidence in merchant selected
Search activities
• Search strategies
• Type of comparisons
• Time to complete task
• Number of vendor sites visited

Results • Those with decision aid perceive online shopping to be more convenient.
• Those with decision aid took less time to complete shopping task.
• Those with decision aid were less confident in their decisions.
Wang (2005)
Theory • Trust
• System restrictiveness
• Custom built RA for digital cameras
• Two tasks: select one digital camera with RA for a friend and another one for a relative
Independent Variables Agent types (decision strategy support)
• Additive compensatory agent
• Elimination by aspect agent
• Hybrid agent
Use of explanations
(with and without)
Dependent Variables • Trust
• Perceived usefulness
• Intention to adopt RA
• Perceived cognitive effort
• Consideration set size
• Decision time
• Perceived agent restrictiveness
• Perceived agent transparency
Results • Explanation use exerted significant effects on perceived agent transparency.
• Decision strategy support had significant effects on perceived cognitive effort.
• Perceived strategy restrictiveness was negatively correlated with both trust and perceived usefulness.
• Perceived agent transparency was positively correlated with trust.
• Perceived cognitive effort was negatively correlated with perceived ease of use.
• There was an interaction effect between decision strategy support and explanation use—the benefits of hybrid
agents were achieved when explanations were provided.
Wang and Benbasat (2004a)

Theory • Trust
• Literature on explanation
• Custom built product-brokering RA based on content filtering
• Two tasks: select one digital camera with RA for a friend and another one for a relative
Independent Variables Type of explanation
• How explanation
• Why explanation
• Decisional guidance
Control variables:
• Trust propensity
• Product experience
• Effort/quality preference
Dependent Variables Trusting beliefs
• Competence
• Benevolence
• Integrity
Results • How explanations affected users’ competence and benevolence beliefs.
• Why explanations affected users’ benevolence beliefs.
• Decisional guidance affected users’ integrity beliefs.
• Trust propensity affects users’ competence beliefs.

Wang and Benbasat (2004b)

Theory • Theories of initial trust formation
• Two tasks: choose a digital camera for a good friend and then select another camera for a close family member
• Subjects were asked to answer several essay questions to justify their trust level in the RA
• How explanation
• Why explanation
Dependent Variables Trust in RA
• Competence
• Benevolence
• Integrity
Results • Different trusting beliefs existed in the initial stage of trust formation.
• Consumers formed different trusting beliefs in different ways.
• Consumers’ trust building and inhibiting processes co-existed in the initial formation stages.
Wang and Benbasat (2005)

Theory Integrated Trust-TAM model
• Two tasks: select one digital camera for a friend and another for a close relative
• How explanation
• Why explanation
Dependent Variables Trust in RA
• Competence
• Benevolence
• Integrity
• Perceived usefulness
• Intention to adopt RA
Results • Consumers’ initial trust and PU had a significant impact on their intentions to adopt RAs, while PEOU did not.
• Consumers’ initial trust and PEOU influenced their PU of the RAs significantly.
• PEOU also influenced consumers’ trust in RA significantly.
• The significant results regarding the impact of trust on PU and on intentions, as well as the impact of PEOU on
trust, confirmed the nomological validity of trust in online RAs.

Appendix B
Theoretical Perspectives
Five theoretical perspectives are proposed to account for the phenomena concerning outcomes of RA use. These perspectives are used in
the “Propositions” section to to generate propositions that are evaluated in light of available empirical evidence. These propositions form
the basis for answering the research questions raised in the “Motivation, Scope, and Contribution” subsection, and for describing the
connections among the key constructs identified in Figure 1.
The following are the theoretical perspectives are utilized:
• Theories of human information processing

• Theory of interpersonal similarity
• Theories of trust formation
• Technology Acceptance Model (TAM)
• Theories of satisfaction
A brief description of each theory follows.
Theories of Human Information Processing
An information processing approach to studying consumer choice is based on the observation that consumers have limited cognitive
capacity to process information (Payne 1982; Payne et al. 1988), which leads to bounded rationality, the notion that individuals seek to
attain a satisfactory, although not necessarily an optimal, level of achievement (Simon 1955).
The effort–accuracy framework epitomized by Payne et al. (1993) is based on the idea that, although consumers have a number of available
strategies in making choices, which strategy is ultimately chosen depends “on some compromise between the desire to make an accurate decision
and the desire to minimize cognitive effort. Since the accuracy and effort characteristics generally differ across strategies for a given decision
environment and across environments for a given strategy, strategy usage will vary depending on the properties of decision task” (Bettman et al.
1998, p. 192). Therefore, a consumer’s decision-making process is seen as a trade-off between the accuracy of the decision and the effort required
to make the decision. There has been mixed evidence on whether the use of electronic decision aids results in better decisions or simply reduces
time and effort. For instance, a series of studies by Benbasat and Todd (1992, 1996) have demonstrated that electronic decision aids are mainly used
to conserve user effort, not necessarily leading to improved decision quality. Other research (Haubl and Trifts 2000) has shown that electronic
decision aids can have favorable effects on both decision quality and decision effort. Punj and Rapp (2003) also suggest that the use of the aid may
increase consumers’ cognitive capacity and help remove cognitive limitations. In line with the theory of bounded rationality, as limitations are
removed, increased effort will lead to better decisions.
Due to the limitation in information processing capacity, it is not feasible for consumers to always evaluate all available alternatives in
detail before making their decisions. Consequently, in order to reduce complexity, it is necessary for them to adopt a two-stage process to
decision making, including an initial screening of available alternatives followed by an in-depth comparison of selected alternatives, before
making their final decision (Payne 1982; Payne et al. 1988).
Furthermore, the theory of constructive preferences postulates that individuals often lack the requisite cognitive resources to form well-
defined preferences that are stable over time and invariant to the context in which decisions are made (Bettman et al. 1998). Instead, they
often construct preferences on the spot when prompted to make decisions (Haubl and Murray 2003). Since these preferences are
constructed, rather than absolute, they are sensitive to the characteristics of the decision environment.
While electronic shopping environments make it possible for consumers to access large amounts of product information, consumers still
need to process increasingly more information with the same limited processing capacity (West et al. 1999). One method of coping with
information overload is to filter and omit information (Farhoomand and Drury 2002; Senecal 2003). Another is to use decision support
tools, such as RAs, to perform resource-intensive information processing tasks, thus freeing up some of the human processing capacity for
decision making (Haubl and Trifts 2000).

Theory of Interpersonal Similarity
Prior research in psychology, sociolinguistics, communication, business, and related fields has supported a positive similarity–attraction
relationship (Byrne and Griffitt 1969), which postulates that the greater degree of similarity between two parties (e.g., in behavior, com-
munication style, attitude, personality, physical appearance, religion, status), the greater the attraction will be. Interpersonal similarity
increases the ease of communication between two individuals who are similar to each other, improves the predictability of the behavior
exhibited by each of them, and fosters relationships of trust and reciprocity among the parties (Levin et al. 2002; Zucker, 1986). Lichtenthal
and Tellefsen (1999) synthesize findings from past research in sales, marketing, and psychology, which indicate that buyers often judge
their degree of similarity with a salesperson in terms of observable characteristics (physical attributes and behavior) and internal
characteristics (perceptions, attitudes, and values). While internal similarity can increase a buyer’s willingness to trust a salesperson and
follow his/her guidance, observable similarity often exerts a negligible influence on buyer perceptions of a salesperson’s effectiveness.
In the same vein, Levin et al. (2002) define interpersonal similarity both in terms of demographics, such as race, gender, education, and
age, and in terms of cognitive processes, such as shared vision and shared language. They have found that, out of the three categories of
variables that affect interpersonal trust (attributes of the relationship between a knowledge seeker and a source of information, attributes of
a knowledge source, and attributes of a knowledge seeker), similarity in the cognitive processes between a knowledge seeker and a source
of knowledge has the most significant effect on trust building, whereas similarity in demographics has little or no effect.
Theories of Trust Formation
RAs are both trust objects and IT artifacts (Gefen et al. 2003). For RAs to be effective, consumers must have confidence in the RAs’ product
recommendations, as well as in the processes by which these recommendations are generated (Haubl and Murray 2003). Therefore, trust remains a
major issue for users of RAs. Furthermore, the relationship between RAs and their users can be described as an agency relationship. By delegating
the task of product screening and evaluation to an RA (i.e., the agent), a user assumes the role of a principal. Thus, as a result of the information
asymmetry underlying any agency relationship, users usually cannot tell whether or not RAs are capable of performing the tasks delegated to them,
and whether RAs work solely for the benefits of the users or the online store where they reside.
Prior trust research (Doney and Cannon 1997; McKnight et al. 1998) has identified many distinct trust formation processes, examples of
which include knowledge, cognition, calculation, transference, prediction, and goodwill. The present paper focuses on three major trust
formation processes, because they cover all of the processes identified in the empirical studies reviewed: knowledge-based processes,
cognitive processes, and transference.
Knowledge-based trust is developed over time as individuals accumulate knowledge that is relevant to trust through experiences with the object of
their trust (Luwicki and Bunker 1995; McKnight et al. 1998). Cognition-based trust is formed by two types of cognitive processes: (1) an
assessment of the trust object’s competence, benevolence, and integrity, and (2) a categorization process, which in turn comprises two sub-
processes (McKnight et al. 1998)—unit-grouping (i.e., when an individual identifies a trust object in the same category as herself or himself) and
reputation categorization (i.e., an assignment of attributes to a trust object based on second-hand information). Trust can also be developed through
transference (Doney and Cannon 1997), that is, it can be transferred from one trusted source to the trust object.
Technology Acceptance Model (TAM)
An RA is essentially a sample of information technology. As such, user intentions to adopt RAs should be explicable in part by the
technology acceptance model (TAM), the parsimony and robustness of which have been demonstrated in numerous empirical studies in IS
literature (see Venkatesh et al. 2003). According to TAM, an intention to adopt a new technology is determined by the perceived usefulness
(PU) of using the technology and the perceived ease of use (PEOU) of the technology.
Theories of Satisfaction
The user satisfaction literature is one of the two dominant research streams (the other being the technology acceptance literature)
investigating perceptions of information system (IS) success (Wixom and Todd 2005). Despite a proliferation of theoretical and conceptual
frameworks related to user satisfaction with information systems, two particularly influential ones are the IS success model (DeLone and
McLean 1992, 2003) and the expectation–disconfirmation paradigm (see McKinney et al. 2002).

DeLone and McLean (1992) presented the IS success model and postulated that information quality (which measures semantic success) and
system quality (which measures technical success) are two key antecedents of user satisfaction. Many empirical undertakings have tested
and validated DeLone and McLean’s IS success model (see DeLone and McLean 2003). Whereas system quality was usually measured in
terms of ease of use, functionality, reliability, flexibility, data quality, portability, integration, and importance, information quality was
generally assessed in terms of accuracy, timeliness, completeness, relevance, and consistency. DeLone and McLean (2003) have updated
their model by adding service quality (usually measured in terms of the dimensions of tangibles, reliability, responsiveness, assurance, and
empathy) to information quality and system quality in their research model.
The expectancy–disconfirmation paradigm has been commonly used in academic marketing studies to measure customer satisfaction (see
McKinney et al. 2002). Based on this paradigm, customer satisfaction is determined by three primary factors: expectation, disconfirmation,
and perceived performance. Expectation is a customer’s “pretrial beliefs” about a product; it is often formed by the customer’s prior
experiences and his or her exposure to vendors’ marketing efforts. Perceived performance is a customer’s perceptions of how product
performance fulfills his or her needs and desires. Disconfirmation occurs when a customer’s evaluations of product performance are
different from his or her expectations about the product (McKinney et al. 2002; Olson and Dover 1979). Satisfaction is achieved when
expectations are fulfilled (confirmed). Negative disconfirmation of expectations will result in dissatisfaction, whereas positive
disconfirmation will result in enhanced satisfaction (Selnes 1998).
Appendix C
Propositions Answering Research Questions
1 How do RA use, RA characteristics, and other factors influence consumer decision-making processes and outcomes?
How does RA use influence consumer decision-making processes and outcomes?
P1: RA use influences users’ decision effort.
P1a: RA use reduces the extent of product search by reducing the total size of alternative sets processed by the users as well as
1.1 the size of the search set, in-depth search set, and consideration set.
P1b: RA use reduces users’ decision time.
P1c: RA use increases the amount of user input.
P2: RA use improves users’ decision quality.
How do the characteristics of RAs influence consumer decision-making processes and outcomes?
P3: RA type influences users’ decision effort and decision quality.
P3a: Compared with pure content-filtering RAs or pure collaborative-filtering RAs, hybrid RAs lead to better decision quality
and higher decision effort (as indicated by amount of user input).
P3b: Compared with non-compensatory RAs, compensatory RAs lead to better decision quality and higher decision effort (as
indicated by amount of user input).
P3c: Compared with feature-based RAs, needs-based RAs lead to better decision quality.
P4: The preference elicitation method influences users’ decision quality and decision effort. The explicit preference elicitation method leads
to better decision quality and higher decision effort (as indicated by amount of user input) than does the implicit preference
elicitation method.
1.2
P5: Included product attribute influences users’ preference function and choice. Included product attributes (in RA’s preference
elicitation interface) are given more weight in the users’ preference function and considered more important by the users than
those not included. Product alternatives that are superior on the included product attributes are more likely to be chosen by
users than are products superior on the excluded product attributes.
P6: Recommendation content influences users’ product evaluation and choice.
P6a: Recommendations provided by RAs influence users’ choice to the extent that products recommended RAs are more likely
to be chosen by users.
P6b: The display of utility scores or predicted ratings for recommended products influences users’ product evaluation and
choice to the extent that products with high utility scores or predicted ratings are evaluated more favorably and are more
likely to be chosen by users.

P7: Recommendation format influences users’ decision processes and decision outcomes.
P7a: Recommendation display method influences users’ decision strategies and decision quality to the extent that sorted
recommendation lists result in greater user reliance on heuristic decision strategies (when evaluating product
alternatives) and better decision quality.
P7b: The number of recommendations influences users’ decision effort and decision quality to the extent that presenting too
many recommendations increases users’ decision effort (in terms of decision time and extent of product search) and
decreases decision quality.
How do other factors (i.e., factors related to user, product, and user-RA interaction) moderate the effects of RA use and RA
characteristics on consumer decision-making processes and outcomes?
P8: Product type moderates the effects of RA use on users’ choice. RA use influences the choice of users shopping for experience
products to a greater extent than that of those shopping for search products.
P9: Product complexity moderates the effects of RA use on users’ decision quality and decision effort. The use of RAs for more complex
products leads to greater increase in decision quality and greater decrease in decision effort than for less complex products.
P10: Product complexity moderates the effect of included product attributes on users’ choice. The inclusion effect is stronger for products with
negative inter-attribute correlations (i.e., more complex products) than for those with positive inter-attribute correlations (i.e.,
less complex products).
1.3
P11: Product expertise moderates the effect of preference elicitation method on users’ decision quality. Preference elicitation method
has less effect on the decision quality of users with high product expertise than on the decision quality of those with low product
expertise.
P12: Perceived product risks moderate the effects of RA use on users’ decision quality and decision effort. When perceived product
risks are high, RA use leads to greater improvements in decision quality and reduction in decision effort than when perceived
product risks are low.
P13: User-RA similarity moderates the effects of RA use on users’ decision quality and decision effort. RA use leads to greater
increase in decision quality and greater decrease in decision effort when the RAs are similar to the users than when the RAs
are not similar to the users.
2 How do RA use, RA characteristics, and other factors influence users’ evaluations of RAs?
2.1 How does RA use influence users’ evaluations of RAs?
How do characteristics of RAs influence users’ evaluations of RAs?
P14: RA type influences users’ evaluations of RAs.
P14a: Compared with pure content-filtering or pure collaborative-filtering RAs, hybrid RAs lead to greater trust, perceived
usefulness, and satisfaction but to lower perceived ease of use.
P14b: Compared with non-compensatory RAs, compensatory RAs lead to greater trust, perceived usefulness, and satisfaction
but to lower perceived ease of use.
P15: The preference elicitation method influences users’ perceived ease of use of and satisfaction with the RAs. Compared to an
explicit preference elicitation method, an implicit preference elicitation method leads to greater perceived ease of use of and
satisfaction with the RAs.
2.2
P16: The ease of generating new or additional recommendations influences users’ perceived ease of use of and satisfaction with
RAs. The easier it is for the users to generate new or additional recommendations, the greater their perceived ease of use of
and satisfaction with the RAs.
P17: User control of interaction with RAs’ preference-elicitation interface influences users’ trust in, satisfaction with, and perceived
usefulness of the RAs. Increased user control leads to increased trust, satisfaction, and perception of usefulness.
P18: The provision of information about search progress, while users await results influences users’ satisfaction with RAs. Users who
are informed about RAs’ search progress (while waiting for recommendations) are more satisfied with the RAs.
P19: Response time influences users’ satisfaction with RAs. The longer the RAs’ response time, the lower the users’ satisfaction with
the RAs.

P20: Recommendation content influences users’ evaluations of RAs.

P20a: Familiar recommendations increase users’ trust in the RAs.
P20b: The composition of the list of recommendations, as reflected by a balanced representation of familiar and unfamiliar (or
new) product recommendations, influences users’ trust in, perceived usefulness of, and satisfaction with RAs.
P20c: The provision of detailed information about recommended products increases users’ trust in, perceived usefulness of,
and satisfaction with RAs.
P20d: The provision of explanation on how the recommendations are generated increases users’ trust in and satisfaction with RAs.
P21: Recommendation format influences users’ perceived usefulness of, perceived ease of use of, and satisfaction with the RAs.
RAs with clear navigational path and layout are considered more useful, easier to use, and more satisfactory than those without.
How do other factors (i.e., factors related to user, product, and user-RA interaction) moderate the effects of RA use and RA
characteristics on users’ evaluations of RAs?
P22: Product type moderates the effects of RA use on users’ trust in and perceived usefulness of RAs. Users have higher trust in
RAs for experience products and higher perceived usefulness of RAs for search products.
P23: Product expertise moderates the effects of RA use on users’ evaluations of RAs (i.e., trust, perceived usefulness, perceived
ease of use, satisfaction). The higher the product expertise of the users, the less favorable the users’ evaluations of RAs.
P24: Product expertise moderates the effects of RA type on users’ evaluations of RAs (i.e., trust, perceived usefulness, perceived ease of
use, satisfaction). The higher the product expertise of the users, the more (less) favorable the users’ evaluations of feature-based
2.3 (needs-based) RAs. The higher the product expertise of the users, the more (less) favorable the user’s evaluations of content-
filtering (collaborative-filtering) RAs.
P25: User-RA similarity moderates the effects of RA use on users’ trust in, satisfaction with, and perceived usefulness of RAs. The more the
RAs are perceived to be similar to their users, the more they are considered to be trustworthy, satisfactory, and useful.
P26: User’s familiarity with RAs moderates the effects of RA use on trust in the RAs. Increased familiarity with RAs leads to
increased trust in the RAs.
P27: The confirmation/disconfirmation of expectations about RAs moderates the effects of RA use on users’ satisfaction with the RAs.
Confirmation or positive disconfirmation of users’ expectations about RAs contributes positively to users’ satisfaction with the RAs.
In contrast, negative disconfirmation of users’ expectations about RAs contributes negatively to users’ satisfaction with the RAs.
How does provider credibility influence user’s evaluations of RAs?
P28: Provider credibility, determined by the type of RA providers and the reputation of RA providers, influences users’ trust in RAs. RAs
2.4
provided by independent third party websites are considered more trustworthy than those provided by vendors’ websites. RAs
provided by reputable websites are considered more trustworthy than those provided by websites that are unknown or non-
reputable.

stats
View
ation
public
Downloaded by University of Rochester At 00:12 17 July 2016 (PT)
The development of e-commerce in Singapore: The impact of government
initiatives Busli Chan Suliman Al-Hawamdeh
B
u
s Busli Chan Suliman Al-Hawamdeh, (2002),"The development of e-commerce in Singapore",
i Business Process Management Journal, Vol. 8 Iss 3 pp. 278 - 288
n http://dx.doi.org/10.1108/14637150210428970
e Downloaded on: 17 July 2016, At: 00:12 (PT)
s References: this document contains references to 10 other documents.
s The fulltext of this document has been downloaded 3681 times since 2006*
P Users who downloaded this article also downloaded:
r
(2001),"E-business e-commerce evolution: perspective and strategy", Managerial Finance, Vol. 27 Iss 7 pp.
o 16-33 http://dx.doi.org/10.1108/03074350110767268
c (2001),"E-commerce marketing strategies: an integrated framework and case analysis", Logistics
e Information Management, Vol. 14 Iss 1/2 pp. 14-23 http://dx.doi.org/10.1108/09576050110360106
(2004),"E-commerce development in China and its implication for business", Asia Pacific Journal
s of Marketing and Logistics, Vol. 16 Iss 3 pp. 68-83 http://dx.doi.org/10.1108/13555850410765230
s
M
a Access to this document was granted through an Emerald subscription provided by emerald-srm:259828 []
n For Authors
a If you would like to write for this, or any other Emerald publication, then please use our Emerald for
g guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
e About Emerald www.emeraldinsight.com
m Emerald is a global publisher linking research and practice to the benefit of society. The company
e manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well
as providing an extensive range of online products and additional customer resources and services.
n Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
t on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
J preservation.
o
u
r
n
a
l
M
M
This article has been cited by:
- Rayed AlGhamdi, Steve Drew, Mohammed AlshehriStrategic government initiatives to promote diffusion of online
retailing in Saudi Arabia 217-222. [CrossRef]
- Siriluck RotchanakitumnuaiDepartment of Management Information Systems, Thammasat University, Bangkok, Thailand. 2008.
Measuring e‐government service value with the E‐GOVSQUAL‐RISK model. Business Process Management Journal 14:5, 724-737.
- Alain Yee‐Loong ChongFaculty of Computing and IT, INTI International University College, Putra Nilai, Malaysia
Keng‐Boon OoiFaculty of Management, Multimedia University, Cyberjaya, Malaysia. 2008. Adoption of
interorganizational system standards in supply chains. Industrial Management & Data Systems 108:4, 529-547.
- Siriluck RotchanakitumnuaiThe Important Risk Factors of E-Government Service Adoption 3652-3655. [CrossRef]
- Vesa P. TaatilaLaurea University of Applied Sciences, Lohja, Finland Jyrki SuomalaDepartment of Psychology,
University of California, Santa Barbara, California, USA Reijo SiltalaDepartment of Teacher Education in Rauma,
University of Turku, Rauma, Finland Soili KeskinenDepartment of Teacher Education in Rauma, University of Turku,
Rauma, Finland. 2006. Framework to study the social innovation networks. European Journal of Innovation Management
9:3, 312-326. [Abstract] [Full Text] [PDF]
Valuable, rare, inimitable resources and organization (VRIO) resources or valuable, rare,
inimitable resources (VRI) capabilities: What leads to competitive advantage?
Article in South African Journal of Business Management · September 2012

DOI: 10.5897/AJBM12.295
CITATIONS READS
24 46,435
2 authors:
Nuno Cardeal Nelson António

Universidade Católica Portuguesa ISCTE-Instituto Universitário de Lisboa
7 PUBLICATIONS 28 CITATIONS 80 PUBLICATIONS 136 CITATIONS
China in Africa View project
Qualstrategy View project
All content following this page was uploaded by Nuno Cardeal on 17 May 2014.

African Journal of Business Management Vol. 6(37), pp. 10159-10170, 19 September, 2012
Available online at http://www.academicjournals.org/AJBM
DOI: 10.5897/AJBM12.295
ISSN 1993-8233 ©2012 Academic Journals
Full Length Research Paper
Valuable, rare, inimitable resources and

organization (VRIO) resources or valuable, rare,
inimitable resources (VRI) capabilities: What leads
to competitive advantage?
1 2
Nuno Cardeal * and Nelson António
1
Católica-Lisbon School of Business Economics, Universidade Católica Portuguesa, Palma de Cima, Lisbon, Portugal.
2
ISCTE-IUL Lisbon, Av. Forças Armadas, Lisbon, Portugal.
Accepted 13 July, 2012
The resource-based view (RBV) argues that valuable, rare, inimitable resources and organization (VRIO)
lead to competitive advantage. Dynamic capabilities (DC) are a comparatively new field and the related
literature is mainly conceptual. Capabilities can be considered as the firm’s routines and processes. We
argue that the “O” in VRIO refers to DC. DCs are the “organization” needed to transform bundles of
resources into competitive advantage. Consequently, does competitive advantage stem from VRIO
resources or from VRI capabilities? Through a case study we analyzed the development of one
capability in a medium-sized Portuguese footwear manufacturer. After reviewing the process of
development of the capability, we performed a VRIO test for each of the resources it exploits and a VRI
test of the capability. We can conclude that none of the resources contributing to the capability are
VRIO, but the capability is VRI.
Key words: Competitive advantage, dynamic capabilities, footwear, resource-based view.
INTRODUCTION
The resource based view (RBV) focuses on specific competitive advantage is achieved and maintained
resources and highlights that competitive advantage is (Ambrosini and Bowman, 2009). The DC literature is quite
based on valuable, rare, inimitable resources and new, which is probably the reason why it is mostly
organization (VRIO) (Barney, 1997). Dynamic capabilities conceptual and provides little empirical evidence, on how
(DC), which represent the firm’s behavioral orientation to overcome the shortcomings of the RBV (Priem and
towards constant integration, reconfiguration, renewal Butler, 2001a).
and recreation of its resources and capabilities and We provide an extensive review of the state of the art of
continuous upgrading and reconstruction of its core the RBV and DC literatures. We conclude that from the
capabilities in response to the changing environment and RBV viewpoint, to have a competitive advantage, firms
to remain competitive (Wang and Ahmed, 2007), focus on need to have VRIO resources. The unit of analysis is the
internal processes or routines. resource. On the other side, from the DC viewpoint,
The RBV does not explain how competitive advantage capabilities in the origin of the competitive advantage
is achieved (Priem and Butler, 2001a). The DC view need to be VRI (being “O” the capability itself). The unit of
explores how the firms’ resources and capabilities evolve analysis is the “O”. Then we argue that DC can be seen
over time and provides a better understanding of how as the “O” in VRIO according to the RBV. Being DCs the
VRIO’s “O”, to achieve competitive advantage, firms need
to have VRI (valuable, rare and inimitable) capabilities. If
we compare the RBV and the DC literature, a tension
*Corresponding author E-mail: nucard@ucp.pt. Tel: +351 217 emerges about whether competitive advantage stems
214 000. from VRIO resources or from VRI capabilities.
10160 Afr. J. Bus. Manage.
We use a case-study methodology to analyze the resources that are not rare or difficult to imitate, then the
development of one capability by a medium sized focal resource cannot be the source of competitive
Portuguese footwear manufacturer. We deeply analyzed advantage (Barney, 2001).
the way the firm renewed, recreated, upgraded and
reconstructed its resources / capabilities in response to
the changing environment. After reviewing the process of Dynamic capabilities (DC
developing the capability, we test its inputs for VRIO and
test the capability for VRI. We find that competitive The DC literature (Adner and Helfat, 2003; Eisenhardt
advantage stems from the VRI capability. We adopt the and Martin, 2000; Helfat et al., 2007; Makadok, 2001;
definition of firm competitive advantage as creating more Teece et al., 1997; Zollo and Winter, 2002) tries to explain
economic value than a marginal competitor that achieves how companies renew their existing capabilities and
only break-even (Peteraf and Barney, 2003). resources to adapt to the changing business
This present study makes several contributions. First, it environment.
provides an in depth analysis of how the capability that So far, the DC literature has focused on conceptual
provides the case firm with its competitive advantage was research which has very little empirical support
developed. This has implications for both managers and (Ambrosini and Bowman, 2009; Barreto, 2010; Di Stefano
scholars/researchers. Second, the findings on VRI DC vs. et al., 2010; Wu, 2010). The few empirical support often
VRIO RBV highlight some of the limitations of the do not address the origins of capabilities development
resource literature concerning the creation of competitive (Felin and Nicolai, 2009). The lack of consensus on DC
advantage. (Arend and Bromiley, 2009; Prieto et al., 2009) has led to
alternative definitions (Barreto, 2010; Døving and
Gooderham, 2008; Eisenhardt and Martin, 2000; Teece et
LITERATURE REVIEW al., 1997; Wang and Ahmed, 2007; Winter, 2003; Zahra et
al., 2006).
Resource based view (RBV) In the DC perspective, and in contrast to the RBV, the
distinction between resources and capabilities is very
Barney (1991) is generally acknowledged as the first clear. The sources of competitive advantage lie in the
scholar to develop the theoretical tool explanatory of RBV context of DC, with greater emphasis on capabilities than
(Newbert, 2008). To have the potential to create on resources because the value of resources in the
sustainable competitive advantage, a resource must be context of dynamic markets tends to depreciate quickly
valuable, rare, inimitable and non-substitutable (VRIN). In (Collis and Montgomery, 2008). Resources remain
the present study, a firm is deemed to have competitive important, not per se, but based on the configuration
advantage when it creates more economic value than its conferred by DC (Cavusgil et al., 2007; Liao et al., 2009;
marginal competitor (Peteraf and Barney, 2003). Morgan et al., 2009; Prieto et al., 2009), because it is not
Contrasting with VRIN, Peteraf and Barney (2003) clear that all companies use their resources in the same
suggest that competitive advantage results from the ways, even those resources that are easily accessible.
existence of critical resources that are used in a superior Instead, firms combine these resources with inside
way. The "higher use" of these resources is aligned to a knowledge, in order to put together the pieces of a puzzle,
new approach that responds to concerns over how to find solutions, and to achieve strategic and operational
resources are transformed into competitive advantage: objectives. The way these resources are interconnected
VRIN resources if managed by unskilled people, unable and the most appropriate combinations and types of
to evaluate their usefulness and/or benefits, or to make interconnections are relevant in DC (Kay, 2010).
appropriate use of them, will provide no benefit to the firm As important as having useful resources is the
(Katkalo et al., 2010). VRIO considers that it is through possession of capabilities, that allows the integration and
the firm’s internal organization that resources are use of these resources (Barney and Wright, 1998;
transformed into competitive advantage (Barney, 1995, Newbert, 2008). Thus, a capability can be seen as the “O”
1997). But what is this “organization”? The existing RVB in VRIO – the way the firms organize a bundle of
literature does not properly address this question. resources.
Organization is seen as “something else”.
Barney (1991) repeatedly uses the term "resource
bundles", resulting in references to integration of VRIO RBV vs. VRI DC
resources; in VRIN, the emphasis is on the resource
level. In VRIO, the emphasis moves downstream to the Value
functionality and/or usefulness of the resources.
However, the unit of analysis is still the resource. Thus, The resource creates value when it allows the company
although the resource may be valuable, rare and difficult to devise and implement strategies that will improve its
to imitate, if there are any strategically equivalent efficiency and effectiveness. Departing from the
Cardeal and António 10161
traditional strengths, weaknesses, opportunities and to the time needed to train employees or to absorb the
threats (SWOT) analysis, an attribute creates value and knowledge necessary to master the resource (Wills-
becomes a resource if it enables the exploitation of Johnson, 2008).
opportunities and/or the neutralization of threats (Barney,
1991). Penrose (1959) identified resources as bundles of
potential services. She also defined services as the result Organization
of how the company uses its resources; they are not
useful unless they are used efficiently. For Amit and Competitive advantage stems from the way firms operate
Schoemaker (1993) companies must have access to and interrelate their strategic and non-strategic resources
adequate capabilities to take advantage of their (Pan et al., 2007), exploiting organizational processes, to
resources. That is, while certain resources may have the produce what can be regarded as intermediate products
potential to create valuable services, the value of these between primary resources and the firm’s final products
services will remain latent until the firm has the (Amit and Schoemaker, 1993).
capabilities needed to deploy them (Newbert, 2008). This The DC literature provides a rich hierarchy of
is consistent with several other definitions of resources, capabilities. For example, Collis (1994) argues that there
for example, resources are valuable when they contribute are first, second and third order, as well as Meta, and ad
to the production of something customers want, at a price infinitum Meta capabilities. Danneels (2002) proposes
they are willing to pay (Collis and Montgomery, 1995); two levels of capabilities (first and second order). Winter
valuable resources enable the firm to do things that lead (2003) proposes the existence of zero-level, first order
to economic value (Fiol, 1991); valuable resources have and higher order capabilities while Zahra et al. (2006)
some capacity to generate profits and prevent losses classify capabilities as substantive and dynamic. They
(Miller and Shamsie, 1996). Words, such as “enable”, have been described also as incremental, renewal and
“contribute” or “some capacity” point to the need for regenerative (Ambrosini et al., 2009), or operational
something else to transform the (valuable) resource into capabilities, portfolios of capabilities and constellations of
the output that provides value. In an attempt to capabilities (Laamanen and Wallin, 2009), or as still
operationalize the concept of valuable resources, capabilities and core capabilities (Wang and Ahmed,
Bowman and Ambrosini (2007) define resources in the 2007). Given the discussion around these hierarchies of
broad sense, to include activities and capabilities. Thus, capabilities, we propose that capabilities are
by themselves, resources cannot be valuable (enablers of organizational processes (skills, expertise, know-how,
the exploitation of opportunities and/or blockers of the management) firms use to explore their resources and
negative effects of threats). lower level capabilities in the conduction of day-to-day
operations. They are intermediary outputs between
resources and lower level capabilities (inputs), and final
Rareness products (outputs). DC is defined as capabilities that
endow firms with competitive advantage in changing
If most competitors hold the same valuable resource,
environments. Both capabilities and DC are
then they will likely explore their use in similar ways, thus “organization”. A DC is the VRIO’s “O”. To be a source of
implementing the same value creating strategy. This
competitive advantage it must be VRI.
would not result in any company achieving competitive
Over time, firms have opportunities to improve their
advantage as a result of owning a valuable resource
organizational processes or routines (which reflect how
(Barney and Zajac, 1994).
things are done in the company). Combining this path
dependency effect with the fact that, due to their strong
Inimitability intangible component, organizational resources or
routines are ambiguous (in the sense that their
If valuable and rare resources are easily imitable, relationship with competitive advantage is not evident and
competitors would quickly copy them and the potential for is often the result of the integration of various packages
competitive advantage would disappear. Resources tend of resources) and are also socially complex, these
to be more difficult to imitate if: (a) they are path processes/routines are difficult for competitors to imitate,
dependent (Dierickx and Cool, 1989; Vergne and Durand, which increases the firm’s potential to achieve
2011); (b) there is an ambiguous relationship between the competitive advantage (Teece et al., 1997).
resources that enhances competitive advantage (Barney, These issues of ambiguity, path dependency and
1995; Dierickx and Cool, 1989; Reed and DeFillipi, 1990); intangibility are so complex that despite having mastered
(c) they are socially complex, for example, if they are the capabilities firms view them as ambiguous and
beyond the ability of firms to manage and manipulate mysterious (Lado and Wilson, 1994; Reed and DeFillipi,
them in a systematic way (Barney, 1991); 1990). However, firms that develop better ways of
h there are legal property rights, such as in the case of integrating their resources have more and stronger
patents (Wills-Johnson, 2008); (e) the process of their sources of competitive advantage.
imitation by other companies is lengthy, for example due To summarize, from the RBV, to achieve competitive
Figure 1. RBV / DC paths to competitive advantage.
advantage, firms need VRIO resources. The unit of individuals (Felin and Nicolai, 2009). It is especially
analysis is the resource. From the DC viewpoint, important to examine how capabilities are developed and
capabilities at the origin of competitive advantage need to to test to what extent they are VRIO, for example,
be VRI (the “O” is the capability) and so the unit of according to the RBV, whether capabilities are enablers
analysis is organization or capability (Figure 1). Which of of competitive advantage (Arend and Bromiley, 2009).
these is dominant? To answer this question we formulate
the research question: Do firms with competitive
advantage have VRIO resources or VRI capabilities? METHODOLOGY
We adopt the definition of firm competitive advantage as creating

more economic value than a marginal competitor (Peteraf and
The current literature Barney, 2003) that achieves only break-even. A firm has competitive
advantage if it systematically achieves net profits. Resources are
Although RBV in general and DC in particular have defined as inputs, and capabilities are defined as intermediary
outputs between resources and competitive advantage. Thus, to
become influential and frequently cited theories in relation
answer our research question, we need to identify which
to theoretical developments in the area of strategy resources/capabilities the firm possesses. Showing how they are
(Kraaijenbrink et al., 2010; Lockett et al., 2009), few developed over time adds to the literature, and allows a more
studies investigate in detail how resources/capabilities comprehensive view of the phenomenon. In relation to capabilities,
are developed over time (Laamanen and Wallin, 2009; understanding how they are developed allows a better
Newbert, 2007), and particularly in more traditional understanding of the resources they exploit. We therefore chose an
inductive case study (Yin, 2009).
industries where there are different conditions and
The case study type can be described as theory building
constraints (Easterby-Smith et al., 2009). The resources (Eisenhardt, 1989; Glaser and Strauss, 1967; Lee, 1999; Lee et al.,
required for competitive advantage often are not readily 1999) and falls within the logic of contemporary description of
identifiable and some studies empirically test the most recent events (Eisenhardt and Graebner, 2007). The research is
identifiable, but probably not the most important ones mainly inductive (Hartley, 2004), based on data analysis to explore
(Lockett et al., 2009) . the phenomenon of interest, which cannot be adequately explained
by the theory.
The literature supports the need for qualitative studies
to understand the origins of capabilities (Felin and
Nicolai, 2009), to understand why companies choose a Data sources and data collection
certain path of development (Laamanen and Wallin,
2009), how firms introduce dynamic components into Data were collected from four main sources (Table 1). Data
existing capabilities (Shamsie et al., 2009), how were collected between January and July 2010. All data were
organizations develop the capabilities/resources that are filed and cataloged in a database. Following the interviews,
the source of their competitive advantage (Ambrosini and some respondents were contacted again for clarifications. The
Bowman, 2009; Easterby-Smith et al., 2009). This is interviews were conducted at the firm’s facilities and lasted
particularly important in slowly changing environments between 40 min and 1 h and 40 min. Interviews were recorded
with infrequent external shocks (Vergne and Durand, and transcribed. In the interview with the head of the firm, we
2011), in smaller companies where resource bundles are
used an adaptation of the critical incident technique (Chell,
less complex (Lockett et al., 2009), and special attention
2004). The main question driving this interview was "what
is needed for capabilities’ implementation (Pablo et al.,
2007) and for decision-making and interactions among were the major strategic moves made by the firm?” Although
this technique can result in information bias due to cognitive
problems on the part of the respondent, it reflects the thinking
of the organization's leader, who, in a medium sized firm, is the
main decision maker.
Table 1. Data sources.
Source Number Example

Interviews 13 Managers, directors
Internal records 57 Sales reports, financial reports, human resources reports
Direct observations 5 Manufacturing processes; informal meetings; employees outside their working context
Press articles 20 Press releases; interviews
Table 2. Quality of the research process.
Tests Case study tactics Research phase Tactics used

Semi-structured interviews; internal records;
To use multiple sources direct observations; press articles.
of evidence (Yin, 2009; Data collection
Data sources were triangulated aiming to reinforce
Eisenhardt, 1989).
the feasibility of their contents.
We respected the sequence: case-study

To establish a chain of questions; protocol; data-bases; and final report.
evidence (Yin, 2009). Data collection
The case-study report was with reference to the
Construct validity data sources from the database.
Revision of case study The case study was subject of a draft report for
Composition review firm staff and also by industry experts.
reports (Yin, 2009).
Codification process
and revision (Miles and Data analysis We used a blind coder to revise our coding process.
Huberman, 1984)
To use case study

External validity Data collection We elaborated a protocol.
protocol (Yin, 2009).
It was developed and included the classification of all

To use case study
Feasibility Data collection the pieces of evidence, including the transcription of all
databases (Yin, 2009).
interviews.
Internal documentation and direct observation constituted fast lead times. In this new era, factors such as flexibility, rapid
sources of data to triangulate the information provided by response, market knowledge, design and marketing have been
respondents and thus control for retrospective bias. We conducted crucial for a firm’s success. New business models are essential for
five direct observations of specific inter- and intra-team problem companies to survive and succeed. Our study case is a medium
solving situations, and of the work methodologies in production. sized Portuguese footwear manufacturer (ABREUS), which has
Employees were observed outside the work context but within the recently developed a new business model. The Portuguese
company, in order to better understand the informal relationships footwear industry fits with what Eisenhardt and Martin (2000)
that existed. describe as moderate dynamic markets. By selecting a medium
In order to assure the quality of the research process, we sized company, we chose a context where resources and
followed the tactics summarized in Table 2. capabilities bundles are less complex. ABREUS was founded by
two brothers in 1994. In the industrial district (Felgueiras), it is
acknowledged to be a company that is “doing a good job”. In 2007,
Research setting and case ABREUS felt the force of changes in customer behavior when they
placed their large orders in other markets. In 2007 some major
Since the 1970s Portugal has been an important player in the ABREUS customers went elsewhere or reduced their purchases
international footwear industry. In the 1970s and 1980s, the industry from ABREUS, which accelerated the strategic reorientation of the
grew based on Portuguese low labor costs. In the 1990s, there was firm Table 3 shows the major economic and financial indicators of
a relocation of large series productions to the Asian countries. By the firm over the last years.The ABREUS brand is GOLDMUD
the 2000s, footwear had become an important fashion accessory, which was established in 2006. From the start, the brand was
and consumers were showing preferences for distinctive and managed by a new firm owned jointly by Miguel Abreu (the firm’s
different products. The industry has evolved towards flexible main leader), his father and his uncle, the owners of ABREUS.
production of small series of distinctive fashion footwear, with very Currently, ABREUS’s source of competitive advantage lies in its
Table 3. Major economic and financial indicators (thousand €).
Year
Financial indicators
2005 2006 2007 2008 2009 2010
Sales 8.037 11.263 10.501 8.819 7.974 7.135
Net profits 86 94 87 23 24 21
Assets 3.332 3.927 5.541 5.897 5.643 6.441
Equity 701 715 952 976 1.109 985
Source: company internal records (ABREUS, 2005 to 2010).
ability to create value for its customers by developing new products, essential for good customer service in relation to NPD.
coupled with the ability to produce them with reasonable quality, at However, there is also a need for the capability to
reasonable cost and on time. That is, the firm creates value by
interpret customers’ ideas and to find innovative
developing solutions for difficult and distinctive shoes. The
milestones leading to ABREUS’s current situation are identified in solutions, in relation to materials and construction or
Table 4. finishing. ABREUS’ NPD is distinguished by the ability to
suggest new solutions:
DATA ANALYSIS "... Today we often suggest new things to the customer.
We do more than the client initially wants. We suggest
ABREUS capabilities development many things, and it makes a big difference. Firstly puts us
at a level that allows us to escape the discussion of price
The changing process of ABREUS was originated from a as a priority ..." (Miguel Abreu)
combination of two circumstances that occurred between
2006 and 2007. In 2006, the company launched its brand. This approach to value creation by proposing new
In 2007, two of its major customers significantly reduced materials, and/or new constructions, had been dominant
the volume of their purchases. In 2008, their purchases in the firm for 2 to 3 years at the time of our study. The
reduced again. ability to propose different solutions begins with the
The period from 2006 to 2009 corresponds to a phase capacity to interpret customers’ ideas and needs, so
of significant investment in human resources and knowledge about customers and the types of shoes they
technology development. In 2005, the development buy is important for the development phase. In the initial
department employed four. In late 2009, the structure was developments for a particular client, ABREUS may not be
led by Miguel Abreu and was composed of 14 people, fully aware of specific needs. It makes efforts to
including 4 modelers and a stylist. In this period, the firm understand better what the customer wants through
bought several new Computer-aided design (CAD) conversations with the client or his representative or
stations. through surveys (for example the Internet). The modeling
Between 2008 and 2009, the company made several team does not have a set guideline about what to do to
investments to adapt the plant building including better understand the development context. Modelers are
increasing the physical space dedicated to new product allocated to specific costumers and it is their own
development (NPD), creating a specific area to install a responsibility to manage the process of interpreting their
samples production line separate from production. During specific needs.
this period ABREUS acquired the necessary equipment The company's decision to launch its own brand had a
to create this line, the need for which had become evident major impact on ABREUS’s current capability in NPD, in
if the company was to be able to respond to increasing particular because the new brand:
requests from GOLDMUD as well as from new customers
with different needs from the firm’s traditional ones. x) Was seen not as a brand belonging to the firm
but as a new customer: From the outset, GOLDMUD was
The human resources dedicated to NPD combined with posi-tioned as a very demanding customer and
the new equipment, provided increased capacity for the contributed to improvement in the factory’s quality levels:
production of samples. It enabled ABREUS to respond
appropriately and to respect clients’ deadlines: "... GOLDMUD is a very demanding customer! It has to
be and pushes everyone to be more demanding too. You
"... Now they have a good capacity of producing samples, cannot close your eyes and non-quality products cannot
at the very best level. It is pretty good..." (Consultant at pass up … every one pair is seen under the microscope"
Technological Centre). (Head of planning department).
This capability to produce samples and prototypes is GOLDMUD’s product development is conducted jointly
Table 4. Major decisions relevant to the firm’s current situation.
Year Decision/fact Rational Impact on actual capability

1995 Integration of modeling To start the NPD Minimum
To reinforce the commercial He was the great mentor and leader in the
1997 Admission of Miguel Abreu
department development process of new capability..
1998 Exhibition at MOCAP Search for new costumers Minimal
Beginning of the relationship Creating a new business model of Business that is in its final stage of the life
2000
with Grupo Sonae agency. cycle. Minimum.
2000 to Integration of several Allows the vertical integration of the production of

Meet different customer needs
2006 mounting systems a greater variety of types of shoes.
Adjust the production structure to

Acquisition of a automatic smaller production series and to Increased production of samples and a series
2004 increase the production capacity of smaller dimension
cutting machine
of samples.
Acquisition of a second
2006 Raise on the production capacity Greater flexibility
assembly and finishing line
Reduce dependence on Strengthening the structure of NPD. Higher

2006 Creation of GOLDMUD intermediate customers. levels of competence in NPD and
Increasing the value added. production. Raise of the production flexibility.
Costumers that had price as Greater openness to pursuing a different

2007 Loss of important costumers key buying factors strategy
New customers with different needs from

2007 Exhibition at MICAM Raise the notoriety of the brand
traditional ones.
Changing the culture, improving the quality

2007 Subcontracting CTCP Support the process of change
and flexibility.
Patent protected, gave the experience and

Creation of Develop a distinctive solution
2008 confidence that allows the company to
reverted mocassin for GOLDMUD
easily develop new assembly systems.
Creating space for the NPD and Allowed the creation of a sample production
2008 Expansion of facilities line.
improve image with new customers.
Increased capacity of NPD, larger installed

Investment in a production Improve the capacity of response to
2009 capacity of samples and faster response in
line for samples samples
sending samples to customers.
together by the stylist and ABREUS’s technical team. A This accumulated technical knowledge contributes to
GOLDMUD collection includes very diverse construc- subsequent NPD for other customers. Learning from
tions, distinctive finishes, innovations, different designs suppliers has become very important for the firm. As
and novel parts. All of this variety forces the technical Miguel Abreu said:
team to extend its skills. They frequently learn from
leather or finishing materials suppliers to accumulate the "... It is necessary to acquire more knowledge and what
knowledge needed to create the models for the brand. we need to do is learning. I have to rely on the relevant
Figure 2. Logical model of NPD development.
people. I know whom to call, people I can rely on ...“ customers, especially those with similar product
concepts, and ABREUS became a partner in the
(G) The new brand has led to the development of a development and production of their new products. The
new shoe assembly process: With GOLDMUD, Miguel emergence of new customers and the disappearance of
Abreu tried to “do something different”. His idea was so old ones, created an opportunity for changes to the whole
different, that the company had to develop a new system NPD area in terms of human resources and its physical
for assembling shoes, called "reverted moccasin”. Miguel space and equipment.
Abreu asked for a different mounting, flexible and Figure 2 is a logic model (Yin, 2009) depicting how
comfortable, and since the factory was familiar with ABREUS created the capability to develop solutions for
production of “moccasins”, the team was able to create difficult and distinctive shoes.
this new mounting system, which leveraged staff skills in
the direction the leader intended.
Once the new mounting system was developed, it was DISCUSSION
decided that it would only be used in GOLDMUD shoes.
Several customers have requested the new construction, Having identified the capability underlying the company's
but ABREUS has not given in to these requests. The competitive advantage and collecting the data needed to
experience of developing the “reverted moccasin” allowed understand how it was created over time, we want to test
ABREUS to develop new mounting systems for specific input resources for VRIO and the capability for VRI. The
customers. analysis starts from the suggestions in Kraaijenbrink et al.
(H) The brand has given the company visibility and (2010) and Locket et al. (2009), which are compatible with the
opened the firm’s doors to new customers with different proposal in Peteraf and Barney (2003), for example, a model in
needs: With the creation of GOLDMUD, the company which competitive advantage stems from how the firm
resumed exhibitions at trade fairs, although only the most
integrates a set of critical resources. Kraaijenbrink et al. (2010)
important footwear fairs. These exhibitions brought new
also suggest that a distinction
Table 5. VRIO analysis of NPD.
Resource Valuable Rare Inimitable

Reverted Mocassin - + +
Creative Design - + -
Technical Design - + +
Technical Skills - - -
Technology relationship with the cluster
New Product Development + + +
Source: Author analysis.
should be made between resources and capabilities, inimitability of those resources considered rare, we
while Locket et al. (2009) suggest that researchers conducted a content analysis of the characteristics that
should invest in understanding the functionality of create barriers to imitability (Barney, 1991; Peteraf and
resources and how they relate to the markets for Barney, 2003).
products/services in which companies compete. Our first finding is that, individually, none of the
Thus, capability is the result of how the company resources is valuable. This is consistent with the DC
integrates a particular set of critical resources. Individual literature (Katkalo et al., 2010).
resources are the inputs to the capability. The capability is In the firm’s business model, the “reverted moccasin”
the "output" of how the company integrates these critical on its own is not valuable because it needs to be coupled
resources to the capability corresponding to the to other resources such as design, to be transformed into
organization, that is, the "O" of the model VRIO (Barney, a shoe that creates value. Property rights on the “reverted
1997). moccasin” determine that this resource cannot be
The content analysis identified a set of resources and imitated, defining it as a unique and rare solution.
capabilities that are the capability’s inputs (Table 5). Design falls into two major phases. The first phase
There were some problems related to what Kraaijenbrink relates to the creative process, which is related to the
et al. (2010) call the impracticality of defining resources, concept of the new product - called creative design. This
which stems from the fact that the definitions in the is where the development and selection of different
literature (Amit and Schoemaker, 1993; Barney, 1991, themes for the collections and the "scribbles", which
2002; Wernelfelt, 1984) are ambiguous, allowing almost correspond to the drawings of the designers, are
anything to be framed as a resource (Steen, 2010). For performed. At ABREUS, this resource is applied only at
example, should "relationship with the cluster” be GOLDMUD. In this business model, creative design alone
considered a resource or a capability? "Relationship with does not constitute a valuable resource because the
the cluster" surely is the result of how the firm, over time, company needs to have other basic resources, such as
develops the integration of other basic resources and, technical design and technology among others, to turn
according to the methodology used in this study, should the creative design into outputs that allow the exploitation
be considered a capability. If “relationship with the of opportunities.
cluster" is a capability, this raises another question, which The creative design associated with GOLDMUD is rare
enters the domain of the hierarchy of capabilities. in the sense that the brand has an identity that is distinct
Since "relationship with the cluster" is an input to the from any other. However, having defined the concept of
analyzed capability, then the analyzed capability, for the brand and developed the first collections that created
example according to Collis (1994), must be a second its identity, creative design with the features of
order capability. To prevent a discussion on the hierarchy GOLDMUD became easily imitable. That is, it would be
of capabilities, which is not the purpose of this study, we relatively easy for another company or designer not
assume that: connected to ABREUS, to develop a creative design for a
shoe that imitates the spirit of GOLDMUD.
*
“Inputs" are isolated resources or capabilities that are The second phase of design is technical design, where the
integrated to produce an output capability; concept is translated from the creative design into a new
*
"Intermediary output" is the capability under analysis. product (design model), the soles are developed, and the
construction and selection of materials is agreed. It reflects
Having identified the capability’s inputs, we analyzed how the company is able to understand the needs of a specific
each resource individually, primarily in terms of value and client with a new product concept (for example, a picture of a
rarity and subsequently in terms of their inimitability. shoe or a draft design), and the ability to create new products
Resources that were not identified as rare were not based on an individual client’s needs. Technical design on its
analyzed through the prism of inimitability. To investigate
own is not
Table 6. NPD’s coding – Inimitability.
Resource Property rights

Path dependent
Reverted Mocassin Patent protected
- "... the brand is important not only for being a new brand but also
because it shows our capabilities. And it began to bring new
customers ... "
- "... the modeling is difficult (GOLDMUD) ... models are difficult,

different constructions, different models, applications. You learn by
working with them as situations where we must use our skills to do
something different are frequent. With this experience you gain
experience for other things with other customers ... "
- "... if the modeler spends a lifetime making some sportive shoes, for
Technical design example, only knows that. We do everything down there, everything
appears: gloves, mounted, pratiques, California, strobel, child, we do
everything. And there are not many modelers that are able to work
with several constructions and to work man shoes and woman’s also.
And as GOLDMUD has many constructions, different designs,
different parts, you learn, and if a customer wants something
different, you are able to develop it… and occasionally it happens ... "
- "... for example, this reverted mocassin we make for GOLDMUD,

but we do one somehow similar to SNIPE, which is a mocassin that
you pull the manual sewed almost to the ground. These are solutions
of ABREUS. They were developed through testing and based on the
experience we are acquiring. The idea came from GOLDMUD ... "
valuable. To take advantage of this resource, for example, for the development of new products includes CAD
the firm would need to transform its input into a technical systems, leather cutting machines and the production
model which could be produced in an industrial machinery to produce samples and to test new models.
environment and subsequently to produce or outsource Technology alone cannot be valuable because the proper
its production. This is a relatively rare resource in the functioning of these devices depends largely on the
industry. operator’s technical skills.
According to many experts we consulted, only a The relationship with the cluster essentially captures
minority of Portuguese footwear firms posses this the firm’s predisposition to work with the most innovative
resource at this level of development. The technical suppliers and adopt new materials and finishing solutions.
design at ABREUS has the characteristics that it is path Without adequate technical capacity, or the technology
dependent; for example, new constructions developed needed to turn final designs using innovative solutions
and suggested to customers result in adaptations to into tested models, the relationship with the cluster does
“reverted moccasin”. Past experience with developments not allow the exploitation of opportunities or the mitigation
at GOLDMUD influence the firm’s technical design in of the effects of possible threats.
relation to high standards and finishing concepts and The analysis on each of the NPD capability resources
materials, where experience with the firm’s original shows that, individually, none of them passes VRIO test
brands benefits ABREUS (Table 6). for competitive advantage. ABREUS’s competitive advan-
Technical skill reflects the team’s knowledge and allows tage lies in how the firm is organized in order to integrate
the design to be transformed into a testable model. It the six resources. The NPD capability represents the way
involves CAD knowledge, as well as the knowledge (the organization) the firm integrates the resources
needed to test new models and the sequence of “reverted moccasin”, creative and technical designs,
productive operations needed to minimize production technical skills, technology and relationship with the
costs based on the ability to mitigate technical modeling cluster. In the ABREUS business model this capability is
and potential quality problems at an early stage. The valuable, because it allows the firm to generate outputs,
technical skill in NPD on its own is not valuable because exploit opportunities and mitigate threats, more precisely
without the right technologies or without a corporate it allows the offer of complete shoe solutions that are
culture already geared to flexibility and quality, perfor- distinctive and difficult to manufacture. The value of this
mance in NPD would be substantially lower. Technology capability is strengthened by its combination with other
resources, for example, production technology, opera- To confirm whether or not these findings are genera-
tional flexibility, and product quality. Those resources and lizable, we need more qualitative and/or quantitative
capabilities already existed at ABREUS and when studies.
improved, allow ABREUS to integrate vertically the
manufacturing of the shoes developed.
Since the company’s NPD capability is the result of the REFERENCES
integration of six resources, its imitability by competitors Adner R, Helfat CE (2003). Corporate effects and dynamic managerial
is made more difficult due to causal ambiguity (Barney, capabilities. Strat. Manage. J. 24(10):1011-1025.
1991). Indeed, the more complex the capability, for Ambrosini V, Bowman C (2009). What are dynamic capabilities and are
example, because it includes a larger set of resources, they a useful construct in strategic management? Int. J. Manag. Rev.
the less evident it becomes, even to the firm and the less 11(1):29-49.
Ambrosini V, Bowman C, Collier N (2009). Dynamic capabilities: an
easy it is to understand the origins of competitive exploration of how firms renew their resource base. Brit. J. Manage.
advantage. 20(S1):S9-S24.
Amit R, Schoemaker PJH (1993). Strategic assets and organizational
rent. Strat. Manage. J. 14(1):33-46.
Arend RJ, Bromiley P (2009). Assessing the dynamic capabilities view:
Conclusion spare change, everyone? Strateg Organ. 7(1):75-90.
Barney JB (1991). Firm resources and sustained competitive
The previous discussion shows that the company has advantage. J. Manage. 17(1):99-120.
one capability (development of new products) that passes Barney JB (1995). Looking inside for competitive advantage. Acad.
Manag. Exec. 9(4):49-61.
VRI DC test. The inputs (resources) of the capability were Barney JB (1997). Gaining and Sustaining Competitive Advantage.
tested according to VRIO in the RBV. None passed the Addison-Wesley: Reading, MA.
test; particularly due to the valuable issue (none of the Barney JB (2001). Is the resource-based "view" a useful perspective for
input resources/capabilities is, separately, valuable, as strategic management research? Yes. Acad. Manage Rev. 26(1):41-
56.
the firm needs some organization to take full advantage Barney JB, Wright P(1998). On becoming a strategic partner: The role
of them). This is consistent with previous RBV literature. It of human resources in gaining competitive advantage. Hum. Resour.
can be concluded that, in this case, the competitive Manage. 37(1):31-46.
advantage stems from a VRI capability and not from Barney JB, Zajac EJ (1994). Competitive organizational behavior:
Toward an organizationally-based theory of competitive advantage.
VRIO resources. The complexity of the analyzed Strat. Manage. J. 15(S1):5-9.
capability makes it socially complex, contributing to the Barreto I (2010). Dynamic capabilities: A review of past Research and
reinforcement of its inimitable characteristic. This finding an agenda for the future. J. Manage. 36(1): 256-280.
is of the outmost relevance for managers. The more Bowman C, Ambrosini V (2007). Identifying valuable resources. Eur
Manag. J. 25(4):320-329.
complex capabilities that lead to competitive advantage Cavusgil E, Serggie SH, Talay MB (2007). Dynamic capabilities view:
are, in terms of bundles of resources they integrate, the Foundations and research agenda. J. Market Theor. Pract. 15(2):159-
more difficult for competitors is to copy them, allowing 166.
competitive advantage to be sustained for longer periods Chell E (2004). Critical Incident Technique. In Essential Guide to
Qualitative Methods in Organizational Research. Cassel C, Symon G
of time. (eds.), SAGE Publications: London.
From a theoretical point of view, this article contributes Collis DJ (1994). Research Note: How Valuable are Organizational
to the RBV literature in general and to work on DC in Capabilities? Strat. Manage J. 15(S1):143-152.
particular based among other reasons, on the findings Collis DJ, Montgomery CA (1995). Competing on resources: strategy in
the 1990s. Harvard Bus Rev. 73:118-128.
related to VRI DC vs. VRIO RBV: Collis DJ, Montgomery CA (2008). Competing on resources. Harvard
Bus. Rev. 86(7,8):140-150.
*
Lack of examination of the VRIO conditions is a Danneels E (2002). The dynamics of product innovation and firm
competences. Strat. Manage. J. 23(12):1095-1121.
limitation in the previous literature. Di Stefano G, Peteraf MA, Verona G (2010). Dynamic capabilities
*
In finding how the capability was developed, and deconstructed: a bibliographic investigation into the origins,
analyzing the underlying micro-dynamics of organiza- development, and future directions of the research domain. Ind. Corp
Change 19(4):1187-1204.
tional activity, this article makes an important contribution Dierickx I, Cool K (1989). Asset stock accumulation and sustainability of
by addressing an important critique of the RBV, which is competitive. Manage. Sci. 35(12):1504-1513.
that it does not address how resources/capabilities are Døving E, Gooderham PN (2008). Dynamic capabilities as antecedents
transformed into competitive advantage. of the scope of related diversification: the case of small firm
* accountancy practices. Strat. Manage. J. 29(8):841-857.
By defining capabilities as the way that the company Easterby-Smith M, Lyles MA, Peteraf MA (2009). Dynamic capabilities:
integrates a bundle of resources, the research defines current debates and future directions. Brit. J. Manage. 20(S1):S1-S8.
capabilities as an intermediate output between resources Eisenhardt KM (1989). Building theory from case study research.
and competitive advantage, contributing to filling a gap in Academy of Management. Acad. Manag. Rev. 14(4):532-550.
Eisenhardt KM, Graebner ME (2007). Theory building from cases:
the resource literature, and resolving some of the opportunities and challenges. Acad. Manage. J. 50(1):25-32.
confusion surrounding the use of the terms resources, Eisenhardt KM, Martin JA (2000). Dynamic capabilities: What are they?
capabilities, skills, and assets to refer to the same thing. Strat. Manage. J. 21(10/11):1105-1121.
The main limitation of this article is that the findings Felin T, Nicolai JF (2009). Organizational routines and capabilities:
Historical drift and a course-correction toward microfoundations.
cannot be generalized because we study only one firm.
Scand J. Manag. 25:157-167. Penrose ET (1959). The Theory of the Growth of the Firm (3 ed.). John
Fiol CM (1991). Managing culture as a competitive resource: An Wiley Sons: New York.
identity-based view of sustainable competitive advantage. J. Manage Peteraf MA, Barney JB (2003). Unraveling the Resource-Based tangle.
17(1):191-211. Manag. Decis. Econ. 24(4):309-323.
Glaser B, Strauss AL (1967). The Discovery of Grounded Theory: Priem RL, Butler JE (2001a). Is the resource-based "view" a useful
Strategies for Qualitative Research. Aldine de Gruyter: New York. perspective for strategic management research? Acad. Manage Rev.
Hartley J (2004). Case Study Research. In Essential Guide to 26(1):22-40.
Qualitative Methods in Organizational Research. Cassel C, Symon G Prieto IM, Revilla E, Rodríguez-Prado B (2009). Building dynamic
(eds.), SAGE Publications Ltd: London. capabilities in product development: How do contextual antecedents
Helfat C, Finkelstein S, Mitchell W, Peteraf M (2007). Dynamic matter? Scand J. Manag. 25(3):313-326.
Capabilities: Understanding Strategic Change in Organizations.
Blackwell: Oxford, UK. Reed R, DeFillipi RJ (1990). Causal ambiguity, barriers to imitation, and
Katkalo VS, Pitelis CN, Teece DJ (2010). Introduction: on the nature and sustainable competitive advantage. Acad. Manag. Rev. 15(1):88-102.
scope of dynamic capabilities. Ind. Corp Change 19(4):1175-1186. Shamsie J, Martin X, Miller D (2009). In the old, in the new: capabilities,
Kay N (2010). Dynamic capabilities as context: the role of decision, strategies, and performance among the Hollywood studios. Strat.
system and structure. Ind. Corp Change 19(4):1205-1223. Manage. J. 30(13):1440-1452.
Kraaijenbrink J, Spencer JC, Groen AJ (2010). The Resource-Based Steen J (2010). Actor-network theory and the dilemma of the resource
View: A review and assessment of its critiques. J. Manage. 36(1):349- concept in strategic management. Scand J. Manag. 26:324-331.
372. Teece DJ, Pisano G, Shuen A (1997). Dynamic capabilities and the
Laamanen T, Wallin J (2009). Cognitive dynamics of capability strategic management. Strat. Manage. J. 18(7):509-533.
development paths. J. Manage. Stud. 46(6):950-981. Vergne JP, Durand R (2011). The path os most persistence: An
Lado AA, Wilson MC (1994). Human resource systems and sustained evolutionary perspective on path dependence and dynamic
competitive advantage: A competency-based perspective. Acad. capabilities. Organ. Stud. 32(3):365-382.
Manag. Rev. 19(4):699-727. Wang CL, Ahmed PK (2007). Dynamic capabilities: a review and
Lee TW (1999). Using Qualitative Methods to Organize Res. SAGE: research agenda. Int. J. Manag. Rev. 9(1):31-51.
Newbury Park, CA.
Lee TW, Mitchell TR, Sablynski CJ (1999). Qualitative research and Wills-Johnson N (2008). The networked firm: a framework for RBV. J.
vocational psycol. J. Vocat. Behav. 55(2):161-187. Manag. Dev. 27(2):214-224.
Liao J, Kickul JR, Ma H (2009). Organizational dynamic capability and Winter SG (2003). Understanding dynamic capabilities. Strat. Manage.
innovation: an empirical examination of internet firms. J. Small Bus. J. 24(10): 991-995.
Manage. 47(3):263-286. Wu L (2010). Applicability of the resource-based and dynamic-capability
Lockett A, Thompson S, Morgenstern U (2009). The development of the views under environmental volatility. J. Bus Res. 63(1):27-31.
resource-based view of the firm: a critical appraisal. Int. J. Manag. Yin RK (2009). Case Study Research: Design and Methods (4 ed.).
Rev. 11(1):9-28. SAGE Inc.: Thousand Oaks.
Makadok R (2001). Toward a synthesis of the resource-based and Zahra S, Sapienza H, Davidsson P (2006). Entrepreneurship and
dynamic-capability views of rent creation. Strat. Manage. J. dynamic capabilities: a review, model and research agenda. J
22(5):387-401. Manage. Stud. 43(4):917-955.
Miles M, Huberman AM (1984). Qualitative data analysis. SAGE: Zollo M, Winter SG (2002). Deliberate learning and the evolution of
Newbury Park. dynamic capabilities. Organ Sci. 13(3):339-351.
Miller D, Shamsie J (1996). The resource-based view of the firm in two
enviroments: The Hollywood film studios from 1936 to 1965. Acad.
Manage. J. 39(3):519-543.
Morgan NA, Vorhies DW, Mason CH (2009). Market orientation,
marketing capabilities, and firm performance. Strat. Manage. J.
30(8):909-920.
Newbert SL (2007). Empirical research on the resource-based view of

the firm: an assessment and suggestions for future research. Strat.
Manage. J. 28(2):121-146.
Newbert SL (2008). Value, rareness, competitive advantage, and
performance: a conceptual-level empirical investigation of the
resource-based view of the firm. Strat. Manage. J. 29(7):745-768.
Pablo AL, Reay T, Dewald JR, Casebeer AL (2007). Identifying enabling
and managing dynamic capabilities in the public sector. J. Manage.
Stud. 44(5): 687-708.
Pan SL, Tan BCC, Huang J, Poulsen B (2007). The development paths
of non-strategic capabilities. Eur. Manag. J. 25(5):344-358.
Briefings in Bioinformatics, 20(4), 2019, 1094–1102
doi: 10.1093/bib/bbx083
Advance Access Publication Date: 31 July 2017
Paper
PATRIC as a unique resource for studying

antimicrobial resistance
Downloaded from https://academic.oup.com/bib/article-abstract/20/4/1094/4056412 by guest on 15 October 2019

Dionysios A. Antonopoulos, Rida Assaf, Ramy Karam Aziz, Thomas Brettin, Christopher
Bun, Neal Conrad, James J. Davis, Emily M. Dietrich, Terry Disz, Svetlana Gerdes,
Ronald W. Kenyon, Dustin Machi, Chunhong Mao, Daniel E. Murphy-Olson, Eric K.
Nordberg, Gary J. Olsen, Robert Olson,
Dionysios A. Antonopoulos is a Microbiologist who is a staff scientist in the Biosciences Division at Argonne National Laboratory and an Assistant Professor in the University of Chicago
Department of Medicine in Illinois, USA.
Rida Assaf is a PhD student in the Department of Computer Science at the University of Chicago in Illinois, USA.
Ramy Karam Aziz is a Professor and Acting Chair at the Department of Microbiology and Immunology, Faculty of Pharmacy, Cairo University, Cairo Egypt.
His research focuses on microbial and viral genomics and metagenomics.
Thomas S. Brettin is a Strategic Program Manager for Computing and Life Sciences within the Computing, Environmental and Biological Sciences Directorate at Argonne National
Laboratory in Illinois, USA.
Christopher Bun has a PhD degree in Computational Biology in the Department of Computer Science, University of Chicago in Illinois, USA.
Neal Conrad is a Software Engineering Associate at Argonne National Laboratory and the University of Chicago Computation Institute who specializes in Web application development
and user experience for bioinformatics.
James J. Davis is a Computational Biologist at Argonne National Laboratory and the University of Chicago Computation Institute in Illinois, USA.
Emily M. Dietrich is a Coordinating Writer/Editor at Argonne National Laboratory and a joint appointment at the University of Chicago Computation Institute in Illinois, USA.
Terrence Disz, PhD, is a Bioinformatics Software Specialist at the Fellowship for Interpretation of Genomes in Illinois, USA.
Svetlana Gerdes, PhD, is a Comparative Genomics Specialist at the Fellowship for Interpretation of Genomes in Illinois, USA.
Ron Kenyon is a Project Director at the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Dustin Machi is a Senior Software Architect at the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Chunhong Mao is a Research Assistant Professor at the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Daniel E. Murphy-Olson is a Cloud Services Team Lead at Argonne National Laboratory and Joint Staff at the University of Chicago Computation Institute in Illinois, USA.
Eric K. Nordberg is a Research Scientist and Software Engineer with the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Gary J. Olsen is a Microbiologist with a particular interest in comparative genome analysis at the University of Illinois at Urbana-Champaign in Illinois, USA. Robert Olson is a Senior
Software Engineer in the Computing, Environment and Life Sciences Directorate of Argonne National Laboratory and the Computation Institute at the University of Chicago, in Illinois,
USA.
Ross Overbeek is a Founding Fellow of the Fellowship to Interpret Genomes, as well as Senior Computational Scientist at the Computation Institute, University of Chicago, in Illinois,
USA.
Bruce Parrello is a Research Professional in the Computing, Environment, and Life Sciences Division at Argonne National Laboratory in Illinois, USA. Gordon D. Pusch has a PhD
degree in Physics. He is a member of the Fellowship for Interpretation of Genomes, and is a codeveloper and co-maintainer of the SEED and RAST genome annotation systems.
John Santerre is a PhD candidate in Machine Learning in the Department of Computer Science, University of Chicago in Illinois, USA.
Maulik Shukla is a Senior Software Engineer, Computing in the Environment and Life Sciences, Argonne National Laboratory in Illinois, USA.
Rick L. Stevens is the Associate Laboratory Director for Computing, Environment and Life Sciences Directorate at Argonne National Laboratory and Professor of Computer Science in
the Computation Institute at the University of Chicago in Illinois, USA. Margo Van Oeffelen is a Technical Assistant at the Fellowship for Interpretation of Genomes.
Veronika Vonstein, PhD, is a Founding Fellow and President of the Fellowship for Interpretation of Genomes.
Andrew S. Warren is a Senior Software Architect at the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Alice R. Wattam is a Research Assistant Professor at the Biocomplexity Institute of Virginia Tech, Blacksburg, Virginia, USA.
Fangfang Xia is a Computer Scientist in the Computing, Environment and Life Sciences Directorate of Argonne National Laboratory and a Research Fellow at Computation Institute of
the University of Chicago in Illinois, USA.
Hyunseung Yoo is a Software Engineer at Argonne National Laboratory and the University of Chicago Computation Institute in Illinois, USA. Submitted: 30 April 2017;
Received (in revised form): 13 June 2017
VC The Author 2017. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which
permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
journals.permissions@oup.com
1094
PATRIC as a unique resource | 1095
Ross Overbeek, Bruce Parrello, Gordon D. Pusch, John Santerre,

Maulik Shukla, Rick L. Stevens, Margo VanOeffelen, Veronika Vonstein, Andrew S.
Warren, Alice R. Wattam, Fangfang Xia and Hyunseung Yoo
Corresponding author: Alice R. Wattam, Biocomplexity Institute of Virginia Tech, 1015 Life Science Circle, Blacksburg, VA 24061 USA. Tel.: 540-231-1263;
Fax: 540-231-2606; E-mail: rwattam@vbi.vt.edu
Abstract
Downloaded from https://academic.oup.com/bib/article

The Pathosystems Resource Integration Center (PATRIC, www.patricbrc.org) is designed to provide researchers with the tools and services that they
need to perform genomic and other ‘omic’ data analyses. In response to mounting concern over antimicrobial resistance (AMR), the PATRIC team
has been developing new tools that help researchers understand AMR and its genetic determinants. To support comparative analyses, we have added
AMR phenotype data to over 15 000 genomes in the PATRIC database, often assembling genomes from reads in public archives and collecting their
associated AMR panel data from the literature to augment the collection. We have also been using this collection of AMR metadata to build ma-chine
learning-based classifiers that can predict the AMR phenotypes and the genomic regions associated with resistance for genomes being submitted to
the annotation service. Likewise, we have undertaken a large AMR protein annotation effort by manually curating data from the literature and public
repositories. This collection of 7370 AMR reference proteins, which contains many protein annotations (functional roles) that are unique to PATRIC
and RAST, has been manually curated so that it projects stably across genomes. The collection currently projects to 1 610 744 proteins in the PATRIC
database. Finally, the PATRIC Web site has been expanded to enable AMR-based custom page views so that researchers can easily ex-plore AMR
data and design experiments based on whole genomes or individual genes.
Key words: antimicrobial resistance (AMR); antibiotic; genome annotation; minimum inhibitory concentration; RAST; the SEED
Background [14]. Here, we describe a set of enhancements introduced to support research
-abstract/20/4/1094/4056412 by guest on 15 October 2019

on AMR.
The Pathosystems Resource Integration Center (PATRIC) is one of four
bioinformatics resource centers (BRCs) funded by the National Institute of
Allergy and Infectious Diseases (NIAID) [1]. The BRC program supports
AMR strategy
research by providing access to data associated with the NIAID Category A–
C pathogenic genera [2], with PATRIC serving as the bacterial database. To The current strategy for integrating AMR data into PATRIC breaks down
provide a rich comparative analysis environment, PATRIC provides access to roughly into two parts: (1) data collection to sup-port analyses of whole
all publicly avail-able genomes and associated metadata for bacterial and genomes and (2) data collection to sup-port analyses of individual proteins
archaeal isolates, which includes >104 000 genomes as of June 2017. (Figure 1). In both cases, the data are drawn from the literature as well as a
number of public resources. Specifics on the data integration, curation and
All of the genomes in PATRIC have been consistently annotated using the tools are described below.
Rapid Annotation using Subsystems Technology toolkit (RASTtk) [3, 4]. This
annotation consistency and subsequent pro-tein family generation [5] serve as
the backbone for many of the comparative analysis tools on the Web site [1].
The PATRIC data-base retains the annotations and identifiers from both AMR—integrating data at the genome level
GenBank [6, 7] and RefSeq [8] to facilitate side-by-side comparisons across
Data collection
the public data, allowing researchers to quickly find genomes and genes with
information that they have gathered from different re-sources. PATRIC also To support an environment for comparative analysis, we inte-grate metadata
provides researchers with a private work-space, where they can access associated with the public genomes at GenBank [7] into the PATRIC
bioinformatics services including genome assembly, annotation, RNA database. This makes it easy to build sets of genomes that are based on
sequencing, variation calling, Tn-Seq, similar genome finder, proteome collection date, geographic location, host, isolation source, etc. These
comparison and meta-bolic model reconstruction. When a user annotates a metadata fields are incorpo-rated both from BioSample [15] and directly
private gen-ome with the PATRIC annotation service, they can compare their from the GenBank file when an assembled genome is added to PATRIC. In
genome with the public collection. This ‘virtual integration’ pro-vides a some cases, metadata are acquired first hand from the NIAID-funded genome
unique analysis experience that is not available at a similar scale at any other sequencing centers and from collaborators wishing to make their data public.
data repository. Given the increasing emphasis on re-search to combat AMR and the
decreasing costs of sequencing, we have been able to collect a large number
of genomes with AMR panel data in the form of minimum inhibitory
Facilitating research on antimicrobial resistance (AMR) has become concentra-tions (MICs) or susceptible, intermediate and resistant (SIR) calls
increasingly important with the recent escalation in re-sistance and the loss of [16]. These panel data provide critical context for AMR research by allowing
effectiveness to first-line drugs [9–13]. This resistance has a human cost, with researchers to quickly build data sets for performing
2 million people being sickened and 23 000 dying annually in the United
States alone
10960 | Antonopoulos et al.
on genotypic data. The complete collection of AMR data in PATRIC can be

downloaded from the PATRIC FTP site: ftp.patricbrc.org/pat
ric2/current_release/RELEASE_NOTES/PATRIC_genomes_AMR.txt.
ML classifiers
As the PATRIC database was rapidly accumulating AMR panel data
associated with sequenced genomes, a small number of studies were being
published that explored using ML algorithms to study AMR [22–24]. With a
sufficient number of genomes and AMR panel data, ML algorithms can be
used to predict AMR phenotypes and the genomic regions associated with
AMR with no a priori knowledge of the underlying mechanisms. This is an
appealing area of exploration for PATRIC because it allows us to leverage our

growing metadata collection to predict AMR phenotypes within the
annotation service and to identify AMR-associated genomic regions with
single-nucleotide polymorph-ism (SNP)-level resolution, a feature that can be
used to inform our ongoing manual protein annotation efforts.
In early 2016, we published a study describing the collection of AMR

metadata for genomes and an ML approach that used the AdaBoost algorithm
[25, 26] to build classifiers for predicting AMR [16]. At the time, we had
sufficient data to make predictions in the species Acinetobacter baumannii,
Mycobacterium tuberculosis, Staphylococcus aureus and Streptococcus
pneumoniae for nine antibi-otics [16] (Table 1). Shortly thereafter, we
collaborated with scien-tists at the Houston Methodist Research Hospital to
build classifiers for Klebsiella pneumoniae covering 13 antibiotics using
1777 genomes collected in their hospital system between 2011 and 2015 [27].
Figure 1. PATRIC annotation process for integrating AMR data in both genomic regions and Using the same protocol as described in the Davis et al. [16] and Long et al.
genes. studies [27], we added 18 additional clas-sifiers to the annotation system that
protein and gene comparisons, novel gene discovery, whole-genome variation have not been previously re-ported, including classifiers for M. tuberculosis,
analyses and machine learning (ML) experi-ments (described below). Peptoclostridium difficile, Pseudomonas aeruginosa, S. aureus and S.
To increase the number of genomes with AMR metadata in PATRIC and pneumoniae (Table 1). Receiver operating characteristic (ROC) curves for the
expand our ability to support AMR-based comparative analyses, we began newly added classifiers are shown in Figure 2.
searching the literature for studies that included sequenced bacterial genomes
and AMR panel data. Oftentimes, panel data from these studies were not
recorded in the public archives, so PATRIC becomes the only place, where To date, we have maintained a policy of adding classifiers to the
both the assembled genomes and metadata are available in the same place. If annotation system when their accuracies and F1 scores ex-ceed 70% and their
a genome was assembled and deposited in GenBank [7], we at-tach the AMR top feature k-mers relate to known AMR genes. The classifiers built in this
metadata directly to the corresponding genome in PATRIC. If the reads for a project and described in Table 1 and Figure 2 are integrated into the
genome were deposited in the Sequence Read Archive (SRA) or the annotation service and can be accessed through PATRIC and RAST.
European Nucleotide Archive (ENA) [17, 18], we assemble and annotate the Phenotype predic-tions and the associated genomic regions are available for
genome using PATRIC services [1, 4, 19]. We then incorporate the genome browsing on both Web sites and are described in tutorials at
into the database along with the metadata (Supplementary Document S1). http://tutorial.theseed.org/.
Our AMR metadata collection and classifier building efforts are ongoing
As laboratory methods for determining MIC values vary, incor-porating at PATRIC. In many cases, the AMR metadata avail-able in published studies
these data into PATRIC requires a significant manual cur-ation effort. When report pan-resistant strains, which can be difficult to classify. In an effort to
information is available from the study, we record how the MIC data were improve the accuracy of the classifiers, we are actively seeking strains with
generated, including the laboratory method, the units of the measurement and AMR metadata that improve the biological diversity of the collection. This
the platform that was used to make the measurements. When an assertion includes collecting strains susceptible to many antibiotics. We are also
about a phenotype is provided in the form of a SIR call, we record the comparing the results from several ML methods and are in the process of
laboratory standard from the European Committee on Antimicrobial adding classifiers based on these other methods when they outperform
Susceptibility Testing (EUCAST) [20] or the Clinical and Laboratory AdaBoost [25]. In this manner, an antibiotic and species would be paired with
Standards Institute [21] and the year of the stand-ard. To date, we have the best ML algo-rithm in the annotation system.
attached metadata to PATRIC genomes for 9165 genomes and have
assembled and annotated 6122 gen-omes from SRA and ENA
(Supplementary Table S1). To date, all AMR metadata in PATRIC are
phenotypes that are derived from la-boratory analyses. Studies often assert
the susceptibility or resist-ance of an organism based on the presence or AMR—integrating data at the gene level
absence of key AMR genes. We do not currently incorporate data that are Data collection
only based
Starting in 2015, the PATRIC annotation team, which also main-tains the
SEED [28] and RAST projects [3], began a focused effort
Table 1. AMR classifiers in the PATRIC annotation system
a b b
Species Antibiotic Resistant genomes Susceptible genomes F1 score Initially described in
Acinetobacter baumannii Carbapenem 122 110 0.95 [16]

Klebsiella pneumoniae Amikacin 1190 364 0.92 [27]
Klebsiella pneumoniae Aztreonam 1377 100 0.75 [27]
Klebsiella pneumoniae Cefoxitin 555 976 0.80 [27]
Klebsiella pneumoniae Ciprofloxacin 119 1435 0.91 [27]
Klebsiella pneumoniae Ertapenem 265 178 0.96 [27]
Klebsiella pneumoniae Gentamicin 786 768 0.86 [27]
Klebsiella pneumoniae Imipenem 1100 453 0.94 [27]
Klebsiella pneumoniae Levofloxacin 246 1307 0.93 [27]
Klebsiella pneumoniae Meropenem 1123 430 0.92 [27]

Klebsiella pneumoniae Piperacillin–tazobactam 322 1230 0.76 [27]
Klebsiella pneumoniae Tetracycline 658 896 0.79 [27]
Klebsiella pneumoniae Tobramycin 501 1053 0.94 [27]
Klebsiella pneumoniae Co-trimoxazole 331 1223 0.87 [27]
Mycobacterium tuberculosis Amikacin 210 350 0.91 This study
Mycobacterium tuberculosis Capreomycin 204 350 0.83 This study
Mycobacterium tuberculosis Isoniazid 250 250 0.88 [16]
Mycobacterium tuberculosis Kanamycin 188 250 0.87 [16]
Mycobacterium tuberculosis Ofloxacin 239 250 0.79 [16]
Mycobacterium tuberculosis Rifampicin 250 250 0.86 [16]
Mycobacterium tuberculosis Streptomycin 250 250 0.71 [16]
Peptoclostridium difficile Azithromycin 213 246 0.97 This study
Peptoclostridium difficile Ceftriaxone 228 86 0.86 This study
Peptoclostridium difficile Clarithromycin 213 246 0.99 This study
Peptoclostridium difficile Clindamycin 310 89 0.74 This study
Peptoclostridium difficile Moxifloxacin 188 271 0.97 This study
Pseudomonas aeruginosa Levofloxacin 192 290 0.85 This study
Staphylococcus aureus Ciprofloxacin 467 762 0.98 This study
Staphylococcus aureus Clindamycin 350 274 0.97 This study
Staphylococcus aureus Erythromycin 484 821 0.96 This study
Staphylococcus aureus Gentamicin 162 1144 0.98 This study
Staphylococcus aureus Methicillin 707 886 0.99 [16]
Staphylococcus aureus Penicillin 886 156 0.96 This study
Staphylococcus aureus Tetracycline 203 1029 0.97 This study
Staphylococcus aureus Co-trimoxazole 142 178 0.96 This study
Streptococcus pneumoniae Beta-lactam 2124 584 0.90 [16]
Streptococcus pneumoniae Chloramphenicol 165 289 0.94 This study
Streptococcus pneumoniae Co-trimoxazole 2124 584 0.88 [16]
Streptococcus pneumoniae Erythromycin 381 324 0.96 This study
Streptococcus pneumoniae Tetracycline 368 290 0.96 This study
a
AMR data in PATRIC may be described as individual antibiotics or classes of antibiotics.
b
Used for building the classifiers.
Figure 2. ROC curves for AdaBoost-based AMR classifiers installed in the annotation service since the publication of the Davis et al. [ 16] and Long et al. papers [27]. Accuracy and F1 scores are
displayed in each inset. ROC curves depict classifiers for (A) P. difficile, (B) S. aureus and (C) K. pneumoniae (Kpn), M. tuberculosis (Mtb), P. aerugi-nosa (Pae) and S. pneumoniae (Spn). Antibiotic
abbreviations are: AZM, azithromycin; CC, clindamycin; CIP, ciprofloxacin; CLR, clarithromycin; CRO, ceftriaxone; E, erythromycin; GM, gentamicin; MFX, moxifloxacin; OX, ofloxacin; P,
penicillin; SXT, trimethoprim sulfamethoxazole; TE, tetracycline.
to incorporate and manually curate protein functions relating to AMR. There or general an annotation is, and if it will project cleanly in the annotation
are several well-known consortia that strive to provide standardized system. When reference proteins from the literature create ambiguous
nomenclature for specific groups of antibiotic resistance genes including BLAST matches or split high-similarity clades in the tree, the nomenclature
tetracycline resistance de-terminants [29, 30], and different classes of b- is retained, but then combined into a single annotation that covers the entire
lactamases main-tained by the Lahey Clinic [31], the University of Stuttgart clade. The train-ing sets of representative AMR sequence variants from
[32, 33] and the Institute Pasteur [34]. There are also several well-respected outside sources and the SEED database [28] are then built. They form the
databases that provide collections of AMR genes covering broad categories basis for each AMR-related functional role. An annotation string for each of
of AMR mechanisms including the Comprehensive Antibiotic Resistance the functional roles is assigned, taking into account the SEED database
Database (CARD) [35], the Bacterial Antimicrobial Resistance Reference internal nomenclature conventions as well as those developed by the AMR
Gene Database [36] hosted by the National Center for Biotechnology research community and ac-cepted by CARD, ResFinder, NCBI and other
Information as part of the National Database of Antibiotic Resistant resources. Signature k-mers (amino acid 8-mers) are built from these
Organisms (NDARO) and ResFinder [37]. These resources main-tain functional roles as described previously [4], and the annotations are then pro-
reference sequences for each AMR gene type, providing each with well- jected to all of the genomes in PATRIC. Trees for the newly annotated AMR

curated informative product names (in the case of NDARO) or a specialized proteins are then manually inspected to identify clades that contain multiple
Antibiotic Resistance Ontology (ARO, provided by CARD). These annotations, indicating a lack of consistency. Inconsistencies are also
collections enable accurate detection and annotation of specific AMR identified by comparing the generation of protein families before and after
determinates in pathogen iso-lates by means of supporting the BLAST-based the addition of a new function. The inconsistent proteins are manually re-
[38, 39] or hidden Markov model (HMM)-based [40] screening of user- annotated and this process is iterated until the annotations pro-ject stably and
submitted sequences against representative sets of AMR sequences. However, accurately across the entire database.
in many cases, these AMR annotations project ambigu-ously because newly
discovered proteins can match representa-tive proteins with differing
annotations at nearly equal BLAST similarities. For example, a novel CTX-
M, SHV or TEM b-lacta-mase could potentially present the researcher with The PATRIC manual curation effort offers a variety of add-itional
over a hun-dred nearly equal BLAST hits against highly homologous but benefits to the field of AMR research. For example, this effort is helping to
clinically different reference sequence variants, making the choice of the alleviate the well-documented problem of miss-annotation and over
most appropriate product name difficult. In many cases, the best choice prediction of AMR annotations [43, 44]. We are doing this by systematically
would be a novel allele designation, rather than one of the existing curated removing erroneous an-notations, which implicate non-AMR-related proteins
product names. We believed that a manual curation effort was necessary to with anti-biotic resistance functions, and by annotating and attaching
integrate AMR sequence variants into distinct functional roles (isofunctional literature references to these closely related proteins to prevent over-
protein families, which are integral for the SEED/PATRIC environ-ment) to projection of AMR roles, and then curating their projection over the PATRIC
ensure that they can be unambiguously projected to the genomes in PATRIC collection as described above.
by the annotation service.
We occasionally discover clades of potential AMR proteins that are
surrounded by solid AMR reference sequences, yet have not been described
in any reference database. In these cases, we describe the protein as a
‘putative’ AMR protein of a given re-sistance type, if the sequence identity
levels are 50% or better over the entire length of the protein, which enables
As many resources focus more heavily on the horizontally transferred functional projection. These are obvious targets for characterization in the
AMR genes, we began our curation effort by building functional roles for laboratory. However, if a newly discovered hypothetical clade has a sequence
AMR-related porin and efflux pump proteins described in the literature that identity that is <50%, we use the less specific annotation string for all its
are often chromosomally encoded, reasoning that this would rapidly add new members. In these cases, we use the following annotations: ‘weak similarity
value to the scientific community. Afterward, this naturally led into an effort to aminoglycoside N(6’)-acetyltransferase’ and ‘weak similarity to
to incorporate annotations for proteins involved in tetra-cycline resistance. aminoglycoside N(3)-acetyltransferase’. These are obvious targets for
The proteins involved in efflux pumps are known to play an important role in character-ization in the laboratory. Finally, having clean sets of AMR-related
this type of resistance [41], and there are well-described annotation rules, functional roles facilitates SNP and other comparative analyses at PATRIC
which have been curated by the community for decades for naming them [30, and elsewhere by providing relevant se-quence peer groups for variation
42]. More recently, we have been annotating class by class using publicly research.
available resources when possible.
As of May 2017, the annotation of AMR determinants confer-ring

resistance to tetracycline, b-lactam, aminoglycoside [45, 46],
chloramphenicol [47] and MLSKO (macrolides, lincosa-mides,
Curation process and k-mer projection streptogramins, ketolides and oxazolidinones) [42, 48, 49] antibiotic classes
Significant manual curation and modification of the existing RAST/RASTtk has been completed. These include 450 functional roles for these five major
automatic annotation pipeline were required to accommodate AMR-related antibiotic classes, as well as 36 roles for closely related non-AMR proteins.
functional roles, as their biology dif-fers significantly from ‘classic’ This collection comprises a combined set of 7370 reference and SEED
functional roles encoding pro-karyotic enzymatic and nonenzymatic proteins with AMR roles and 36 424 proteins with related non-AMR roles.
housekeeping functions. The process of creating projectable AMR The collection projects consistently to 1 610 744 AMR proteins with AMR
annotations starts with the incorporation of reference proteins from the roles and 2 518 252 proteins with related non-AMR roles in PATRIC. We
literature and public resources. BLAST searches are used to compare have also associated literature references with the majority of the newly
reference sequences against the SEED database and PATRIC [1]. The sub- curated AMR functional roles in PATRIC, totaling 411 references. The
sequent matching proteins are used to build alignments and trees, which are curation effort is ongoing and is focusing on proteins conveying resistance to
manually inspected to understand how specific quinolone,
Downloaded from https://academic.

Figure 3. Summary information for the antibiotic methicillin at PATRIC. The antibiotic interface provides a summary of the antibiotic, its synonyms and actions, and also provides links via separate tabs
for AMR phenotypes, genes and regions across all the data available in PATRIC.
vancomycin, fosfomycin, rifampin/rifamycin, nitroimidazole, bleomycin and from which the genome has been sequenced. For each taxo-nomic level with
oup.com/bib/article-abstract/20/4/1094/4056412 by guest on 15 October 2019

other antibiotic classes. associated AMR data, PATRIC provides several summaries. A bar graph
summarizing the antibiotics, the AMR phenotype (resistant, intermediate or
susceptible) and the num-ber of genomes that match that phenotype is
Visualization of AMR data at PATRIC available on the overview tab at the top of the main landing page for each
Several new interfaces have been developed on the PATRIC Web site to allow taxon (Figure 4A). Clicking on any of the antibiotics displayed in the graph
researchers to fully explore the AMR data available in the resource. These will open a new page that summarizes all the genomes from that taxon level
interfaces include information that is summarized across all genomes for the that have the particular AMR phenotype. An alternate tabular view of the data
available antibiotics, at the taxon level, and for individual genomes and is also available (Figure 4B). The taxon-level summary page also includes an
genes. Details on each of these interfaces are described below. AMR phenotype tab that lists all of the genomes within the selected taxon
that have an AMR phenotype, and the data that are asso-ciated with it,
including specific treatments, phenotypes or la-boratory methods. All tables
in PATRIC include a dynamic filter for rapid filtering of the genomes based
Antibiotic view
on metadata selections.
Data from PubChem [50] are now integrated for nearly 100 spe-cific
antibiotics that can be viewed on landing pages designed especially to display
this information. Each individual antibiotic has a landing page with several
tabs that provide a general over-view, specific information on the AMR
phenotype, the genes associated with that phenotype and the regions within Gene view and predicted regions associated with AMR
the indi-vidual genes or genomes that are linked to resistance or suscep- phenotypes
tibility to that specific drug (Figure 3). PATRIC provides a summary of data at the gene level, where the physical
characteristics of a gene, its functional role(s), available experimental data
The overview tab includes a general description of the drug, the chemical and associated publications are provided. This view also includes information
structure, the mechanism of action, a description of the pharmacological on homology to genes known to be important in AMR. In addition, PATRIC
activity and class and known synonyms. The AMR phenotype tab provides a provides a view for predicted regions within some genes that are associ-ated
list of all the genomes that have been identified as being susceptible or with AMR phenotypes. The k-mer regions predicted by the ML classifiers are
resistant to that antimicrobial. This tab also includes the laboratory typing visually indicated and their genomic region can be seen on the genome
method and platform, and the testing standard if that informa-tion is browser (Figure 5).
available. A third tab, called AMR genes, displays infor-mation on the genes
associated with resistance. The final tab, AMR regions, includes the location
of the specific k-mers that are associated with the genome’s phenotype.
Future improvements
We continue to peruse resources and publications to identify new genomes
and AMR genes to incorporate into PATRIC. These will be used to expand
Taxon-level view the AMR phenotype predictions and AMR gene analysis to new genera and
PATRIC organizes relevant data for all the available sequenced bacterial and new antibiotics. We plan to map AMR properties to the genus-specific
archaeal genomes according to NCBI taxonomy [51]. Data are summarized at families (PLfams) to support comparative analysis of AMR genes,
each level, from the highest (the Superkingdoms: Bacteria and Archaea) to incorporate new AMR gene trees and allow users to build nucleotide-based
the strain (or isolate)

Figure 4. A taxon-level summary on the PATRIC Web site describing AMR phenotype data across all of the genomes that are part of the Staphylococcus genus. (A) A bar graph summarizes the
antibiotics, the AMR phenotype (resistant, intermediate or susceptible) and the number of genomes that match that phenotype. (B) The AMR phenotype tabular view, which shows all the genomes that
have associated AMR data, includes a dynamic filter for rapid selection of genomes based on the metadata.
Figure 5. AMR predicted regions, located in the genome of S. aureus strain 08S00974, as visualized in the PATRIC JBrowse viewer [ 57]. These predicted regions, numbered sequentially by their
occurrence in the genome as ‘classifier_predicted_regions 12–15’, were predicted by the ML algorithm that is being used to predict AMR pheno-types. The predicted regions are located in and around a
gene (figj1280.11691.peg.56) that is annotated as ‘Tetracycline resistance, MFS efflux pump ¼ > Tet(K)’. The an-notation for this gene came from the focused manual curation effort at PATRIC to
incorporate and propagate information for specific genes that were known to play an important role in AMR.
multiple sequence alignments to identify SNPs and their associ- annotation propagation strategy described above. These include
ation with AMR phenotypes. We are acutely aware that several antibiotic targets, which are largely cellular proteins performing
important types of AMR determinants are not amenable to essential household cellular functions, and such proteins are
being encoded and automatically propagated via the automated grouped into ‘classic’ functional roles in SEED/PATRIC. They
o Benson DA, Cavanaugh M, Clark K, et al. GenBank. Nucleic Acids Res
2017;45:D37.
carry functional annotations that are unrelated to AMR. Antibiotic e
susceptibility in these target proteins is determined by a few, or even a single, p O’Leary NA, Wright MW, Brister JR, et al. Reference sequence (RefSeq) x
non-synonymous mutation in the corresponding gene [52–54]. Likewise, database at NCBI: current status, taxonomic p
single mutations in non-coding DNA regions, including promoters, operators a
and at-tenuators, can lead to dramatic increase in MIC, or an increase in n
resistance levels to particular antimicrobials [55, 56]. These cases will be s
treated separately in PATRIC. We are in the process of designing tools i
specific for SNP detection and analysis tar-geted at the gene level. While o
PATRIC does not currently enable examining AMR data from metagenomes n
or from population-based studies, this is something that we plan to provide in ,
fu-ture releases. a
n
d
f
u
Key Points n
PATRIC includes AMR information at both the genome and gene level, c
and uses manual curation and ML to in-tegrate these data into the ti
annotation service. A large collection of AMR-specific functional roles o
has been manually curated, and this information is propagated by the n
annotation service. With summaries of the available data across all a
taxonomic levels and new interfaces, re-searchers can quickly locate l
and examine these data in their private genomes and compare with the a
n
PATRIC collection.
n
o
t
a
ti
Supplementary Data o
Supplementary data are available online at http://bib.oxford n
journals.org/. .
N
u
Funding c
The NIAID, National Institutes of Health, Department of Health and l
e
Human Services (grant number HHSN272201400027C to R.L.S.).
i
c
A
References c
i
i Wattam AR, Davis JJ, Assaf R, et al. Improvements to PATRIC, the all- d
bacterial Bioinformatics Database and Analysis Resource Center. Nucleic s
Acids Res 2017;45:D535–42. R
j Greene JM, Collins F, Lefkowitz EJ, et al. National Institute of Allergy

e
s
and Infectious Diseases bioinformatics resource cen-ters: new assets for
2
pathogen informatics. Infect Immun 2007; 75:3212–9.
0
1
k Aziz RK, Bartels D, Best AA, et al. The RAST Server: rapid an-notations 6
using subsystems technology. BMC Genomics 2008; 9:1. ;
4
4
l Brettin T, Davis JJ, Disz T, et al. RASTtk: a modular and ex-tensible :
implementation of the RAST algorithm for building custom annotation D
pipelines and annotating batches of gen-omes. Sci Rep 2015;5:8365. 7
3
m Davis JJ, Gerdes S, Olsen GJ, et al. PATtyFams: protein families 3
–
for the microbial genomes in the PATRIC database. Front Microbiol
4
2016;7:118.
5
.
n Clark K, Karsch-Mizrachi I, Lipman DJ, et al. GenBank. Nucleic Acids xi)
Res 2016;44:D67–72.
Wor
l
d
Health Organization. Antimicrobial Resistance. Draft Global Action
xxix) Long SW, Olsen RJ, Eager TN, et al. Population genomic ana-
Plan on Antimicrobial Resistance. Geneva: WHO, 2015.
lysis of 1,777 extended-spectrum beta-lactamase producing Klebsiella
xii) Eurosurveillance Editorial Team. WHO member states adopt pneumoniae, Houston, Texas: unexpected abun-dance of clonal group
global action plan on antimicrobial resistance. Euro Surveill 2015;20. 307. mBio, vol. 8, 2017.
xiii) Fauci AS, Collins FS. New strategies in battle against antibi-
otic resistance. https://directorsblog.nih.gov/2014/09/18/ new-strategies-
in-battle-against-antibiotic-resistance/.
xiv) Roca I, Akova M, Baquero F, et al. The global threat of anti-
microbial resistance: science for intervention. New Microbes New
Infect 2015;6:22–9.
xv) Chen L. Notes from the field: pan-resistant New Delhi metallo-
beta-lactamase-producing Klebsiella pneumoniae— Washoe County,

Nevada, 2016. MMWR Morb Mortal Wkly Rep 2017;66:33.
xvi) Centers for Disease Control and Prevention. Antibiotic resist-

ance threats in the United States, 2013. Centres for Disease Control and
Prevention, US Department of Health and Human Services, 2013.
xvii) Barrett T, Clark K, Gevorgyan R, et al. BioProject and

BioSample databases at NCBI: facilitating capture and organ-ization of
metadata. Nucleic Acids Res 2012;40:D57–63.
xviii) Davis JJ, Boisvert S, Brettin T, et al. Antimicrobial resistance
prediction in PATRIC and RAST. Sci Rep 2016;6:27930.
xix) Kodama Y, Shumway M, Leinonen R. The Sequence Read
Archive: explosive growth of sequencing data. Nucleic Acids Res
2012;40:D54–6.
xx) Leinonen R, Akhtar R, Birney E, et al. The European

Nucleotide Archive. Nucleic Acids Res 2010;gkq967.
xxi) Bankevich A, Nurk S, Antipov D, et al. SPAdes: a new genome
assembly algorithm and its applications to single-cell sequencing. J
Comput Biol 2012;19:455–77.
xxii) European Committee on Antimicrobial Susceptibility Testing.
EUCAST guidelines for detection of resistance mech-anisms and
specific resistances of clinical and/or epidemiolo-gical importance.
EUCAST, Basel, Switzerland, 2013. http:// www/eucast.
org/clinical_breakpoints.
xxiii) Patel J, Cockerill F, Alder J, et al. Performance standards for

antimicrobial susceptibility testing; twenty-fourth informa-tional
supplement. In: CLSI Standards for Antimicrobial Susceptibility
Testing. Clinical and Laboratory Standards Institute, Wayne, PA, vol.
34, 2014, 1–226.
xxiv) Bradley P, Gordon NC, Walker TM, et al. Rapid antibiotic-
resistance predictions from genome sequence data for Staphylococcus
aureus and Mycobacterium tuberculosis. Nat Commun 2015;6:10063.
xxv) Drouin A, Gigue` re S, Sagatovich V, et al. Learning interpret-

able models of phenotypes from whole genome se-quences with the Set
Covering Machine. arXiv, preprint arXiv:1412.1074 [q-bio.GN], 2014.
xxvi) Santerre JW, Davis JJ, Xia F, et al. Machine learning for anti-
microbial resistance. arXiv:1607.01224 [stat.ML], 2016.
xxvii) Freund Y, Schapire RE, A decision-theoretic generalization of
on-line learning and an application to boosting. In: European
Conference on Computational Learning Theory. Springer, 1995, 23–37.
xxviii) Freund Y, Schapire R, Abe N. A short introduction

to boosting. J Jpn Soc Artif Intell 1999;14:1612.
D, et al. The SEED and the Rapid Annotation of microbial genomes using Subsystems Technology (RAST). Nucleic Acids Res 2014;42:D206–
14.
) Ov (J) Levy SB, McMurry LM, Barbosa TM, et al. Nomenclature for new tetracycline resistance determinants. Antimicrob Agents Chemother
er 1999;43:1523–4.
be (K) Chopra I, Roberts M. Tetracycline antibiotics: mode of action, applications, molecular biology, and epidemiology of bacter-ial resistance.
ek Microbiol Mol Biol Rev 2001;65:232–60. second page, table of contents.
R, (L) Bush K, Pazkill T, Jacoby J. ß-lactamase classification and amino acid sequences for TEM, SHV and OXA extended-spectrum and
Ol inhibitor resistant enzymes. http://www.lahey.org/Studies/.
so
(M) Thai QK, Bo¨ s F, Pleiss J. The lactamase engineering database: a critical survey of TEM sequences in public databases. BMC Genomics
n
2009;10:390.
R,
Pu
(N) Fischer M, Thai QK, Grieb M, et al. DWARF–a data warehouse sys-tem for analyzing protein families. BMC Bioinformatics
sc 2006;7:495.
h (O) Pasteur I. Klebsiella sequence typing. http://bigsdb.pasteur.fr/ klebsiella/klebsiella.html.
G (P) McArthur AG, Waglechner N, Nizam F, et al. The comprehen-sive antibiotic resistance database. Antimicrob Agents Chemother
2013;57:3348–57.
(Q) NCBI. Bacterial antimicrobial resistance reference gene data-base, 2017. https://www.ncbi.nlm.nih.gov/bioproject/?term¼31 30472017.
(R) Zankari E, Hasman H, Cosentino S, et al. Identification of acquired antimicrobial resistance genes. J Antimicrob Chemother
2012;67:2640–4.
(S) Boratyn GM, Camacho C, Cooper PS, et al. BLAST: a more effi-cient report with usability improvements. Nucleic Acids Res
2013;41:W29–33.
(T) Madden T. The BLAST sequence analysis tool. In: The NCBI Handbook [Internet], 2nd ed., National Center for Biotechnology
Information, Bethesda, MD, 2013.
(U) Haft DH, Selengut JD, Richter RA, et al. TIGRFAMs and genome properties in 2013. Nucleic Acids Res 2013;41:D387–95.
(V) Sun J, Deng Z, Yan A. Bacterial multidrug efflux pumps: mechanisms, physiology and pharmacological exploitations. Biochem Biophys
Res Commun 2014;453:254–67.
(W) Roberts MC, Sutcliffe J, Courvalin P, et al. Nomenclature for macrolide and macrolide-lincosamide-streptogramin B resistance
determinants. Antimicrob Agents Chemother 1999; 43:2823–30.
nham N, Garavelli JS, Apweiler R, et al. Missing in action: enzyme functional annotations in biological databases. Nat Chem Biol 2009;5:521–
5.
*
F Schnoes AM, Brown SD, Dodevski I, et al. Annotation error in public databases: misannotation of molecular function in en-zyme superfamilies.
u PLoS Comput Biol 2009;5:e1000605.
*
r Ramirez MS, Tolmasky ME. Aminoglycoside modifying en-zymes. Drug Resist Updat 2010;13:151–71.
*
Shaw KJ, Rather PN, Hare RS, et al. Molecular genetics of ami-noglycoside resistance genes and familial relationships of the aminoglycoside-
modifying enzymes. Microbiol Rev 1993; 57:138–63.
*
van Hoek AH, Mevius D, Guerra B, et al. Acquired antibiotic re-sistance genes: an overview. Front Microbiol 2011;2:203.
*
Roberts MC. Update on macrolide-lincosamide-streptogramin, ketolide, and oxazolidinone resistance genes. FEMS Microbiol Lett
2008;282:147–59.
*
Roberts MC. Nomenclature for tetracycline genes/nomencla-ture center for MLS genes, 2017. http://faculty.washington. edu/marilynr/2017.
*
Kim S, Thiessen PA, Bolton EE, et al. PubChem substance and compound databases. Nucleic Acids Res 2015;4:44.
*
Federhen S. Type material in the NCBI taxonomy database. Nucleic Acids Res 2015;43:D1086–98.
*
Maness MJ, Sparling PF. Multiple antibiotic resistance due to a single mutation in Neisseria gonorrhoeae. J Infect Dis 1973;128: 321–30.
*
Mac Aogain M, Kilkenny S, Walsh C, et al. Identification of a novel mutation at the primary dimer interface of GyrA con-ferring
fluoroquinolone resistance in Clostridium difficile. J Glob Antimicrob Resist 2015;3:295–9.
*
Santos-Lopez A, Bernabe-Balas C, Ares-Arroyo M, et al. A naturally occurring single nucleotide polymorphism in a multicopy plasmid
produces a reversible increase in antibi-otic resistance. Antimicrob Agents Chemother 2017;61: e01735–16.
*
Martinez J, Baquero F. Mutation frequencies and antibiotic re-sistance. Antimicrob Agents Chemother 2000;44:1771–7.
*
Suzuki S, Horinouchi T, Furusawa C. Prediction of antibiotic resistance by gene expression profiles. Nat Commun 2014;5.
*
Buels R, Yao E, Diesh CM, et al. JBrowse: a dynamic web plat-form for genome visualization and analysis. Genome Biol 2016; 17:66.
Downloaded by INSEAD At 17:40 05 December 2018 (PT)
Industrial Management & Data Systems
The effect of e-service quality on Jordanian student’s e-loyalty: an
empirical study in online retailing
Rami Mohammad Al-dweeri, Antonia Ruiz Moreno, Francisco Javier Llorens Montes, Zaid
Mohammad Obeidat, Khaldoon M. Al-dwairi,
Rami Mohammad Al-dweeri, Antonia Ruiz Moreno, Francisco Javier Llorens Montes, Zaid
Mohammad Obeidat, Khaldoon M. Al-dwairi, (2018) "The effect of e-service quality on Jordanian
student’s e-loyalty: an empirical study in online retailing", Industrial Management & Data
Systems, https://doi.org/10.1108/IMDS-12-2017-0598
https://doi.org/10.1108/IMDS-12-2017-0598
Downloaded on: 05 December 2018, At: 17:40 (PT)
References: this document contains references to 104 other documents.
The fulltext of this document has been downloaded 2 times since 2018*
Access to this document was granted through an Emerald subscription provided by
emerald-srm:178665 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The
company manages a portfolio of more than 290 journals and over 2,350 books and book series
volumes, as well as providing an extensive range of online products and additional customer
resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative
for digital archive preservation.

The effect of e-service quality onEffectquality

of e-service
Jordanian student’s e-loyalty: an

empirical study in online retailing
Rami Mohammad Al-dweeri Received 27 December
Department of Marketing, University of Jordan, Amman, Jordan 2017 Revised 24 June
Antonia Ruiz Moreno and Francisco Javier Llorens Montes 2018 3 September 2018
Accepted 31 October 2018
Department of Business Organization, Universidad de Granada, Granada, Spain
Zaid Mohammad Obeidat
Department of Marketing, University of Jordan, Amman, Jordan, and
Khaldoon M. Al-dwairi
Department of Library and Information Management,
Al-Balqa’ Applied University, Al-Salt, Jordan
Abstract
Purpose – The purpose of this paper is to examine the dimensions of e-service quality, its effects on e-
satisfaction and e-trust and its impact on behavioural and attitudinal loyalty in Jordanian youth users of
online retailing.
Design/methodology/approach – The scale proposed in this study has been specifically constructed using
the four models most recognised for measuring e-service quality: E-S-QUAL, WebQual, eTransQual and
eTailQ. The dimensions used in this study are efficiency, privacy, reliability, emotional benefit and
customer service. The research model was statistically tested by students in Jordan, using Amazon.com.
Findings – It is found that privacy, reliability, emotional benefit and customer service are important
elements to measure the e-service quality, but efficiency is not. E-trust was found to be an antecedent of e-
satisfaction, and behavioural loyalty an antecedent of attitudinal loyalty.
Originality/value – The contribution of this study is thus the investigation of the causal relationship
between the e-service quality dimensions, e-satisfaction, e-trust, behavioural loyalty and attitudinal loyalty,
where it is necessary to consider the subject in more depth and to examine e-service quality dimensions
based on a proposed model constructed from the four most common models.
Keywords E-satisfaction, Attitudinal loyalty, Behavioural loyalty, E-trust, E-service quality, Online
retailing Paper type Research paper
1. Introduction
With the fast growth of the internet and e-commerce, more customers nowadays engage in
online shopping. As a result, firms depend on online marketing to sell their products through
websites, and to obtain a competitive advantage (Venkatesh et al., 2012). Over the last years,
many retailers have added an online channel so becoming multichannel retailers (Frasquet et
al., 2017). In recent years, service quality and e-service quality have been among the most
important research areas in marketing due to their apparent influence on the financial
performance of the firm (Kim and Lennon, 2017; Zahedifard et al., 2014), in addition to
gaining a competitive advantage through superior electronic services (Yarimoglu, 2015).
Generally, the literature implies that the greater the e-service quality, the greater the
electronic customer satisfaction (Hsu and Hsu, 2008; Ou et al., 2011). In addition, it was found
that e-service quality also increases customers’ e-trust in the service provider (Mohd Kassim
and Ismail, 2009). Moreover, considering that e-trust was found to be a main predictor of Industrial Management & Data
Systems
© Emerald Publishing Limited
0263-5577
This research was supported by Erasmus Mundus. The authors thank the Dunia Beam project, which DOI 10.1108/IMDS-12-2017-0598
provided financial support that greatly assisted the research.

IMDS online shopping (e.g. Cristobal et al., 2007), significant links were also found
between it and e-loyalty (e.g. Frasquet et al., 2017; Chung and Shin, 2010;
Kantsperger and Kunz, 2010). However, the causal relationship between e-service
quality, e-satisfaction, e-trust and e-loyalty has also been debated in recent years.
Previous studies showed that e-service quality did not have a consistent impact on e-
loyalty (Ou et al., 2011; Ghane et al., 2011; Ho and Lee, 2015; Kao and Lin, 2016;
Swaid and Wigand, 2009; Wolfinbarger and Gilly, 2003; Yaya et al., 2011), which
implies that this relationship could be influenced by other variables. As a result, some
studies analysed the relationship between the four constructs (e-service quality, e-
satisfaction, e-trust and e-loyalty). However, Ho and Lee (2015) discussed this
relationship but, instead of e-service quality, they examined information quality,
while Ghane et al. (2011) and Ziaullah et al. (2014) focussed on this relationship
without considering the two kinds of loyalty, behavioural and attitudinal.
Consequently, there is a lack of empirical studies examining the causal relationship
between e-service quality, e-satisfaction, e-trust and e-loyalty all together. Although
previous links have been found between these constructs and e-loyalty individually
(Ghane et al., 2011), a comprehensive examination of all the direct and indirect
effects that e-service quality, e- satisfaction and e-trust have on e-loyalty and on each
other is needed.
In addition, limited research attention has been given to examining the influence of this
sequence on different types of loyalty (Ziaullah et al., 2014). Generally, customer loyalty
is the definitive aim of relationship marketing, and particularly in electronic transactions,
to avoid customers switching to other companies (Cheng, 2011). In addition, loyalty will
encourage customers to pay higher prices and spread positive word of mouth about the
firm (Dowling and Uncles, 1997; Romadhoni et al., 2015). Fandy and Chandra (2008)
note that customer e-loyalty is a dynamic concept that is influenced by a number of
factors, one of which is the e-service quality. However, when examining the influence of
the e-service quality on e-loyalty, all of the previous studies considered e-loyalty to be a
single construct, with no regard to the different types of e-loyalty (e.g. Ho and Lee, 2015;
Kao and Lin, 2016). Consequently, there is another gap in the literature in examining this
causal relationship and taking into consideration those two aspects of e-loyalty:
behavioural and attitudinal. The literature has highlighted important differences between
these two types of loyalty that can have important implications for businesses (Russell et
al., 2007). For example, behavioural loyalty often leads the customer to make repeat
purchases from the firm and its website (Cheng, 2011). On the other hand, attitudinal
loyalty does not really mean that the customer actually purchases from the website,
despite having a positive attitude towards it (Kang et al., 2015). In this case, the customer
could feel positively about the brand or the website, leading to positive word of mouth and
referrals to friends; however, actual behavioural loyalty could be hindered due to a lack of
ability to afford the brand or to shop there themselves. As a result, in many cases, one type
of loyalty could exist without the other (Russell et al., 2007). Therefore, online companies
must take into account not only that users revisit websites, but also that they recommend
them to other users, and that this is gaining more importance, especially in the era of
social media.
The contribution of this study is thus the investigation of the causal relationship
between the e-service quality dimensions (efficiency, privacy, reliability, emotional benefit
and customer service), e-satisfaction, e-trust, behavioural loyalty and attitudinal loyalty,
where it is necessary to consider the subject in more depth and to examine e-service
quality dimensions based on a proposed model constructed from the four most common
models. Consequently, this comprehensive examination of the e-service quality will allow
for a better examination of these factors and their direct and indirect relationships.
Moreover, while the concept of service quality has been examined in a number of
contexts, limited
research
attention was
given to
examining it in
the e-retailing
context (Ou et
al., 2011; Ghane
et al., 2011) and
in the middle
east context (Al-
dweeri et al.,
2017). Because
customers of
electronic
services are harder to please and retain (Kao and Lin, 2016). This examination would Effect of
allow for a better examination of the main factors that would encourage their satisfaction e-service
and loyalty. Thus, increasing the sales volume and market share, and providing
quality
opportunities for cross-selling and improving customer retention for electronic service
providers (Gera et al., 2017). The purposes of this study are to:
← establish the dimensions (efficiency, privacy, reliability, emotional benefit and
customer service) that affect e-service quality in the Jordanian context; and
← examine the relationships between e-service quality dimensions, e-satisfaction, e-
trust, behavioural loyalty and attitudinal loyalty in the Jordanian context.
q Literature review
Academic studies of electronic services have sparked intense debate in recent years. Two
of the main areas of analysis have been the quality of service, as a multidimensional
construct, and the development of various measurement scales. Previous reviews have
been published of the concept and measurement models of service quality in general
(Manuel, 2015; Salameh and Bin Hassan, 2015), the factors that impact the user
experience of electronic services (Rowley, 2006), the marketing of services through the
internet (Parasuraman and Zinkhan, 2002), satisfaction with electronic services (Chen et
al., 2008) and the causal relationship between electronic service quality, e-satisfaction and
e-loyalty (Águila et al., 2013; Giovanis and Athanasopoulou, 2014).
There is no unanimous agreement about the concept of e-service quality, and therefore there
is no definition that is widely accepted among researchers. The sub-field of e-service quality
under study also differs from one article to another. Some studies focus on the quality of
websites (Loiacono et al., 2007), while others take a broader view of the quality of electronic
services, including the personal contact with customer service (Yang et al., 2003). Rolland and
Freeman (2010) propose a broad definition of e-service quality from the business to customer
(B2C) perspective, suggesting that it is the excellence of customer service throughout the
buying process, from initial contact to delivery of the service or product. It may be noted that
this definition does not included after-sales. Zeithaml et al. (2002) defined e-service quality as
the degree to which a website facilitates effective purchases and deliveries. This definition
refers to the ability of a website to meet the needs of customers without problems. Santos
(2003), from an online marketing environment, defines e-service quality in e-commerce as the
comprehensive assessment of excellence and the quality of electronic services offered in virtual
markets. This definition includes the complete experience of a visit to a website. Taking this
approach, we can propose a definition of e-service quality in mathematical form as follows: the
e-quality of service is the difference between the value of service perceived and the value of
customer expectations, as seen in the following equation: eSQ ¼ Vsp−Vce. This equation
represents the e-service quality of a particular transaction.
The development of the scale proposed in this study for measuring e-service quality was
specifically the result of a literature review (see Figure 1), and has been constructed from the four
models most recognised for measuring e-service quality, which are E-S-QUAL (Parasuraman et al.,
2005), WebQual (Loiacono et al., 2007), eTransQual (Bauer et al., 2006) and eTailQ (Wolfinbarger
and Gilly, 2003). These models were examined in several studies, but there was no agreement on
which dimensions measure e-service quality and the results varied from one study to another
(Casielles et al., 2009; Ha and Stoel, 2009; Shia et al., 2016; Zehir et al., 2014).
In our study, after a detailed review of the dimensions of those models, we grouped them
into five dimensions. This study will focus on the five dimensions of (efficiency, privacy,
reliability, emotional benefit and customer service), as will be explained in more detail in the
next section. However, in our study we chose these five dimensions for two main reasons.
IMDS
Efficiency
Behavioural
E-satisfaction
Privacy loyalty
Reliability
Emotional
benefit
Attitudinal
E-trust loyalty
Customer
Figure 1. service
Proposal model
First, as seen in Table I, these were the most studied and influential factors in the e-service
quality literature (Loiacono et al., 2007; Parasuraman et al., 2005). Second, all of our five
dimensions are constructs that encompass a variety of factors related to e-service quality –
for example, efficiency and privacy are often related to aspects such as the information
content, ease of use and security of a website (Zeithaml et al., 2002); reliability is mainly
related to compliance, which is one of the strongest predictors of e-satisfaction
(Wolfinbarger and Gilly, 2003); and emotional benefit is tied to the visual and emotional
appeal of the website. Finally, the customer service dimension is also related to variables
such as communication, customization, responsiveness, compensation and contact, all of
which were found to be related to the e-service quality (Loiacono et al., 2007).
Considering that customers perceive the service quality of a website as the overall
excellence of the website (Parasuraman et al., 2005), the overall excellence of a website
should be considered in terms of design, technical and emotional factors.
xxx) Theoretical background and research hypotheses 3.1

E-service quality
Electronic services quality is becoming important not only for the success or failure of e-
commerce, but also in terms of providing consumers with an internet experience that is
more than the mere interactive flow of information (Santos, 2003). E-services add value to
the exchange of information that the internet provides.
Models
Proposal model E-SERVQUAL WebQual eTransQual eTailQ
Efficiency X X X X
Privacy X X X
Table I. Reliability X X X
Dimensions of Emotional benefit X X
e-service quality Customer service X X X
Since the beginning of this century, the literature has begun to include studies that have aimed Effect of
to develop scales to measure perceived e-service quality. Parasuraman et al. (2005), building on e-service
the previous research of Zeithaml et al. (2002), developed a multi-item scale to measure the
quality of electronic services. This scale was called E-S-QUAL, and is one of the most
quality
comprehensive available today. Its development was based on the realisation of an exploratory
study to determine the need for two different scales to assess the quality of electronic services.
The E-S-QUAL scale takes into account aspects related to basic or primary service and
distinguishes four dimensions: efficiency, compliance, system availability and privacy.
Another model, WebQual, was developed by Loiacono et al. (2007), with a scale created
following a literature review and models for measuring the quality of existing e-services. The
WebQual scale was originally conceived as an instrument to assess perceptions of quality in e-
marketing that took place through websites. There were 12 dimensions used in this model, grouped
in turn into four sub-dimensions: usefulness (information fit-to-task, interactivity, trust, response
time); ease of use (ease of understanding, intuitive operations); entertainment (visual appeal,
innovativeness, flow, emotional appeal); and complementary relationship (consistent image, online
completeness and better than alternative channels). Bauer et al. (2006) developed a model of
transaction processing for electronic services called eTransQual, which includes all stages of the
delivery of electronic services. After conducting an analysis of exploratory and confirmatory factors,
five qualitative dimensions were identified: responsiveness, reliability, process, functionality/design
and enjoyment.
Another scale measuring e-service quality is the eTailQ model, developed by Wolfinbarger
and Gilly (2003), who were trying to determine the dimensions of e-service quality related to
internet retailing. The authors distinguish four dimensions: fulfilment/ reliability, website
design, privacy/security and customer service. The eTailQ scale ignored aspects such as
entertainment.
The first dimension of the scale of this study is efficiency. During the literature review of studies
that had used this dimension to measure e-service quality, it was noted that Zeithaml et al. (2002)
include information content, design, updating the website and ease of use in this dimension, as do
Parasuraman et al. (2005) for e-SERVQUAL and E-S-QUAL. Aspects related to information content
and updating the website are frequently cited as determining factors in deciding whether or not to
purchase online (Li et al., 2002; Wolfinbarger and Gilly, 2003). The information that consumers look
for is diverse, including information about service, after-sales and customer service (Aladwani,
2002). Another aspect to consider is related to the design of the website, which should be well
structured and organised. Fassnacht and Koese (2006) distinguish between graphic quality and
clarity of design, the former visually representing various elements of the user interface (text, icons,
digital images or website background) and the latter defined as the structural design of the user
interface that helps customers to find the information they need.
The structure of the website and ease of use reduce customer costs for searching and information
processing, and also decrease the time spent on the purchase (Lee and Lin, 2005). Cristobal et al.
(2007) confirmed that good design helps the client to make a quick navigation by providing
structured information. Swaid and Wigand (2009) confirmed that the information has positive effects
on e-satisfaction, while Herington and Weaven (2009) noted that although efficiency is found to be
most important overall, and rated highest by respondents, it has the least impact upon satisfaction.
Efficiency has a positive impact on customer trust, however (Kao and Lin, 2016; Kim et al., 2009).
Hansen and Jonsson (2013) found that efficiency has no positive impact on e-trust. The following
hypotheses have thus been developed:
H1a. Efficiency is positively related to customer satisfaction online.
H1b. Efficiency is positively related to customer trust online.
IMDS The second hypothesis concerns privacy, which is the dimension most studied in the field of e-
service quality. Numerous investigations have focussed on the analysis of the particularities of
privacy features in relationships over the internet (Zeithaml et al., 2002; Yang et al., 2003).
This is due to the insecurity of consumers when making transactions through this medium.
Generally, when online customers are unable to interact with employees or the physical
facilities of a firm, then trust must be established in another way. Some issues have prevented
electronic commerce prospering completely, the most obvious being the lack of confidence and
insecurity perceived in making or receiving a payment via the internet (Bush and Kraft, 1998).
Several websites have thus been developed and have deployed consumer privacy policies to
protect users from the risk of fraud and financial loss as a result of using credit card or other
financial information (Malhotra and Segars, 2005). Confidence in an online context more than
ever affects consumer willingness to be vulnerable to a company, in the belief that the company
will fulfil its promises and not exploit that vulnerability to their advantage (Ranaweera et al.,
2005). The most tangible aspect of the privacy of a website is its privacy policy, which
describes how personal information is managed to provide services on the website.
There is no unanimous agreement among researchers when considering the role played by the
privacy evaluation of a website. Some authors have found that privacy does not exert a significant
influence on perceptions of e-service quality (Fassnacht and Koese, 2006; Wolfinbarger and Gilly,
2003), or on satisfaction with the website (Cristobal et al., 2007; Kim and Stoel, 2004). The results
obtained by Kim et al. (2009) suggest that privacy is a driver of e-trust, which in turn leads to
customer e-satisfaction. Ribbink et al. (2004) found a high impact of privacy on customer trust in the
online environment. The following hypotheses are thus formulated:
H2a. Privacy is positively related to customer satisfaction online.
H2b. Privacy is positively related to customer trust online.
The third hypothesis refers to reliability, and the process between the ordering and delivery of
goods or services. For some authors, the term “reliability” is synonymous with compliance
(Wolfinbarger and Gilly, 2003). To provide perfect performance of the service, the company
should consider several aspects, such as the delivery of orders on time, fulfilment of the
characteristics indicated on the website and shipping processes, which have to be fast and
accurate (Nath and Zheng, 2004). This dimension makes reference to the delivery of goods
and/or services on time, and that no errors occur during transactions (Wolfinbarger and Gilly,
2003). Consumers should also be able to return an item if they do not find it to their liking (Ho
and Lee, 2015). The progress of the order should be known by consumers in all its stages. For
example, some websites provide a service that informs customers about the status of their order
via e-mail or SMS. This dimension has been considered, in some studies, the most important
factor in e-service quality. For example, Wolfinbarger and Gilly (2003) and Zeithaml et al.
(2002) show that reliability is the most powerful indicator in customer satisfaction in an online
context. Yen and Lu (2008) confirm that reliability is positively related to satisfaction and that
satisfaction is positively associated with loyalty. Lee and Lin (2005) suggest that reliability is a
significant predictor of customer satisfaction in online shopping. Swaid and Wigand (2009)
show the influence of reliability on a customer’s intention to return to the website and make a
repeat purchase. Luarn and Lin (2003) demonstrate that commitment plays a crucial
intervening role in the relationship between customer satisfaction and perceived loyalty value.
Other studies have also confirmed that reliability is the strongest predictor of both e-
satisfaction and e-trust (Giovanis and Athanasopoulou, 2014; Kim et al., 2009). In the online
retailing context, consumers need to be assured that their products will be delivered as
expected. Therefore, we hypothesise that:
H3a. Reliability is positively related to customer satisfaction online.
H3b. Reliability is positively related to customer trust online.
The dimension of emotional benefit can refer to visual appeal, innovation, emotional appeal, Effect of
the image (Loiacono et al., 2007) or simply whether something is attractive (Cao et al., 2005). e-service
As mentioned previously, perceived service quality is defined as the judgement of consumers
regarding the overall excellence of a website (Parasuraman et al., 2005). When consumers
quality
assess a website, they refer to their feelings as well as their cognition in the process of service
consumption. The feeling of entertainment can have a positive influence on the evaluation of a
service if it is also able to meet customer needs. According to Wirtz et al. (2000), consumers
remember the feelings experienced during consumption.
Bauer et al. (2006) studied emotional benefit in their model (eTransQual), with sub-
criteria such as service customization, fun to use the website, online shopping enthusiasm
and enjoyment provided by the website. They confirm that emotional benefit is a key
factor influencing both the length of a relationship and purchase intention, and is a factor
in the experience of the first visit that encourages customers to revisit the website. Along
the same lines, the authors confirm that entertainment during navigation has a positive
influence on e-service quality. Otherwise, Ab Hamid and McGrath (2005) suggest that
emotional benefit has a small effect on e-service quality and e-satisfaction. Therefore:
H4a. Emotional benefit is positively related to customer satisfaction online.

H4b. Emotional benefit is positively related to customer trust online.
The last criterion of the proposed model is customer service. Customer service is the ability of a
website to maintain relationships with customers when problems occur in transactions, and the
ability to provide various modes of communication with customers. Communication is defined
as keeping customers adequately informed and using a language they can understand (Santos,
2003). Communication in electronic services may be online communication (e-mail or chat) or
traditional communication (telephone, fax and regular mail). Gounaris and Dimitriadis (2003)
confirm that customer service is an important factor in measuring e-service quality, considering
how easily customers can communicate with the staff of a customer service department and
receive answers quickly to requests and e-mails. This same perception is shared by
Wolfinbarger and Gilly (2003), who consider customer service as the level of response or
support, and the desire to respond quickly to a customer when a question is asked. Cox and
Dale (2001) note that customer service adds value to the customer experience and helps to build
relationships with them, offering additional services and information. According to these
authors, it possible to establish long customer relationships by offering a customised service,
encouraging frequent purchases and offering services that add value.
Gounaris et al. (2010) considered customer service as an important dimension when
measuring e-service quality, and the same perception is shared by Wolfinbarger and Gilly
(2003). Zeithaml et al. (2002) consider customer service to be a dimension of the first
order in their scale, which refers to the ability to offer quick responses and to provide help
if there is any type of problem. Parasuraman et al. (2005) changed this traditional view,
demonstrating that a second scale, E-RECs-QUAL, containing three dimensions –
responsiveness, compensation and contact – focussed on dealing with problems and
complaints. Lee and Lin (2005) found customer service to affect e-satisfaction mildly. The
following hypothesis is proposed, related to this dimension:
H5a. Customer service is positively related to customer satisfaction online.
H5b. Customer service is positively related to customer trust online.
3.2 E-satisfaction, e-trust and e-loyalty

The conceptual and operational definitions of e-satisfaction are inconsistent in the literature.
Authors use several concepts to refer to satisfaction: Oliver (1997) uses the term “emotion”,
IMDS Wang and Huarng (2002) use the term “affective response”, and Anderson and Srinivasan
(2003) refer to “experience”. Hassan and Herrero (2006) suggest that if a user does not
achieve their objectives, or if a website does not meet their needs, they will simply
abandon it and look for an alternative. We can thus say that customer satisfaction starts
from the entrance point of a website.
One of the purposes of this study is to test the causal relationship between the e-satisfaction,
e-trust and e-loyalty constructs, which has not yet been determined very clearly in the
literature. The literature emphasises the importance of achieving customer satisfaction to
encourage long-term relationships (B2C), and studies its impact on purchase intention and
commitment (Kim et al., 2017; Herington and Weaven, 2009; Lee and Lin, 2005; Al-dweeri et
al., 2017). The relationship between satisfaction and loyalty seems almost intuitive, as several
researchers have tried to confirm in their research (Kim et al., 2009).
Cox and Dale (2001) confirmed that e-service quality can be a way to gain competitive
advantage through customer satisfaction in e-commerce, while Wolfinbarger and Gilly
(2003) argued that perceived quality is the second most important determinant when it
comes to generating customer loyalty to a website and that the first is e-satisfaction.
Thaichon and Quach (2015) noted that their results demonstrate that satisfaction and
commitment are significantly related to behavioural loyalty. The effects of trust,
satisfaction and commitment on attitudinal loyalty are also confirmed.
The definition of e-loyalty, according to Anderson and Srinivasan (2003), is a
customer’s favourable attitude towards an electronic business, resulting in repeat buying
behaviour. Cristobal et al. (2007) argue that achieving electronic loyalty requires a service
quality that satisfies the customer. According to Gommans et al. (2001), e-loyalty has
parallelism with the website fidelity concept, regarding the generation of buying
behaviour and revisit. Smith and Rupp (2002) suggest that the most important objective
for any website is to maximise the loyalty of the end users and the time value of visit and
future purchases. According to those authors, the revisit has the same importance for a
website success as is the fidelity to the purchase transaction.
In this sense, one can say satisfaction mediates the other two constructs in this
relationship. In this study, we test the relationship between e-satisfaction and the two
kinds of e-loyalty: behavioural, in which customers recommend a website to others, and
attitudinal, in which customers make repeat purchases from the same online provider.
The concept of trust is one of the most important elements leading to the creation of
long-term relationships with customers, especially in maintaining the confidentiality of
information relating to clients, and a commitment to providing the best service/product
over time. The concept of trust is generally formed between an organisation and its
customers through: efficiency, or competence in providing the service, and attention to
every client’s interest (Singh and Sirdeshmukh, 2000). Chu (2009) defines trust as a
human characteristic that is based on the assessment of another’s personality traits.
There are several definitions of trust in the online environment as a set of distinct
beliefs, consisting of integrity, benevolence and ability (Gefen et al., 2003). According to
Yoon (2002), e-trust is security assurance, reputation, web searching, fulfilment (e.g.
willingness to customise), presentation (e.g. web quality), technology and interactions
(e.g. e-forums). Kim et al. (2009) claim that online retailers should realise that in order to
build e-loyalty and e-satisfaction, there has to be a prior development of e-trust. A study
by Kao and Lin (2016), which aimed to test whether a simultaneous relationships exists
between trust and loyalty, suggested that loyalty has a positive relationship with trust and
trust exerts a positive impact on loyalty.
Some authors (Ghane et al., 2011; Ribbink et al., 2004) claimed that e-trust not only has a
direct impact on e-loyalty, but also has an indirect influence through e-satisfaction. We suppose
that e-trust has an indirect influence on e-loyalty through e-satisfaction. Singh and
Sirdeshmukh (2000) argue that in any buyer–seller relationship, consumer trust Effect of
evaluations before a specific exchange episode are found to have a direct influence on e-service
their post-purchase satisfaction. In previous studies, e-trust has been a strong predictor of
quality
e-satisfaction (Harris and Goode, 2004; Jin and Park, 2006; Kim et al., 2009). This study
therefore examines the relationship between e-satisfaction and e-trust with the two aspects
of loyalty (behavioural and attitudinal). Hypotheses are thus formulated as follows:
H6. E-trust is positively related to customer satisfaction online.
H7a. E-satisfaction is positively related to customer behavioural loyalty online.
H7b. E-satisfaction is positively related to customer attitudinal loyalty online.
H8a. E-trust is positively related to customer behavioural loyalty online.
H8b. E-trust is positively related to customer attitudinal loyalty online.
H9. Behavioural loyalty online is positively related to customer attitudinal loyalty online.
The hypotheses specified above define the research model, while the conceptual model
that we intend to examine is depicted in Figure 2.
(X) Data and estimation

results 4.1 Sample characteristics
In order to test for causal relationships, a quantitative approach was used using a web-based
questionnaire addressed to students at Jordan University who were users of Amazon.com.
Following similar approaches from the literature (e.g. Lee and Lin, 2005; Parasuraman et al.,
2005; Flavián and Guinalíu, 2006; Al-dweeri et al., 2017), a student sample was selected due to
students representing a high proportion of users of the internet and new technologies (Peterson
and Merunka, 2014; Obeidat et al., 2017). We followed the recommendations of Roberts et al.
(2003) and Flavián and Guinalíu (2006) by choosing Amazon.com as a research setting.
Nevertheless, to get the needed responses a number of banners and posts were posted on
Facebook student pages regarding which of the following websites they chosen at least once for
online shopping? The websites proposed were: Amazon.com, MarkaVIP.com, eBay.com,
Alibaba.com and Souq.com, which are the top ranking sites in the e-shopping category in
Jordan (Alexa.com, 2017). Amazon.com was the most used, including 56 per cent of the sample
(see Table AI). Consequently, in addition to
Efficiency H1a
Behavioural
E-Satisfaction H7a
H2a loyalty
Privacy H3a
H7b
H4a
H5a
Reliability
H6 H9
H1b
Emotional benefit H3b H2b H8a
H4b
Attitudinal
E-trust H8b
Figure 2.
Customer service H5b loyalty
Research hypothesis
IMDS being the mostly used website by Jordanian students, we also chose it for a number of
reasons, first, Amazon.com is one of the few online retailers if not the only one to provide
variety of functions including a Marketplace, Amazon Services and Web Services, and the
collaboration on digital text platforms with other firms such as Apple (Ritala et al., 2014).
Second, Amazon can capture and provide more value to consumers that is jointly created
with third-party sellers by providing direct entry into their product spaces (Zhu and
Qihong, 2016). Finally, it is the biggest online retailer in the world (Zhu and Qihong,
2016). Nevertheless, since the data were gathered from a convenience sample (Garg and
Dhar, 2016; Lee et al., 2016; Pienaar and Willemse, 2008), the results are not
generalisable to the whole population of Jordanian students.
The survey was conducted between February and March 2017 (see Table II). The
online questionnaire consisted of a total of 38 items using a 1–5 scale in which 1 was
“strongly disagree” and 5 “strongly agree”, which were later refined through a pre-test with
two researchers. A total of 316 students filled out the questionnaire. This sample size is
similar to other e-service quality studies (384 in Bauer et al., 2006; 273 in Kim and Stoel,
2004; 359 in Águila et al., 2013). To test the suitability of the measurement scales, SPSS
17.0 and AMOS 20 software were used.

Primary data collection focussed on undergraduate students in Jordan: 61.7 per cent
were female and 38.3 per cent were male; 59.5 per cent were aged between 21 and 25; one
quarter had visited the website in the last 12 months between 6 and 12 times; 35.1 per cent
made a purchase in the last 12 months at least three times.
4.2 Empirical assessment of research constructs

We used Cronbach’s α to investigate the reliability of each construct; all α coefficients
achieved at least 0.70. The dimensions of e-service quality have the following values:
efficiency 0.906, privacy 0.871, reliability 0.860, emotional benefit 0.802 and customer
service 0.834. The other constructs in the model had the following values: e-satisfaction
0.835, e-trust 0.772, attitudinal loyalty 0.753 and behavioural loyalty 0.734. The resulting
scales were thus acceptable and sufficiently reliable (Nunnally, 1967). The overall
Cronbach’s α was 0.963.
In order to ensure the adequacy of the measurement model, the discriminant validity
was tested by the correlation between the variables in the confirmatory model; the results
showed an acceptable level of discrimination (Table III) as they were not greater than 0.8
(Bagozzi, 1994).
Items Reference models
Efficiency Adapted from:

Loiacono et al. (2007), Wolfinbarger and Gilly (2003), Li et al. (2002), Parasuraman et al.
(2005), Bauer et al. (2006)
Privacy Wolfinbarger and Gilly (2003), Parasuraman et al. (2005), Bauer et al. (2006), Collier and
Bienstock (2006)
Reliability Swaid and Wigand (2009), Parasuraman et al. (2005), Wolfinbarger and Gilly (2003),
Aladwani and Palvia (2002), Bauer et al. (2006)
Emotional benefit Wolfinbarger and Gilly (2003), Loiacono et al. (2007), Bauer et al. (2006)
Customer service Wolfinbarger and Gilly (2003), Parasuraman et al. (2005), Loiacono et al. (2007)
e-satisfaction Anderson and Srinivasan (2003), Cristobal et al. (2007), Thaichon and Quach (2015)
e-trust Yoon (2002), Kim et al. (2009), Kao and Lin (2016), Ghane et al. (2011), Ribbink et al. (2004)
Attitudinal loyalty Ho and Lee (2015), Kao and Lin (2016), Ghane et al. (2011) and Ziaullah et al. (2014)
Table II. Behavioural Ho and Lee (2015), Kao and Lin (2016) Kim et al. (2009), Ghane et al. (2011) and
Measurement scale loyalty Ziaullah et al. (2014)
Effect of
RELI CSERV EMOT PRIV EFFI TRU SATIS BEHA ATTIT
e-service
RELI 1.00 quality
CSERV 0.66 1.00
EMOT 0.53 0.49 1.00
PRIV 0.62 0.60 0.54 1.00
EFFI 0.69 0.61 0.61 0.77 1.00
TRU 0.65 0.62 0.54 0.63 0.62 1.00
SATIS 0.64 0.64 0.65 0.62 0.63 0.67 1.00
BEHA 0.54 0.47 0.48 0.45 0.46 0.62 0.66 1.00 Table III.
ATTIT 0.60 0.52 0.55 0.54 0.57 0.64 0.64 0.61 1.00 Sample correlations
Summary values for the variables were calculated by averaging the items in the scales. All
variables were sufficiently normally distributed, with skewness and kurtosis coefficients within the
range of −2.00 to +2.00. The correlations are presented in Table III. The correlation coefficients
were positive and significant at the 0.01 level for all variable pairings.
Figure 2 presents the structural equation modelling results specified in the AMOS 20.0
output. The results relating to the fit of the structural model generally support a claim of good
fit. Table IV provides a summary of the goodness-of-fit statistics (Figure 3).
As shown in Table III, all of the indices fall within the recommended ranges, which support a
2
claim of good fit for the model. In particular, the relative normed χ value of 2.68 is less than the
recommended maximum value of 3.00 (Bagozzi et al., 1998; Kline, 1998),
Indicator Recommended value Value

2
χ p 0.05 p ¼ 0.000
RMSEA 0.08 0.073
GFI 0.90 0.982
AGFI 0.90 0.919
NFI 0.90 0.986
IFI 0.90 0.991
CFI 0.90 0.991 Table IV.
2
Normed χ 1–2 2.684 Tests of model fit
Efficiency
0.03 e6
0.77
0.63 0.49
0.09 0.44
0.69
Privacy 0.14 Satisfaction Behavioural e8
0.61 0.62 0.22 0.32
0.61 0.54 Reliability 0.26 0.22 0.23

0.29
0.60 0.53 0.29 0.05
Emotional 0.30
0.66 0.15 Trust Attitudinal e9
0.49 0.19 0.55 0.52
Figure 3.
CService 0.21 AMOS model
e7
IMDS which represents a good fit. The RMSEA value of 0.073, which is below the recommended
maximum of 0.080 suggested by Browne and Cudeck (1993), also suggests that the
measurement model fits well. The GFI value of 0.982 and the AGFI value of 0.919 are both
above the 0.900 level recommended by Byrne (2016). This research also used IFI and CFI to
measure the goodness-of-fit of the models; the IFI (0.991) and CFI (0.991) index values for the
measurement model both exceed the recommended level of 0.900 (Byrne, 2016), which
suggests adequate fit of the model (Hu and Bentler, 1999). The NFI value of 0.986 also
suggests a reasonable fit. From all of the values outlined above, it is inferred that the
structural model represents an acceptable fit.
To test the hypotheses, a structural equation model was used (Hair et al., 1998). All
measurements of the equations were statistically significant at a significance level of 0.05,
and the measures of reliability and variance extracted from the constructs were all
significant (Table V ). The estimated standardized regression coefficients are shown
(all t-values exceed the reference values of 1.96 for a significance level of 0.05) (Figure 4).
5. Discussion
To summarise, using a student sample of Jordanian Amazon.com users (n ¼ 316), this study
examined the influence of e-service quality dimensions (i.e. efficiency, privacy, reliability,
emotional benefit and customer service), mediated by the influence of e-trust and
satisfaction on attitudinal and behavioural loyalty. Generally, for our sample e-trust was
found to be a stronger mediator between the e-service quality dimensions and behavioural
and attitudinal loyalty, with all the hypotheses regarding this sequence supported except in
the case of emotional benefit and e-trust. With regard to the role of e-satisfaction as a
mediator, a positive influence was found on attitudinal and behavioural loyalty, in addition
to emotional benefit and customer service. However, insignificant links were found between
e-satisfaction and efficiency, privacy and reliability.
With regard to the e-service quality dimensions in this study and context, the data
obtained for the proposed model confirm that these sub-dimensions, except for efficiency, all
have a significant influence on e-trust, which in turn also positively influenced both attitudinal
and behavioural loyalty. Regarding the path mediated by e-satisfaction, only the emotional
benefits and customer service were significant. Moreover, in relation to our first dimension of
Estimated Standardized estimated

Causal path coefficient SE p coefficient Hypothesis
Efficiency→Satisfaction 0.032 0.058 0.578 0.03 Not supported

Privacy→Satisfaction 0.080 0.049 0.099 0.09 Not supported
Reliability→Satisfaction 0.141 0.056 0.012 0.14 Not supported
Emotional→Satisfaction 0.287 0.045 *** 0.29 Supported
Customer Service→Satisfaction 0.201 0.053 *** 0.19 Supported
Satisfaction→Trust 0.211 0.049 *** 0.22 Supported
Efficiency→Trust 0.050 0.067 0.451 0.05 Not supported
Privacy→Trust 0.199 0.055 *** 0.22 Supported
Reliability→Trust 0.281 0.063 *** 0.26 Supported
Emotional→Trust 0.159 0.051 0.002 0.15 Supported
Customer service→Trust 0.234 0.060 *** 0.21 Supported
Satisfaction→Behavioural 0.463 0.057 *** 0.44 Supported
Satisfaction→Attitudinal 0.298 0.060 *** 0.29 Supported
Trust→Behavioural 0.322 0.054 *** 0.32 Supported
Trust→Attitudinal 0.299 0.054 *** 0.30 Supported
Table V. Behavioural→Attitudinal 0.231 0.054 *** 0.23 Supported
Structural parameters Notes: SE, standard error; p ¼ p-value. ***p-value less than 0.001
Effect of
Efficiency e-service
0.03
quality
*
0.56
Privacy 0.09 E-Satisfaction 0.44 Behavioural

t 1.65 t 8.12 loyalty
0.14
*
2.52
0.29 0.19
t 6.38 t 3.78 0.29
Reliability t 4.99
0.22 0.23
t 4.32 t 4.30
0.22 0.05
0.32
t 3.62 t 0.75
0.26 t 5.96
Emotional benefit t 4.49
0.15 E-trust 0.30 Attitudinal
t 3.10 t 5.50 loyalty

0.21 Figure 4.
Customer service t 3.89 Final model
e-service quality, it can also be confirmed that efficiency does not influence satisfaction and
trust in online shopping, as consistent with Hansen and Jonsson (2013). Neither the information
needed to perform transactions nor that required to create a good and attractive design had an
effect on the satisfaction and trust of consumers in our sample; perhaps the sample students had
enough information about the products before going to the website. This result is in direct
contrast to previous studies, which confirmed that efficiency influenced both e-satisfaction and
e-loyalty (e.g. Kao and Lin, 2016; Swaid and Wigand, 2009).
The second dimension relates to privacy. Despite being one of the most influential
dimensions in e-service quality, no relationship was found between privacy and e-satisfaction.
However, in terms of e-trust, a positive significant influence was found. Consequently, this
result could mean that perceptions of privacy were more influential on trust due to risk and
privacy concerns being one of the main reasons behind the lack of adoption of electronic and
mobile commerce in Jordan (Al Masarweh et al., 2016). In other words, due to the lack of
secure electronic infrastructure which hinders the adoption of electronic transactions on
Jordanian websites (Al-Bakri and Katsioloudes, 2015), Jordanian consumers are often found to
be more concerned about the privacy and security issues than about being satisfied. Thus, when
these privacy measures are found on a website, Jordanian consumers are more likely to be
trusting. Generally, there is no unanimous agreement among researchers about the role privacy
plays in assessing services offered by a website. Fassnacht and Koese (2006) found that privacy
does not exert a significant influence on perceptions of e-service quality. However, according to
the results, it can be concluded that privacy positively influences the trust of consumers more
than satisfaction: ensuring the protection of personal and financial data and the security of
transactions has a positive influence on the assessment of the quality of services offered by a
website, and these factors also have an indirect effect on behavioural and attitudinal loyalty.
The important role of privacy suggested in our study supports previous studies on e-trust
development (Kim et al., 2009), where it has been argued that a guarantee of privacy reduces
customer concerns about the illegal disclosure of personal data, which leads to a higher level of
e-trust.
Reliability is the most important factor in the assessment of electronic services, in
agreement with authors such as Lee and Lin (2005), Wolfinbarger and Gilly (2003),
IMDS Yang and Fang (2004) and Zeithaml et al. (2002). The results show that delivering
goods/services in the promised time and fulfilling the promises made have a positive
influence on e-satisfaction, e-trust and behavioural and attitudinal loyalty. Consumers
considered that the delivery of items within a suitable time frame positively influenced e-
trust and attitudinal loyalty, and meant they would say positive things about the website to
other people and would recommend the website to others.
Our results confirm that the visual appeal, innovation, emotional appeal, image and
attractive nature of a website has an influence on e-satisfaction and e-trust, in accordance
with other studies (Bauer et al., 2006; Cao et al., 2005). These authors confirm that
emotional benefit is a key factor in influencing the length of the relationship, as is
purchase intention. Our results are consistent with this, but we notice that emotional
appeal has much more influence on e-trust than on satisfaction.
Based on our findings, in the Middle East and Jordan, Jordanian managers should show a
willingness to respond to customer needs and answer questions promptly is important to obtain
the trust of customers. Firms online should show a sincere interest in solving the problems
faced by customers in transactions and offer support after-sales to obtain e-trust and e-
satisfaction, and thus a competitive advantage. Despite the study being limited to a Jordanian
sample, these results are in accordance with other studies, such as those by Parasuraman et al.
(2005) and Wolfinbarger and Gilly (2003). We also found that customer service mildly affects
behavioural and attitudinal loyalty. Other conclusions of this research is that reliability is the
most important factor in ensuring that customers trust a website, and that emotional benefit is
the most important factor for satisfied customers. We attribute this to the fact that the sample of
this study is comprised of young Jordanian people.
In this research, we can confirm that there is a positive relationship between the main
constructs of the measurement model, e-satisfaction and e-trust, and the two types of e-loyalty
(behavioural and attitudinal). E-trust is an antecedent of e-satisfaction, and Jordanian
consumers can be satisfied if a website has given them reasons to trust. The results of this
research confirm those obtained by Singh and Sirdeshmukh (2000), who proposed that
consumer trust evaluations have a direct influence on their post-purchase satisfaction.
When a customer perceives an improvement in the quality of service, their buying
behaviour will be influenced in a favourable way in their future purchase decisions. Their
intention to speak well about the website, recommend it or remain loyal, will increase and
their intentions to complain will decrease. This leads to a positive psychological customer
engagement with the website that is considered essential for successful businesses in the
long term.
The results of the empirical study show how quality electronic services form a
multidimensional structure, based on a hierarchical structure, where the perception of
quality is defined by clear dimensions, made up, in turn, of various sub-dimensions.
6. Managerial implications and future research lines

To provide an excellent service quality, Jordanian and Middle East firms in particular and
global firms in general should understand the patronage motives which drive consumers to use
their online websites. As a result, understanding the basic attributes consumers consider when
evaluating the service quality of websites is the first step in providing superior e-service quality
(Parasuraman et al., 2005). With regard to e-loyalty, when loyalty is measured along the two
dimensions studied here, firms can identify which stage of loyalty the customer is in; thus, if
the customer is only in an attitudinal loyalty stage (e.g. spreading positive word of mouth but
not using the service), special promotional programmes (e.g. special discounts, free samples)
could be used to enhance his loyalty to make it become behavioural, whereas if a customer is
already behaviourally loyal, customised offerings, price cuts and increasing the switching costs
could further increase customer behavioural loyalty.
As seen in the results of this study, in order to build e-loyalty, Jordanian and global firms Effect of
should improve their web platform infrastructure in terms of these five e-service quality e-service
dimensions. A superior customer service in the shape of a continuous online presence on social
media, quick recovery actions, call centre services and after-sales services would go a long way
quality
in increasing the service quality of their websites. In addition, one of the most important
findings of the research is the fact that e-trust was seen as a more important predictor of e-
loyalty in both of its aspects. Consequently, due to the slow adoption rate of electronic
commerce in the Arab region and specifically in Jordan (Al Masarweh et al., 2016), Jordanian
and international firms with a popular online presence should focus on highlighting the safety
and privacy of their transactions when dealing with customers who tend to value e-trust more.
Moreover, establishing a legal framework to ensure the protection of the personal data of
customers could also enhance perceptions of privacy.
In terms of emotional benefits, Jordanian and international firms should also work on
increasing customer emotional engagement with their websites by increasing the visual appeal
and innovativeness of their websites. Customization of a website in terms of colours, images
and font size with respect to different cultures should also enhance the perceptions of e-service
quality in these countries. Regarding reliability, firms should focus on succeeding in making
on-time deliveries, providing information on the shipping processes and procedures and
fulfilling delivery of the exact product as characterised on their page (Nath and Zheng, 2004).
All of these elements help in gaining the customer’s e-trust, which leads to a high level of e-
satisfaction and e-loyalty to the online company. Consequently, adopting a customer-oriented
approach would help firms to achieve a competitive advantage over competitors.
The limitations of this study include the fact that only Amazon.com was analysed.
Further investigation of other companies or countries is recommended to enable further
empirical testing of the methodology developed in this study, could also use different
samples other than students to increase the generalisability of the findings, and conduct
and test this model in more countries seeing as this study was only conducted in Jordan.
The model was validated in this study with reference to the measurement of e-service
quality in the case of private customers; other studies could attempt to validate this model
with business customers. Moreover, future research could incorporate other dimensions
that could influence the perception of e-service quality. For instance, examining the effect
of certain features of the interactivity of a website, such as 3D modelling and design
aesthetics, on e-service quality has not been yet examined. In addition, examining the
dimensions of e-service quality in a social media setting rather than a website setting could
provide further insights as to whether differences exist between these dimensions in
different settings. Furthermore, with the majority of studies examining e-service quality
employing self-administered surveys, future research could employ experimental designs
in order to see how manipulation of the service quality dimensions influences behavioural
and attitudinal loyalty. In other hand, other researchers could examine the impact of e-
service quality on e-loyalty through the mediation role of e-satisfaction and e-trust.
Additionally, with the prevalent influence of personality traits on consumer behaviour,
examining the influence of certain personality traits on the perception of e-service quality,
e-loyalty and e-satisfaction could provide both theoretical and managerial contributions.
References
Ab Hamid, N.R. and McGrath, M. (2005), “The diffusion of internet interactivity on retail websites: a
customer relationship model”, Communications of the IIMA, Vol. 5 No. 2, pp. 35-46.
Águila, A.R., Meléndez, A.P. and Al-dweeri, R.M. (2013), “The influence of electronic service
quality on loyalty in postal services: the mediating role of satisfaction”, Total Quality
Management and Business Excellence, Vol. 24 No. 9, pp. 1111-1123.
IMDS Aladwani, A. and Palvia, P. (2002), “Developing and validating an instrument for measuring user-
perceived web quality”, Information & Management, Vol. 39 No. 6, pp. 467-476.
Aladwani, A.M. (2002), “The development of two tools for measuring the easiness and usefulness of
transactional websites”, European Journal of Information Systems, Vol. 11 No. 3, pp. 223-231.
Al-Bakri, A.A. and Katsioloudes, M.I. (2015), “The factors affecting e-commerce adoption by
Jordanian SMEs”, Management Research Review, Vol. 38 No. 7, pp. 726-749.
Al-dweeri, R.M., Obeidat, Z.M., Al-dwiry, M.A., Alshurideh, M.T. and Alhorani, A.M. (2017), “The
impact of e-service quality and e-loyalty on online shopping: moderating effect of e-
satisfaction and e-trust”, International Journal of Marketing Studies, Vol. 9 No. 2, pp. 92-103.
Alexa.com (2017), available at: www.alexa.com/topsites/countries/JO ( accessed July 2017).
Al Masarweh, M., Al-Masaeed, S., Al-Qaisi, L. and Hunaiti, Z. (2016), “E-commerce adoption at
customer level in Jordan: an empirical study of Philadelphia general supplies”, International
Journal of Advanced Computer Science and Applications, Vol. 7 No. 11, pp. 196-205.
Anderson, R.E. and Srinivasan, S.S. (2003), “E-satisfaction and e-loyalty: a contingency framework”,
Psychology and Marketing, Vol. 20 No. 2, pp. 99-121.
Bagozzi, R.P. (1994), Structural Equation Models in Marketing Research: Basic Principles of
Marketing Research, Blackwell, Oxford.
Bagozzi, R.P., Yi, Y. and Nassen, K.D. (1998), “Representation of measurement error in marketing
variables: review of approaches and extension to three-facet designs”, Journal of
Econometrics, Vol. 89 No. 1, pp. 393-421.
Bauer, H., Falk, T. and Hammerschmidt, M. (2006), “eTransQual: a transaction process-based
approach for capturing service quality in online shopping”, Journal of Business Research, Vol.
59 No. 7, pp. 866-875.
Browne, M.W. and Cudeck, R. (1993), “Alternative ways of assessing model fit”, in Bollen, K.A. and
Long, J.S. (Eds), Testing Structural Equation Models, Sage, Beverly Hills, CA, pp. 136-162.
Bush, I. and Kraft, K. (1998), “Accountable welfare reform: what consumers think”, Public
Administration Review, Vol. 58 No. 5, pp. 406-415.
Byrne, B.M. (2016), Structural Equation Modeling with AMOS: Basic Concepts, Applications, and
Programming, Routledge, New York, NY.
Cao, M., Zhang, Q. and Seydel, J. (2005), “B2C e-commerce website quality: an empirical
examination”, Industrial Management & Data Systems, Vol. 105 No. 5, pp. 645-661.
Casielles, R., Álvarez, L. and Lanza, A.B. (2009), “Customer satisfaction and switching barriers:
effects on repurchase intentions, positive recommendations, and price tolerance”, Journal of
Applied Social Psychology, Vol. 39 No. 10, pp. 2275-2302.
Chen, Q., Rodgers, S. and He, Y. (2008), “A critical review of the e-satisfaction literature”, American
Behavioral Scientist, Vol. 52 No. 1, pp. 38-59.
Cheng, S.I. (2011), “Comparisons of competing models between attitudinal loyalty and behavioral
loyalty”, International Journal of Business and Social Science, Vol. 2 No. 10, pp. 149-166.
Chu, K. (2009), “The construction model of customer trust, perceived value and customer loyalty”,
The Journal of American Academy of Business, Vol. 14 No. 2, pp. 98-103.
Chung, K.H. and Shin, J.I. (2010), “The antecedents and consequents of relationship quality in internet
shopping”, Asia Pacific Journal of Marketing and Logistics, Vol. 22 No. 4, pp. 473-491.
Collier, J.E. and Bienstock, C.C. (2006), “Measuring service quality in e-retailing”, Journal of
Service Research, Vol. 8 No. 3, pp. 260-275.
Cox, J. and Dale, B. (2001), “Service quality and e-commerce: an exploratory analysis”, Managing
Service Quality, Vol. 11 No. 2, pp. 121-132.
Cristobal, E., Flavián, C. and Guinalíu, M. (2007), “Perceived e-service quality (PeSQ): measurement Effect of
validation and effects on consumer satisfaction and website loyalty”, Managing Service e-service
Quality: An International Journal, Vol. 17 No. 3, pp. 317-340.
quality
Dowling, G.R. and Uncles, M. (1997), “Do customer loyalty programs really work?”, Sloan
Management Review, Vol. 38 No. 4, pp. 71-83.
Fandy, T. and Chandra, G. (2008), “Service management”, mewujudkan layanan prima, Andi
Yogyakarta, Hal, 36.
Fassnacht, M. and Koese, I. (2006), “Quality of electronic services: conceptualizing and testing a
hierarchical model”, Journal of Service Research, Vol. 9 No. 1, pp. 19-31.
Flavián, C. and Guinalíu, M. (2006), “Consumer trust, perceived security and privacy policy: three
basic elements of loyalty to a website”, Industrial Management & Data Systems, Vol. 106 No.
5, pp. 601-620.
Frasquet, M., Descals, A.M. and Ruiz-Molina, M.E. (2017), “Understanding loyalty in multichannel
retailing: the role of brand trust and brand attachment”, International Journal of Retail &
Distribution Management, Vol. 45 No. 6, pp. 608-625.
Garg, S. and Dhar, R.L. (2016), “Extra-role customer service: the roles of leader–member exchange
(LMX), affective commitment, and psychological empowerment”, International Journal of

Hospitality & Tourism Administration, Vol. 17 No. 4, pp. 373-396.
Gefen, D., Karahanna, E. and Straub, D.W. (2003), “Trust and TAM in online shopping: an integrated
model”, MIS Quarterly, Vol. 27 No. 1, pp. 51-90.
Gera, R., Mittal, S., Batra, D.K. and Prasad, B. (2017), “Evaluating the effects of service quality,
customer satisfaction, and service value on behavioral intentions with life insurance customers
in India”, International Journal of Service Science, Management, Engineering, and Technology
(IJSSMET), Vol. 8 No. 3, pp. 1-20.
Ghane, S., Fathian, M. and Gholamian, M.R. (2011), “Full relationship among e-satisfaction, e-trust,
e-service quality, and e-loyalty: the case of Iran e-banking”, Journal of Theoretical and Applied
Information Technology, Vol. 33 No. 1, pp. 1-6.
Giovanis, A.N. and Athanasopoulou, P. (2014), “Gaining customer loyalty in the e-tailing
marketplace: the role of e-service quality, e-satisfaction and e-trust”, International Journal of
Technology Marketing, Vol. 9 No. 3, pp. 288-304.
Gommans, M., Krishnan, K.S. and Scheffold, K.B. (2001), “From brand loyalty to e-loyalty: a
conceptual framework”, Journal of Economic and Social Research, Vol. 3 No. 1, pp. 43-58.
Gounaris, S. and Dimitriadis, S. (2003), “Assessing service quality on the web: evidence from
business-to-consumer portals”, Journal of Service Marketing, Vol. 17 No. 5, pp. 529-548.
Gounaris, S., Dimitriadis, S. and Stathakopoulos, V. (2010), “An examination of the effects of service
quality and satisfaction on customer’s behavioural intentions in e-shopping”, Journal of
Services Marketing, Vol. 24 No. 2, pp. 142-156.
Ha, S. and Stoel, L. (2009), “Consumer e-shopping acceptance: antecedents in a technology
acceptance model”, Journal of Business Research, Vol. 62 No. 5, pp. 565-571.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis, Prentice
Hall International, Upper Saddle River, NJ.
Hansen, E. and Jonsson, E. (2013), “E-loyalty in fashion e-commerce: an investigation in how to
create e-loyalty”, master’s thesis, Linnaeus University, Småland.
Harris, L.C. and Goode, M.M. (2004), “The four levels of loyalty and the pivotal role of trust: a study
of online service dynamics”, Journal of Retailing, Vol. 80 No. 2, pp. 139-158.
Hassan, Y.M. and Herrero, S.V. (2006), “Improving tag-clouds as visual information retrieval
interfaces”, Proceedings of the International Conference on Multidisciplinary Information
Sciences and Technologies, Mérida, October 25–28.
Herington, C. and Weaven, S.K.W. (2009), “E-retailing by banks: e-service quality and its importance to
customer satisfaction”, European Journal of Marketing, Vol. 43 Nos 9/10, pp. 1220-1231.
IMDS Ho, C.I. and Lee, P.L. (2015), “Are blogs still effective to maintain customer relationships?
An empirical study on the travel industry”, Journal of Hospitality and Tourism Technology,
Vol. 6 No. 1, pp. 5-25.
Hsu, J. and Hsu, C. (2008), “The relationships between service quality and customer satisfaction in a
leading Chinese Web 2.0 company”, The Business Review, Vol. 11 No. 1, pp. 84-90.
Hu, L. and Bentler, P. (1999), “Cut off criteria for fit indexes in covariance structure analysis:
conventional criteria versus new alternatives”, Structural Equation Modeling, Vol. 6 No. 1, pp.
1-55.
Jin, B. and Park, J.Y. (2006), “The moderating effect of online purchase experience on the evaluation
of online store attributes and the subsequent impact on market response outcomes”, Advances
in Consumer Research, Vol. 33, pp. 203-212.
Kang,J., Alejandro, T.B. and Groza, M.D. (2015), “Customer–company identification and the
effectiveness of loyalty programs”, Journal of Business Research, Vol. 68 No. 2, pp. 464-471.
Kantsperger, R. and Kunz, W.H. (2010), “Consumer trust in service companies: a multiple mediating
analysis”, Managing Service Quality: An International Journal, Vol. 20 No. 1, pp. 4-25.
Kao, T.W. and Lin, W.T. (2016), “The relationship between perceived e-service quality and brand
equity: a simultaneous equations system approach”, Computers in Human Behavior, Vol. 57,
pp. 208-218.
Kim, J., Jin, B. and Swinney, J.L. (2009), “The role of retail quality, e-satisfaction and e-trust in
online loyalty development process”, Journal of Retailing and Consumer Services, Vol. 6 No.
4, pp. 239-247.
Kim, J.H. and Lennon, S.J. (2017), “Descriptive content analysis on e-service research”, International
Journal of Service Science, Management, Engineering, and Technology, Vol. 8 No. 1, pp. 18-
31.
Kim, S. and Stoel, L. (2004), “Dimensional hierarchy of retail website quality”, Information &
Kim, R. and Peterson, R. (2017), “A meta-analysis of online trust relationships in e-commerce”,
Journal of Interactive Marketing, Vol. 38, pp. 44-54.
Kline, R.B. (1998), Principles and Practice of Structural Equation Modeling, The Guilford Press,
New York, NY.
Lee, G.G. and Lin, H.F. (2005), “Customer perceptions of e-service quality in online shopping”,
International Journal of Retail and Distribution Management, Vol. 33 No. 2, pp. 161-176.
Lee, S., Kyung, S. and Kim, K. (2016), “Information and communication technology overload and
social networking service fatigue: a stress perspective”, Computers in Human Behavior, Vol.
55, pp. 51-61.
Li, Y.N., Tan, C.K. and Xie, M. (2002), “Measuring web-based service quality”, Total Quality
Loiacono, E., Watson, R.T. and Goodhue, D. (2007), “WebQual: an instrument for consumer evaluation of
websites”, International Journal of Electronic Commerce, Vol. 11 No. 3, pp. 51-87.
Luarn, P. and Lin, H.H. (2003), “A customer loyalty model for e-service context”, Journal of
Electronic Commerce Research, Vol. 4 No. 4, pp. 156-167.
Malhotra, A. and Segars, A.H. (2005), “Investigating wireless web adoption patterns in the US”,
Communications of the ACM, Vol. 48 No. 10, pp. 105-117.
Manuel, P. (2015), “A trust model of cloud computing based on quality of service”, Annals of
Operations Research, Vol. 233 No. 1, pp. 281-292.
Mohd Kassim, N. and Ismail, S. (2009), “Investigating the complex drivers of loyalty in e-commerce
settings”, Measuring Business Excellence, Vol. 13 No. 1, pp. 56-71.
Nath, A. and Zheng, L. (2004), Perception of Service Quality in E-commerce: An Analytical Study
of Internet Auction Sites, Lulea University of Technology, Lulea.
Nunnally, J.C. (1967), Psychometric Theory, McGraw Hill, New York, NY. Effect of
Obeidat, Z.M.I., Xiao, S.H., Iyer, G.R. and Nicholson, M. (2017), “Consumer revenge using the e-service
internet and social media: an examination of the role of service failure types and cognitive quality
appraisal processes”, Psychology & Marketing, Vol. 34 No. 4, pp. 496-515.
Oliver, R.L. (1997), Satisfaction: A Behavioral Perspective on the Consumer, McGraw-Hill, New York, NY.
Ou, W.M., Shih, C.M., Chen, C.Y. and Wang, K.C. (2011), “Relationships among customer loyalty
programs, service quality, relationship quality and loyalty: an empirical study ”, Chinese
Management Studies, Vol. 5 No. 2, pp. 194-206.
Parasuraman, A. and Zinkhan, G.M. (2002), “Marketing to and serving customers through the
internet: an overview and research agenda”, Journal of the Academy of Marketing Science,
Vol. 30 No. 4, pp. 286-295.
Parasuraman, A., Zeithaml, V. and Malhotra, A. (2005), “E-S-QUAL: a multiple-item scale for
assessing electronic service quality”, Journal of Service Research, Vol. 7 No. 3, pp. 213-233.
Peterson, R.A. and Merunka, D.R. (2014), “Convenience samples of college students and research
reproducibility”, Journal of Business Research, Vol. 67 No. 5, pp. 1035-1041.
Pienaar, J. and Willemse, S.A. (2008), “Burnout, engagement, coping and general health of service
employees in the hospitality industry”, Tourism Management, Vol. 29 No. 6, pp. 1053-1063.
Ranaweera, C., McDougall, G. and Bansal, H. (2005), “A model of online customer behavior during
the initial transaction. Moderating effects of customer characteristics”, Marketing Theory, Vol.
5 No. 1, pp. 148-160.
Ribbink, D., Van Riel, C.R., Liljander, V. and Streukens, S. (2004), “Comfort your online customer:
quality, trust and loyalty on the internet”, Managing Service Quality: An International Journal,
Vol. 14 No. 6, pp. 446-456.
Ritala, P., Golnam, A. and Wegmann, A. (2014), “Coopetition-based business models: the case of
Amazon. com”, Industrial Marketing Management, Vol. 43 No. 2, pp. 236-249.
Roberts, K., Varki, S. and Brodie, R. (2003), “Measuring the quality of relationships in consumer
services: an empirical study”, European Journal of Marketing, Vol. 37 Nos 1/2, pp. 169-196.
Rolland, S. and Freeman, I. (2010), “A new measure of e-service in France”, International Journal of
Retail and Distribution Management, Vol. 38 No. 7, pp. 497-517.
Romadhoni, B., Hadiwidjojo, B., Noermijati and Aisjah, S. (2015), “Relationship between e-service
quality, e-satisfaction, e-trust, e-commitment in building customer e-loyalty: a literature
review”, International Journal of Business and Management Invention, Vol. 4 No. 2, pp. 1-9.
Rowley, J. (2006), “An analysis of the e-service literature: towards a research agenda”, Internet
Research, Vol. 16 No. 3, pp. 339-359.
Russell, B.R., McColl-Kennedy, J.R. and Coote, L.V. (2007), “Involvement, satisfaction, and brand loyalty
in a small business services setting”, Journal of Business Research, Vol. 60 No. 12, pp. 1253-1260.
Salameh, A.A. and Bin Hassan, S. (2015), “Measuring service quality in m-commerce context: a
conceptual model”, International Journal of Scientific and Research Publications, Vol. 5 No. 3,
pp. 1-9.
Santos, J. (2003), “E-service quality: a model of virtual service quality dimensions”, Managing
Service Quality, Vol. 13 No. 3, pp. 233-246.
Shia, B., Chen, M. and Ramdansyah, A. (2016), “Measuring customer satisfaction toward localization
website by WebQual and importance performance analysis: case study on Aliex Press site in
Indonesia”, American Journal of Industrial and Business Management, Vol. 6 No. 2, pp. 117-
128.
Singh, J. and Sirdeshmukh, D. (2000), “Agency and trust mechanisms in consumer satisfaction and
loyalty judgments”, Journal of the Academy of Marketing Science, Vol. 28 No. 1, pp. 150-167.
Smith, A. and Rupp, W. (2002), “Communication and loyalty among knowledge workers: a resource
of the firm theory view”, Journal of Knowledge Management, Vol. 6 No. 3, pp. 250-261.
IMDS Swaid, S.I. and Wigand, R.T. (2009), “Measuring the quality of e-service: scale development and
initial validation”, Journal of Electronic Commerce Research, Vol. 10 No. 1, pp. 13-28.
Thaichon, P. and Quach, T.N. (2015), “The relationship between service quality, satisfaction, trust,
value, commitment and loyalty of internet service providers’ customers”, Journal of Global
Scholars of Marketing Science, Vol. 25 No. 4, pp. 295-313.
Venkatesh, V., Chan, F. and Thong, J. (2012), “Designing e-government services: key service
attributes and citizens’ preference structures”, Journal of Operations Management, Vols 1-2
No. 30, pp. 116-133.
Wang, M. and Huarng, S.A. (2002), “An empirical study of internet store customer post shopping
satisfaction”, Special Issues of Information Systems, Vol. 3 No. 5, pp. 632-638.
Wirtz, J., Kun, D. and Lee, K. (2000), “Should a firm with a reputation for outstanding service quality offer
a service guarantee?”, The Journal of Services Marketing, Vol. 14 No. 6, pp. 502-512.
Wolfinbarger, M.F. and Gilly, M.C. (2003), “eTailQ: dimensionalizing, measuring and predicting
etail quality”, Journal of Retailing, Vol. 79 No. 3, pp. 183-198.
Yang, Z. and Fang, X. (2004), “Online service quality dimensions and their relationship with
satisfaction: a content analysis of customer reviews of securities brokerage services”,

International Journal of Service Industry Management, Vol. 15 No. 3, pp. 302-326.
Yang, Z., Peterson, R.T. and Cai, S. (2003), “Service quality dimensions of internet retailing: an
exploratory analysis”, Journal of Service Marketing, Vol. 17 No. 7, pp. 685-700.
Yarimoglu, E.K. (2015), “A review of service and e-service quality measurements: previous literature
and extension”, Journal of Economic and Social Studies, Vol. 5 No. 1, pp. 169-200.
Yaya, L., Marimon, F. and Casadesus, M. (2011), “Customer's loyalty and perception of ISO 9001 in
online banking”, Industrial Management & Data Systems, Vol. 111 No. 8, pp. 1194-1213.
Yen, C.H. and Lu, H.P. (2008), “Effects of e-service quality on loyalty intention: an empirical study
in online auction”, Managing Service Quality, Vol. 18 No. 2, pp. 127-146.
Yoon, S.J. (2002), “The antecedents and consequences of trust in online purchase decisions”, Journal
of Interactive Marketing, Vol. 16 No. 2, pp. 47-63.
Zahedifard, E., Mohebi, S. and Dastranj, H.A. (2014), “Investigating effect of services quality
dimensions on insurant satisfaction; studying domain: customers of social security insurance ”,
Journal of Life Science and Biomedicine, Vol. 4 No. 6, pp. 541-549.
Zehir, C., Sehitoglu, Y., Narcikara, E. and Zehir, S. (2014), “E-S-quality, perceived value and loyalty
intentions relationships in internet retailers”, Procedia–Social and Behavioral Sciences, No.
150, pp. 1071-1079.
Zeithaml, V., Parasuraman, A. and Malhotra, A. (2002), “Service quality delivery through websites: a
critical review of extant knowledge”, Academy of Marketing Science Journal, Vol. 30 No. 4,
pp. 362-375.
Zhu, F. and Qihong, L. (2016), “Competing with complementors: an empirical look at Amazon.com”,
working paper.
Ziaullah, M., Feng, Y., Akhter, S.N. and Khan, M.F. (2014), “An empirical study on exploring
relationship among information quality, e-satisfaction, e-trust and young generation ’s
commitment to Chinese online retailing”, Journal of Competitiveness, Vol. 6 No. 4, pp. 3-18.
Further reading
Mobile Marketing Statistics, available at: www.smartinsights.com/mobile-marketing/mobile-
marketing-analytics/mobile-marketing-statistics/ (accessed 26 April 2016).
Sweeney, J.C. and Lapp, W. (2004), “Critical service quality encounters on the web: an exploratory
study”, Journal of Services Marketing, Vol. 28 No. 4, pp. 276-289.
purchased on this webs
Appendix 350 My
decision in
choosing this
site is correct
351
I
Efficiency a
1. This site makes it easy to find what I need Adapted from: m
2. It makes it easy to get anywhere on the site Loiacono et al. (2007), Wolfinbarger and Gilly (2003), Li
s
a
- It enables me to complete a transaction quickly et al. (2002), Parasuraman et al. (2005), Bauer et al. (2006) t
i
s
- Information at this site is well organised
f
i
- It loads its pages fast e
d
- This site is simple to use w

i
- This site is well organised t
h
Privacy
8. It protects information about my web- Wolfinbarger and Gilly (2003), Parasuraman et al. (2005), t
shopping behaviour Bauer et al. (2006), Collier and Bienstock (2006) h
- It does not share my personal information
e
with other sites
- This site protects information about my credit p
card u
- Website appears symbols and messages gives r
a signal that the site is safe and protected c
h
Reliability a
12. It delivers orders when promised Swaid and Wigand (2009), Parasuraman et al. (2005), s
13. This site makes items available for delivery Wolfinbarger and Gilly (2003), Aladwani and Palvia e
within a suitable time frame (2002), Bauer et al. (2006)
- It quickly delivers what I order e
x
- It sends out the items ordered p
- It is truthful about its offerings e
r
- It makes accurate promises about delivery i
of products e
Emotional benefit n
18. I feel cheerful when I use the website Wolfinbarger and Gilly (2003), Loiacono et al. (2007), c
19. Excitement when shopping online Bauer et al. (2006) e
20. I feel happy when I use this website
a
Customer service t
21. This website is ready and willing to respond to Wolfinbarger and Gilly (2003), Parasuraman et al. (2005),
customer needs Loiacono et al. (2007) t
- Customer service personnel are always h
willing to help you i
- Inquiries are answered promptly s
- When you have a problem, the website shows
a sincere interest in solving it w
- After sale support at this site is excellent e
b
E-satisfaction s
26. I am overall very happy with this website Anderson and Srinivasan (2003), Cristobal et al. (2007), i
27. I am satisfied with my decision to be Thaichon and Quach (2015)
t
e Effect of e-service quality
(continued )
Table AI. Measurement model

IMDS
E-trust
30. I trust what this website says about its Yoon (2002), Kim et al. (2009), Kao and Lin (2016),
products Ghane et al. (2011), Ribbink et al. (2004)
31. This website is reliable
32. I trust the claims and promises this website
makes about a product
Attitudinal loyalty
33. I consider myself to be a loyal patron of this Ho and Lee (2015), Kao and Lin (2016), Ghane et al. (2011)
website and Ziaullah et al. (2014)
34. I would say positive things about this website
to other people
35. I would recommend this website to someone
who seeks my advice
Behavioural loyalty
36. I would consider this website as my first choice Ho and Lee (2015), Kao and Lin (2016) Kim et al. (2009),
to buy services/goods Ghane et al. (2011) and Ziaullah et al. (2014)

37. I would do more business with this website in
the next few years
38. I am willing to put in extra effort to buy from
Table AI. this website
Rami Mohammad Al-dweeri can be contacted at: r.dweeri@ju.edu.jo
Technological intensity and Innovation

capability in
innovation capability in firms
industrial firms
Paulo Antônio Zawislak and Edi Madalena Fracasso 189
School of Management, Federal University of Rio Grande do Sul,
Porto Alegre, Brazil, and Received 3 June 2016
Accepted 30 November 2017
Jorge Tello-Gamarra
School of Chemistry and Food, Federal University of Rio Grande,
Santo Antônio da Patrulha, Brazil
Abstract
Purpose – Over time, technological intensity has been used as a proxy for innovation capability of firms in an
industrial sector. However, not only firms belonging to the stratum of high technological intensity are able to
innovate. Therefore, this study aims to explore a potential association between technological intensity and
innovation capability in firms from different industrial sectors, using the Organization for Economic Cooperation
and Development (OECD)’s classification and the components of innovation capability proposed by Zawislak et
al. (2012, 2013).
Design/methodology/approach – The authors conducted an exploratory research with four case studies
focusing on the innovation capability of Brazilian firms.
Findings – The results show that the four firms, each belonging to one stratum of technological intensity, have
innovation capability, and the differences regarding this feature can be explained by the balance and development
of all firms’ capabilities (technological, operational, managerial and transactional).
Originality/value – In the literature, studies that relate technological intensity and innovation capability are
scarce. Therefore, the originality of this research is to relate these two concepts. The most important is that firms
can be innovative regardless of their stratum of technological intensity, which shows the importance of other
capabilities to ensure the innovation’s success.
Keywords Firm, Innovation capability, Technological intensity
Paper type Research paper
1. Introduction
Technological intensity is defined as the level of knowledge incorporated in companies’
products in every industrial sector, and this indicator is typically measured by dividing the
© Paulo Antônio Zawislak, Edi Madalena Fracasso and Jorge Tello-Gamarra. Published in
Innovation & Management Review. Published by Emerald Publishing Limited. This article is
published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce,
distribute, translate and create derivative works of this article (for both commercial and non-
commercial purposes), subject to full attribution to the original publication and authors. The full
terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
JEL classification – D23, L6, M2, O3
Innovation &
The authors received financial support from the Research Foundation of the State of Rio Grande do Management Review
Sul (FAPERGS) and the Brazilian National Council for Scientific and Technological Development Vol. 15 No. 2, 2018
pp. 189-207
(CNPq). Emerald Publishing Limited
Conflicts of interest: The authors declare no conflicts of interest. 2515-8961
DOI 10.1108/INMR-04-2018-012
INMR average R&D spending by the firm’s revenue. The Organization for Economic Cooperation
15,2 and Development (OECD) is responsible for the classification of industrial sectors according
to their level of technological intensity [Organization for Economic Cooperation and
Development (OECD), 2003, 2007]. Through this indicator, it classifies industrial sectors in
four levels: high, medium-high, medium-low and low.
Because of OECD’s credibility, this classification is used by scholars whenever they need
r specific information about firms’ technological structure and strategic innovation behavior, in different
industrial sectors. Moreover, it has been commonly accepted that high-technology-
intensive firms are more innovative, more efficient, pay higher wages and are more
successful than low-technology-intensive firms (Hatzichronoglou, 1997; Markwald,
2004).
However, Griliches and Mairesse (1984) found a negative R&D elasticity in company
performance for low- and medium-technology firms (LMTs), while the elasticity for high-
tech firms (HTs) was positive. Other studies (Kafouros, 2005; Wang and Tsai, 2003)
achieved similar results, in which the effect of R&D activities on firm performance was
considerably higher for HTs than for LMTs. In other words, the effect of technological
innovation efficiency on firm performance may be less important for LMTs, as their
innovations are not usually the result of the latest technological knowledge, but rather rest
on their ability to turn different types of knowledge (their general knowledge stock) into
innovation (economically useful knowledge) (Bender and Laestadius, 2005). In the same
line of thought, Santamaría et al. (2009) found that the quantity of innovation was
statistically much larger for HTs than for LMTs.
Based on theoretical arguments and empirical evidence, we arrived at some questions.
In high-technology-intensive sectors, all firms should be considered more innovative?
Conversely, when a firm belongs to a low-technological intensity sector, should it be
considered a priori as less innovative? In other words, do different technological intensity
firms show different innovation capabilities?
By asking these questions, we shed light on two issues – technological intensity and
innovation capability – that have been constantly studied in the past decades. However,
the literature has dealt with them separately. Until now, very few authors (Segarra-Ciprés
et al., 2012) have suggested this conceptual relation as we propose.
Most of the studies consider technological intensity as the result of a positive relationship
between R&D spending and revenue, which means that the more high-tech a sector is, the
higher will be its performance [Hatzichronoglou, 1997; Organization for Economic
Cooperation and Development (OECD), 2003, 2007]. However, this only gives a partial view
of how sectoral agents actually behave, that is, which are the innovation capabilities behind
companies of a given sector (Hirsch-Kreinsen et al., 2005; Von Tunzelmann and Acha, 2005).
Indeed, that relationship does not consider how a company gets and uses its innovation
capability and becomes competitive, even in low-technology-intensive sectors.
To identify the innovation capability of firms in different technological intensity
stratum, we propose to correlate OECD’s typology (which implies different levels of
technology mastering for distinct sectors) with the innovation routines and behavior of
any firm (its innovation capability) in a specific sector. To do so, we carried out an
exploratory case study of four companies, each belonging to an industrial sector whose
technological intensity was defined by Organization for Economic Cooperation and
Development (OECD) (2003, 2007) typology.
After this introduction, the paper is divided in five sections. Section 2 presents the
classification of industrial sectors based on technological intensity, proposed by
Organization for Economic Cooperation and Development (OECD) (2003). Section 3
discusses the different approaches of innovation capabilities, with special emphasis on the Innovation
innovation capability model proposed by Zawislak et al. (2012, 2013). Section 4 describes capability in
the research method. Section 5 presents and discusses the results of the four selected
cases, and Section 6 shows the conclusion of the study. firms
2. Technological intensity
In a scenario where competition is fierce and firms struggle to keep and possibly increase 191
their market share, technological level and its evolution play a key role. In this context,
firms with high-technological levels would have a potential advantage over those with
low-technological levels. The acceptance of this assumption has confronted researchers,
public administrators and companies’ owners and managers with a challenge: after all,
what is, in fact, high technology? (Felsenstein and Bar-El, 1989).
Many authors have attempted to answer this question, including Palda (1986),
Felsenstein and Bar-El (1989), Hatzichronoglou (1997) and Organization for Economic
Cooperation and Development (OECD) (2003, 2007); all of them address high and low
technology as parts of the concept of technological intensity.
According to Felsenstein and Bar-El (1989), technological intensity has a
multidimensional character consisting of three dimensions, two referring to the industrial
inputs (labor and capital) and one to the output (product). The technological intensity of
the production factor “labor” relates to the amount of experience and skill levels of the
workforce in industry, while the technological intensity of the production factor “capital”
regards the quality of capital invested. The product’s technological intensity refers to those
industries that make large investments for developing new products and processes. At the
end, this vision of technological intensity directly relates to the traditional analysis of the
ratio between capital and labor, in which the more capital-intensive an industrial sector is,
the more technologically intensive it will be, and vice-versa.
Palda (1986) defined technological intensity as the degree to which scientific research
efforts contribute to increasing productivity and therefore revenue. Thus, technological
intensity could be measured as the ratio between R&D and the firm’s revenue, and not
only the capital–labor ratio. The more a firm invests in R&D, the higher its technological
intensity and the higher its revenue[1].
OECD, inspired by these studies, and especially by Hatzichronoglou’s (1997), offers a
comprehensive classification of technological intensity by industrial sectors, using the
concept of technological intensity to take into account both the level of technology
specific to the sector (measured by the ratio between R&D spending and value added) and
the technology embodied in acquisitions of intermediate and capital goods. Four groups of
industries were identified according to the degree of technological intensity, as shown in
Table I.
Although it serves as the basis for defining industrial policies and international trade,
various authors have found several limitations in OECD’s classification. Hatzichronoglou
(1997) highlights an important limitation regarding the classification criterion that
emphasizes R&D. He believes that although R&D is extremely important for high-
technology industries, it is not that important for other sectors. Attributes such as technical
and scientific staff, technology acquired through patents, licenses and know-how and
cooperation between companies, among other factors, could also have a significant role
when assessing technological intensity.
Moreover, authors like Furtado and Carvalho (2005), Kafouros (2005), Wang and Tsai
(2003), Srholec (2007), Santamaría et al. (2009), Mendonça (2009) and Reichert et al. (2016)
discuss different versions of technological intensity and firms’ behavior, inferring that even
INMR
Technological intensity strata Industries
15,2
High-technology industries (a) Aircraft and spacecraft; (b) Pharmaceuticals; (c) Electronics
components (d) Office, accounting and computing machinery; (e)
Radio, TV and communications equipment; (f) Medical, precision
and optical instruments
Medium-high-technology industries (a) Electrical machinery and apparatus; (b) Motor vehicles, (c)
192 Trailers and semi-trailers; (d) Chemicals excluding
pharmaceuticals; (e) Railroad equipment and transport equipment;
machinery and equipment
Medium-low-technology industries (a) Building and repairing of ships and boats; (b) Rubber and
plastics products; (c) Coke, refined petroleum products and
Table I. nuclear fuel; (d) Other non-metallic mineral products; (e) Basic
metals and fabricated metal products
Classification of
Low-technology industries (a) Manufacturing, recycling; (b) Wood, pulp, paper, paper
manufacturing products, printing and publishing; (c) Food products, beverages
industries according and tobacco, (d) Textiles, textile products, leather and footwear
to their technological
intensity Source: OCDE (2003, 2007)
low-tech sectors may have high-performance firms. This is especially true for developing
countries.
However, despite being quite relevant, these criticisms do not affect the essence of
OECD’s classification, that is, the development of a coherent framework for the analysis
of industry’s technological structure. Moreover, it has been widely used in recent years,
both by academics and policy makers. For all these reasons, OECD’s classification makes
the necessary stratification, with enough differences between the four groups to suggest
the existence of different arrangements of innovation capability.
3. Innovation capability
Innovation as a source of firm’s advantage is a well-consolidated topic in the literature.
However, studies that attempt to identify the sources of innovation are still ongoing. One
of the more advanced perspectives is technological capability. As a matter of fact, the
different capabilities’ approach intends to describe the innovation process that occurs
inside the firm boundaries.
These studies began with Katz (1984), Desai (1984), Lall (1992) and Bell and Pavitt
(1995), among others. For these authors, innovation is a process that depends only on
technological capability. Moreover, they showed a positive association between
technological capability, innovation and firm performance.
While technology is a relevant dimension of innovation (Shafia et al., 2016), if we
consider innovation exclusively as the outcome of scientific and technological advances,
the spectrum of how change and innovation occur in the vast majority of firms is central
(Alves et al., 2017). Nevertheless, despite the relationship between technological
capability and innovation being positive, several researchers (Teece, 1986; Patel and
Pavitt, 1997) observe that, for a firm to be innovative, technological capability is a
relevant feature, but not sufficient.
Innovation may be the result of a complex process and depend on a set of capabilities
that, although often dispersed throughout the company’s structure, can still be aligned with
its strategic requirements. This set of capabilities form a meta-capability known as
innovation capability. It is the firm’s ability to rapidly introduce new products and adopt
new processes, which are critical for competing with other firms (Guan and Ma, 2003; Innovation
Wang et al., 2008). capability in
Innovation capability has been studied by using three approaches: assets (Christensen,
1995), processes (Chiesa et al., 1996; Burgelman et al., 2004) and abilities (Guan and Ma,
firms
2003; Yam et al., 2004). Further studies have evolved to consider it as the result of a set of
complementary capabilities (Wonglimpiyarat, 2010; Forsman, 2011).
This complementarity is, in fact, the combination of the aforementioned technological
driver with the firm’s business driver. In other words, the firm is viewed as a technological 193
set of products and processes that operate under a specific business model, to trade and
profit from the market. Therefore, as suggested by Zawislak et al. (2012, 2013), every firm
has the following general capabilities to some extent: technological, operational,
managerial and transactional capabilities (Figure 1).
3.1 Technological capability

Technological capability refers to the skills, knowledge, experience and routines that the
firm needs to develop new products (goods and/or services). Technological capability
concerns directly R&D activities, which facilitate the creation of new products. This
capability is composed of monitoring technological advances, assimilating new
technologies and formalization of the development process.
Technological monitoring means that leading firms in their respective industries are able to
monitor the range of technological options available in the external environment, thus
identifying and choosing the most appropriate technology for their needs ( Rush et al., 2007).
Once the existing technologies have been identified, firms need to select and bring them inside.
This process of assimilating technologies occurs in two ways, through acquisition or learning
(Wong et al., 1998). For Christiansen and Varnes (2009), the formalization of the development
process suggests that structured approaches for managing the new product development
process are fundamental for a successful innovation. These formal rules are part of the best
practices in new product development (Griffin, 1997; Davila, 2000).
3.2 Operational capability

Operational capability concerns the organization of the production of goods and services on a
commercial scale. It has been defined as the skills, knowledge, experience and routines needed
to produce goods and services in a flexible way, with quality and at the lowest possible cost.
Operational capability is responsible for executing the ideas (products and
Figure 1.
Innovation capability
INMR processes) that originate from technological capability. It may be summarized by production
15,2 planning, the quality system used and the objectives of reducing production costs. Competition
requires that firms seek ways to optimize their production process. Production planning is
critical for the refinement of the productive process (Duchessi et al., 1989). In parallel, the
quality system “refers to all the physical aspects of the process and the product or service
delivered” (Corbett and Wassenhove, 1993, p. 109). The quality of the
xxxi) product or service is closely linked to firm performance and the system designed to achieve it (Capon
et al., 1990; Roth and Miller, 1992). Finally, reducing production costs is at the heart of
any competing firm. It refers to all direct and indirect costs involved in the production
process. Low production costs are positively associated with firm’s performance, whether
it is an old or a new company (Terjesen et al., 2011).
3.3 Managerial capability

Managerial capability refers to the skills, knowledge, experience and routines that a firm
uses to efficiently coordinate the capabilities regarding its other activities. This capability
also aims to minimize the internal frictions in different areas of the company, and a firm
with well-developed management skills can achieve efficiency gains in all areas. Strategy
planning, human resources and norms and procedures are the core elements of the
managerial capability.
Strategy planning refers to the pattern of decisions made by the firm to define and
disclose its objectives, intentions or goals, which creates the main policies and plans to
attain them (Andrews, 1980). This is usually applied throughout the firm, while business
strategy, being less comprehensive, defines the choice of products or services and
individual business markets within the firm (Ansoff, 1965). According to Penrose (1959),
a firm is a collection of production assets (human and non-human) under administrative
coordination. Resources, particularly human resources, are the essence of firms’ growth.
For Barnard (1938), companies are organizations that have formal procedures to achieve
their goals. A formal organization is a cooperation system among men that is conscious,
deliberate and intentional, hence more effective. Norms and procedures are the key to an
organization’s formalization.
3.4 Transactional capability

Finally, transactional capability is defined as a set of skills, knowledge, experience and
routines of a firm that allows it to minimize its transaction costs, whether regarding its
acquisitions from suppliers or its sales to customers (Tello-Gamarra and Zawislak, 2013).
As transactions are a complex activity that goes beyond the simple relationships of buying
and selling, this capability refers to gathering information from suppliers and consumers
to find the best sources and markets, as well as the most suitable prices (Tello-Gamarra
and Zawislak, 2013). In short, a firm’s success necessarily involves the ability to place its
products on the market, on a commercial scale at the lowest cost. Transactional capability
ensures that the firm’s path to the market is more efficient through customer relationship,
bargaining power and contract management.
Managing customer relationship is critical to firm’s success (Reinartz et al., 2004), and
establishing actions to enhance this relationship should be a priority task. Programs focused on
improving loyalty and affective commitment influence both customer retention and the amount
of purchases by each customer (Verhoef, 2003). Bargaining power is the ability to influence the
terms and conditions of a contract (with both customers and suppliers) for one ’s own benefit
(Argyres and Liebeskind, 1999). The bargaining power of a firm reduces its transaction costs
and its governance structure (Bosse and Alvarez, 2010). According to
Aoki et al. (1989), the firm is defined as a network of contracts. Therefore, to ensure its Innovation
existence, it needs to organize a set of contracts (with suppliers and customers) at the capability in
lowest possible cost. The contract is a legal instrument used by firms as a safeguard for
transactions. These occur both in the acquisition of inputs and in the sale of finished firms
products.
As each of these firm’s capability refers to a different set of knowledge and routines
that involve different stages of the techno-economic process of adding value, we present
in Table II four constructs (one for each capability) with their defining components, that 195
reveal the contours of a firm’s innovation capability.
These components were the basis for the analysis of the four case studies of selected
firms. Assuming their technological-intensity levels according to Organization for
Economic Cooperation and Development (OECD) (2003, 2007), classification and
contrasting them to the innovation capability profiles of each firm, it was possible to
identify if there is an association between these two constructs.
Capability Components Authors

Technological 1. Technological monitoring Griffin (1997), Davila (2000), Wong et al. (1998),
capability 2. Technology assimilation Huergo (2006), Rush et al. (2007), Christiansen and
3. Formalization of the Varnes (2009), Zawislak et al. (2012, 2013)
development process
Operational 1. Production planning Duchessi et al. (1989), Capon et al. (1990), Roth and
capability 2. Quality system Miller (1992), Corbett and Wassenhove (1993),
3. Reducing production costs Terjesen et al. (2011), Zawislak et al. (2012, 2013)
Managerial 1. Strategy planing Penrose (1959), Barnard (1938), Ansoff (1965),
capability 2. Human resources Andrews (1980), Zawislak et al. (2012, 2013)
3. Norms and procedures
Transactional 1. Customer relationship Williamson (1985, 1999), Aoki et al. (1989), Verhoef Table II. The
capability 2. Bargaining power (2003), Reinartz et al. (2004), Argyres and Liebeskind different
3. Contract (1999), Bosse and Alvarez (2010), Tello-Gamarra and components of a
Zawislak (2013), Zawislak et al. (2012, 2013) firm’s capabilities
4. The method
A multiple case study was conducted to explore the potential association between
technological intensity and innovation capability in firms from different industrial sectors.
This research design was chosen because innovation capability is a complex phenomenon,
according to the theoretical model presented in Section 3, which is the result of a
combination of technological, operational, managerial and transactional capabilities,
shown through their components (Table II).
Following Yin’s (2003) and Eisenhardt and Graebner’s (2007) observations regarding
exploratory research, we chose to study four Brazilian companies in industries whose
levels of technological intensity followed the Organization for Economic Cooperation and
Development (OECD) (2003, 2007) classification presented in Table I.
The selected companies, whose names have been replaced by the name of their
respective industrial sector, are shown in Table III. These companies were chosen by
convenience from a larger research database (Núcleo de Gestão da Inovação Tecnologica
[NITEC], 2015). First, because they belong to different sectors and
INMR also because each of them meets the major features that represent the distinct
15,2 technological intensities, which should lead to different innovation capability
arrangements.
Technological intensity Company Main product No. of employees Market

196
High Electronics Capacitors 1,700* International
Medium-high Chemicals Thermoplastic resins 1,600* International
Table III. Medium-low Metal Products Bolts for motors, cars and trucks 228** National
Characteristics of Low technology Textiles Sportswear 120** National
firms selected for the
study Notes: *Total employees in the subsidiary under study; **total employees in the entire company
We collected data in four stages. First, from secondary sources (information in firms’
websites, articles, annual reports, etc.), before the visits and interviews, to understand each
firm’s behavior.
Second, we carried out in-depth interviews with professionals that had an extensive
knowledge of the business, such as the owners, CEOs or directors. We used a semi-
structured questionnaire (Appendix) derived from the distinct components of each
innovation capability (as shown in Table II), according to the conceptual model by
Zawislak et al. (2012, 2013). These interviews were scheduled in advance and were
recorded. Before starting to record, we told the interviewee that data would be treated with
confidentiality, without disclosing the companies’ names.
Third, we visited the companies’ facilities. During the visits, we collected further
information on issues that were not fully covered during the interviews. The visits were
guided by the interviewees, and we could know the different areas of the company and
observe the existence of the four aforementioned capabilities. In addition, we asked more
questions about the firm’s behavior regarding innovation capability.
Shortly after interviewing and visiting the firms, we wrote visit reports that followed
the same structure of the research instrument, as part of the fourth stage. We also looked
for additional secondary information to provide a more complete picture of the firms and
their capabilities.
The four capabilities of the companies listed in the reports are presented in the Results
section. Each of the results was evaluated based on the level of each component of the
four capabilities, according to a consensual judgment by the three authors.
We defined the scores based on the triangulation of the collected information, through
an exhaustive review of secondary data prior to the interviews, the recorded interviews
(with directors, CEOs or owners), data collection during visits to the factories and
additional secondary data gathered after the interviews. From this set of information, we
defined three scores: high, medium and low. As firms had, in principle, the four
capabilities, and each capability was formed by three components, we evaluated each of
these components according to the following criteria:
(Y) High (score: 3): When the firm’s component exists and is visibly developed. For
example, regarding the technological monitoring component, if the firm has a vast
collection of routines that indicate that it monitors new technologies within the
sector.
*
Medium (score: 2): When the component exists but is moderately developed. For Innovation
technological monitoring, if the firm has some routines for doing it. capability in
*
Low (score: 1): When the component of the firm is scarce or does not exist; if the firms
firm has few or no technological monitoring routines.
The following section presents the results and discussion, which are organized according
to each firm’s capability, which, in turn, are made up of three components, as presented
before.
197
5. Results and discussion

We present and discuss the results based on the scores of each of the four capabilities that
compose the innovation capability as shown in Table II.
We carried out an analysis of each capability and its components. Thus, we made
parallel evaluations of the four capabilities in all firms, in each stratum of technological
intensity. At the end of the scores’ description for each type of capability and intensity, we
provide a summary of the association between technological intensity and innovation
capability. Table IV, at the end of this section, helps to summarize the scores of each
company and its innovation capability.
5.1 Technological capability

5.1.1 Technological monitoring. The four firms claimed to monitor new technologies. For
example, when asked about the attributes of the firm’s technological capability in relation
to such monitoring, the interviewee from Textiles commented:
[. . .]. We travel abroad, at least three times a year to Paris, New York or London, looking for
technology. Today the world is flat, and with the internet you can see the windows across the
country, you can search for fashion, and observe what’s new. So, before looking into magazines,
soap operas and customers [. . .] today it’s important to subscribe to websites, because through
the computer you can travel everywhere! So, we mix these travels increasingly, to open our
horizons and see what’s happening out there [. . .].
5.1.2 Technology assimilation. In the four firms, all interviewees claimed to assimilate
technologies for the business. However, each firm does it in different ways – by purchasing the
technology or developing its own. For example, the interviewee from Metal Products company,
which corresponds to the medium-low technological intensity stratum, first made the following
comment: “So, with the evolution of the company, its technology has changed a lot compared
to a few years ago”. Further on he continued: “Today, the machinery that we have in the
assembly line, with computer numerical control (CNC) machines together with conventional
ones, makes up a structure that hardly any competitor has”.
5.1.3 Formalization of the development process. The interviewees from the four firms,
regardless of their technological intensity stratum, claimed to keep records of the
development process of their products. However, there are different levels of
formalization. While Chemicals has a complex software to monitor the whole process of
new products’ development, Textiles conducts several routine activities in search of new
information to help the development process. When referring to the formalization of the
firm’s development process, the Metal Products interviewee commented:
So, this is how we develop the process: we have at least one electrical engineer (who knows
everything about electronics), a production engineer, a mechanical engineer (who understands of
mechanical assemblies, belts, gears, pulleys, shafts, various types of materials, their hardness,
who can help develop the product). Then we have 2 to 4 technicians, who work based on their
Brazil
Grande Do Sul -
of four firms in Rio
innovation capability
intensity and
technological
Table IV. Levels of
19
8
,2I
15
Technological capability Operational capability
Formalization of Technological Reducing
Technological Technology the development capability Production production Operational
Level of technological intensity/Company monitoring assimilation process total planning Quality costs capability total
High-technological intensity/Electronics 3 3 3 9 3 3 1 7
Medium-high-technological intensity/
Chemicals 3 3 3 9 3 3 1 7
Medium-low-technological intensity/
Metal Products 3 2 2 7 3 3 1 7
Low-technological intensity/Textiles 2 2 1 5 3 3 2 8
(continued)
Managerial Transactional
capability capability
Managerial
Production Human Norms and capability Corporate Bargaining Transactional Innovation
Level of technological intensity/Company planning resources procedures total strategy power Contract capability total capability
High-technological intensity/Electronics 3 2 3 8 3 3 3 9 33
Medium-high-technological intensity/
Chemicals 3 3 3 9 3 3 3 9 34
Medium-low-technological intensity/
Metal Products 2 2 2 6 2 2 2 6 26
Low-technological intensity/Textiles 1 2 2 5 3 2 1 6 24
Table IV.
199
yInnova
rmsfiinc
apabilit
tion
INMR knowledge, their experience [. . .]. Since we are always doing different things and sometimes
there is no time to set a big meeting, we have short meetings of just a few minutes with the
15,2 operators, who will produce the piece, modify it, etc., often in the production area. We also have
a designer in the group, to help develop a unit or a part.
5.2 Operational capability

*
5.2.1 Production planning. Production planning was found to be present in the four firms,
with some specific features in each one, such as the level of formalization, planning time,
dependence on inputs, optimization of equipment use, production start and planning
meetings. Regarding production planning, Electronics’ interviewee said:
Our production strategy is to use the equipment at its maximum capacity. To do that, we focus on
having qualified people, and this group is quite restricted. Today, we prefer people graduated in
Engineering and Administration. We focus especially on the Theory of Constraints, on the
bottlenecks, so we can use the resources at their best. In the factory, there are daily production
meetings around this “board” [. . .] what and how was the production the previous day at each
position within the factory; how many pieces were produced; what was the waste; were there any
problems with quality? So, these meetings of twenty minutes, half an hour at most, are attended
by team leaders, supervisors, technicians, engineers, managers, everyone around the “board”.
5.2.2 Quality system. Regarding this issue, all interviewees stressed their companies’
commitment to product quality. Quality is considered as an “admission ticket” to their
respective sectors. Firms believe that having high-quality standards is more than a matter
of choice, it is the minimum requirement for competing.
Electronics’ interviewee said:
Our company has always developed all its raw materials. As quality is our main focus, we need
raw materials that will lead to quality products. So, we established verticalization precisely in the
quality division, in order to have products with the highest quality.
In turn, Metal Products’ interviewee, when asked about the firm’s main policy, replied:
The issue of quality! First, due to the machinery we have in the production line today, which
many of our competitors do not possess. Thus, they cannot achieve better quality, earn as much
per piece, dedicate the same time as we do to each part, pass it through several stages until it
reaches a final quality. Our quality control inspects every piece (there are tolerance limits
regarding threads, gauges, depth), and if any item is outside the standards, which doesn ’t comply
with what was required by the design, it is rejected. The product is reworked or it ’s considered
useless. And a new one is made.
The interviewee from Chemicals said: “[. . .] quality control checks the properties to see if
they meet the product specifications; if so, it is packed and sent for sale”. Textile
representative stressed: “We don’t worry about the price, we worry about quality. For us it
is very clear! [. . .]. We start out from this principle: quality”.
5.2.3 Reducing production costs. Of the four firms, only one, Textiles, said it has a policy
for cost reduction. The other three (Electronics, Metal Products and Chemicals)
commented that their production costs were high. Electronics’ interviewee mentioned:
Our costs are higher. They are higher because we have some disadvantages compared to our
competitors. The first relates to the workforce, which is more expensive in Brazil than abroad.
The cost of an operator in Brazil is higher than in Hungary. The cost of an engineer in Brazil is
almost 1.5 times that of Hungary. These costs are a considerable weight. Another problem
regards our raw materials, of which 95 per cent come from Asia or Europe. In fact, our raw
material is from Asia. For us, it is very expensive compared to our competitors. And it is due to Innovation
transportation costs.
capability in
Metal Products’ interviewee said: firms
I believe our costs are a little higher. Today, I don’t know exactly what percentage they would be,
but our work hour costs a little more than that of the competition.
201
5.3 Managerial capability
5.3.1 Strategy planning. The four firms have established strategies, expressed in different
ways; either in terms of the market, as is the case of Electronics; as a result of an activity
(quality), as is the case of Metal Products; in terms of defined terms, as in the case of
Chemicals; or the focus on activity, as in Textiles.
Electronics’ strategy was described as follows:
Our company wants to be the market leader, the number one in the market, and feels that the
Asian market has become small for it. Our competitors are also expanding to other areas. That is
why we bought this unit in Brazil, because we were seeking to buy another company that was
already strong in a market where we wanted to grow.
Regarding its strategy, Metal Products interviewee said: “[. . .]. There are several, but the
main thing is to keep the company’s standard of quality, through quality management, cost
management, and process management”.
For Chemicals, strategy requires a period:
There is always a planning cycle for the next five years. We review the planning cycle and the
strategy each year, but the strategy horizon is ten years and the the planning cycle horizon is five
years.
On the other hand, Textiles mentioned:
We have been structuring ourselves, automating the company, since the beginning of 2006.
Everything is automated. Our parts are tracked down. From the moment we thought about it,
there was a need to cut costs immediately. We are prepared for the new Brazil that is emerging.
And we know there will be much competition ahead. So, for us this is the company’s strategy.
5.3.2 Human resources. With regard to human resources, the four companies confirmed
their key role. However, two of them gave greater emphasis to the importance of
employees for the company. The employee is not seen as a production factor to be simply
allocated, but as a key contributor to the company, who must develop along with the
company. That is, the firms need to grow, while ensuring the development of their
employees. Chemicals’ interviewee commented:
I would say that the major concern of the company, today, is that it will continue to grow, face a
lot of opportunities, so how does the company grow? Through people. Since we have a culture
based largely on people, to sustain our growth we must have outstanding people, well informed
and with an entrepreneurial profile. I think the challenge is to attract and keep the right people
aligned with our culture, with good training, and help the client grow the way we are growing.
Metal Products’ interviewee said that training employees is very important, that it is an
advantage over the competition. Further on, he said:
[. . .] when faced with more difficult products or services, our customers or even competitors
look for us, because they are not capable to do, they do not have the machines, the training or the
personnel for that, so we do the development for them.
INMR 5.3.3 Norms and procedures. Regarding this component, all four companies highlighted
15,2 different features of their norms and procedures that help to formalize daily work.
For example, Metal Products said: “For each production stage within each area of the
company, we have procedures and evaluation criteria”. This same firm commented on its
production process: “We have a schedule for production and an engineering room where I put
all the production schedules, with all the customers, within one month”. Regarding these
- items, Chemicals indicated a high degree of formalization.

Of the four companies, what draws most attention is the low-technological intensity
stratum. The Textiles firm states that its different areas are planned, and information flows
through a managerial system. Moreover, this firm attempts to control all its areas. The
interviewee said that:
Today our company is still a family business [. . .] but things are very well organized, even in the
administrative areas, specifically finance, accounting [. . .]. We have outside firms that do the
accounting, but everything is shown in our system.
And adds:
Everything is in the system [. . .]. It was developed in 2007 [. . .] since then we have made
changes that today help the general control of production and sales; it is not closely linked to
marketing, which is done separately, but internally, we are 100 per cent led by the control system,
everything goes through it.
5.4 Transactional capability

5.4.1 Customer relationship. The four companies carry out activities designed to keep a
good customer relationship and loyalty. Opposite to what we previously thought, both
firms of the high- and medium-high-technological intensity strata attempt to keep a good
relationship with their customers. That is, even though these firms belong to the highest
strata of technological intensity, they do not neglect the attributes regarding customer
relationship. For example, Electronics’ interviewee commented:
In our case we have a marketing center in Germany. And this marketing center is the main
contact with our corporate customers. For example, if I have a big client in Germany, and he is
developing a project, he calls us during the development stage (as he also calls our competitors)
and we provide technical and commercial support at the same time. But the solution is not given
(obviously) by the central, it comes from all over the world. So the engineering sta ff is always
involved with it, and always visits clients. It is part of our daily functions to visit and work with
customers at their facilities, to carry out the commercial and technical tasks. We make a lot of
visits to clients. And these visits may be everywhere in the world, so to speak. And we have this
focus, to try to understand what our customer will need in the future.
Concerning this same issue, Chemicals’ interviewee said:
The head of the hierarchy is the customer. It is the client who drives our actions. So, we have to
understand his needs and then bring him into the company.
5.4.2 Bargaining power. Regarding this component, the four firms state that they have
bargaining power with both customers and suppliers. This finding strengthens the idea that
transactional capability is present in all strata of technological intensity. To illustrate it, we
present comments made by Textiles interviewee, who theoretically would be the firm with
the lowest bargaining power. He says: “we don’t change a supplier because of a few reals
(Brazilian currency) on some raw material, we negotiate with him over the price”.
5.4.3 Contract. In relation to contracts, three of the four firms put great emphasis on this Innovation
activity – Electronics, Chemicals and Metal Products – whereas Textiles gives less capability in
importance to this transactional aspect. Electronics’ interviewee said: “Today our contracts
with customers have a failure rate of 1 per million. This is our goal and we achieve it”.
firms
It should be noted that contracts depend on several aspects, among them is the business
area, as in some sectors, the use of contracts is more common than in others. On this
subject, Metal Products’ interviewee emphasized: “We sign the contract directly with the
customer, only after that we start to work”. 203
5.5 The association between technological intensity and innovation capability

Table IV summarizes the main results of this exploratory study. Four companies were
chosen from the strata of technological intensity attributed to their respective industrial
sectors, according to classification of Organization for Economic Cooperation and
Development (OECD) (2003, 2007). The innovation capability of each company was
assessed based on the sum of its components, that is, the firm’s technological, operational,
managerial and transactional capabilities.
We observed that both the Electronics and Chemicals companies that belong to the high-and
medium-high-technological intensity strata have high levels of innovation capability. On the
other hand, the Metal Products company, from the medium-low-technological intensity
stratum, has a level of innovation capability similar to that of the Textiles company, which
belongs to a low-technological-intensity stratum, as seen in Table IV.
However, the association between technological intensity and innovation capability is
weak. Because firms belonging to strata of low-technological intensity also have the
innovation capability. Differences between firm’s innovation capability may be a
consequence of other aspects (e.g. development of other capabilities and scale firm).
6. Conclusion
Technological intensity is measured by the ratio of R&D spending to firm’s revenue in
different industries and was developed by the OECD to classify different industrial
sectors, in developed countries, in four levels: high-, medium-high-, medium-low- and
low-technological intensity.
Many innovation studies focus exclusively on high-technology companies that tend to
be innovative. This led to the assumption that firms from low-technological intensity strata
are not innovative.
Our research, which consisted of an exploratory case study of four companies, each
belonging to one of the four strata of technological intensity developed by OECD, and the
components of innovation capability contained in the model by Zawislak et al. (2012,
2013), were used to identify the level of innovativeness of the four companies.
The aim of this article was to find whether there is an association between technological
intensity and innovation capability. However, the association between technological intensity
and innovation capability is weak. Thus, the main result is that firms of the four strata of
technological intensity have innovation capability. Although firms with the highest
technological intensity have a greater innovation capability, firms of the low-tech strata also
have it. Therefore, firms can be innovative regardless of their stratum of technological intensity,
which shows the importance of other capabilities to ensure their performance.
We believe that the greater innovation capability of some firms is not necessarily because of
their technological intensity. These firms have previously accumulated other capabilities
(technological, operational, managerial and transactional) regardless of the stratum of
technological intensity to which they belong. This accumulation of capabilities allows them to
INMR achieve scale and grow in others markets. Coincidentally, in this study, firms that have the
15,2 greatest innovation capability are big companies with factories in several countries (Table III).
Although it is possible to notice differences in the arrangement of capabilities, according to the
sectors’ level of technological intensity, the four Brazilian companies show some homogeneity,
that is, there are no substantial differences between the capabilities of the sectors. This research
has some limitations that provide an opportunity for future studies. To this
- end, we suggest a survey with a much larger sample of companies, if possible representing
the whole population of firms, through a quantitative study that measures the
technological intensity of the sample, as well as the intensity of each component of the
four capabilities that comprise innovation capability. In addition, sectoral analyses could
be done, to seek different features that can justify innovation capability other than
technological intensity.
Note
← Palda (1986) advises to be careful when using the term “high technology”, since it has a very
popular and positive meaning for a firm or industry. Although the terms “high technology” or
“high-tech” do not have a simple or precise definition (computers are seen as “high-tech” but
atomic reactors are not), they can be considered equal to high technological intensity.
References
Alves, A., Barbieux, D., Tello-Gamarra, J., Reichert, F. and Zawislak, P. (2017), “Innovation and
dynamic capabilities of the firm: defining a model”, Revista De Administração De Empresas,
Vol. 57 No. 3, pp. 232-244.
Andrews, K. (1980), The Concept of Corporate Strategy, revised ed., Richard D. Irwin, Homewood, IL.
Ansoff, H.I. (1965), Corporate Strategy: An Analytic Approach to Business Policy for Growth and
Expansion, McGraw-Hill, New York, NY.
Aoki, M., Gustafsson, B. and Williamson, O.E. (1989), The Firm as a Nexus of Treaties, Sage
Publications, London.
Argyres, N. and Liebeskind, J. (1999), “Contractual commitments, bargaining power, and
governance inseparability: incorporating history into transaction cost theory”, Academy of
Management Review, Vol. 24 No. 1, pp. 49-63.
Barnard, C. (1938), The Functions of the Executive, Harvard University Press, Cambridge.
Bell, M. and Pavitt, K. (1995), “The development of technological capabilities”, in Ul Haque, I. (Ed.), Trade,
Technology and International Competitiveness, The World Bank, Washington, DC, pp. 69-101.
Bender, G. and Laestadius, S. (2005), “Non-science based innovativeness: on capabilities relevant to
generate profitable novelty”, Perspectives on Economic, Political and Social Integration, Vol.
11 Nos 1/2, pp. 123-170.
Bosse, D. and Alvarez, S. (2010), “Bargaining power in alliance governance negotiations: evidence
from the biotechnology industry”, Technovation, Vol. 30 Nos 5/6, pp. 367-375.
Burgelman, R., Maidique, M.A. and Wheelwright, S.C. (2004), Strategic Management of
Technology and Innovation, McGraw Hill, New York, NY.
Capon, N., Farley, J.U. and Hoenig, S. (1990), “Determinants of financial performance: a Meta-
analysis”, Management Science, Vol. 36 No. 10, pp. 1143-1159.
Chiesa, V., Coughlan, P. and Voss, C. (1996), “Development of a technical innovation audit”, Journal
of Product Innovation Management, Vol. 13 No. 2, pp. 105-136.
Christensen, J.F. (1995), “Asset profiles for technological innovation”, Research Policy, Vol. 24 No.
5, pp. 727-745.
Christiansen, J.K. and Varnes, C.J. (2009), “Formal rules in product development: sensemaking of Innovation
structured approaches”, Journal of Product Innovation Management, Vol. 26 No. 5, pp. 502-519.
Corbett, C. and Wassenhove, L. (1993), “Trade-offs? What trade-offs? Competence and competitiveness in
capability in
manufacturing strategy”, California Management Review, Vol. 35 No. 4, pp. 107-122. firms
Davila, T. (2000), “An empirical study on the drivers of management control systems’ design in new
product development”, Accounting, Organizations and Society, Vol. 25 Nos 4/5, pp. 383-409.
Desai, A. (1984), “India’s technological capability: an analysis of its achievements and limits”,
Research Policy, Vol. 13 No. 5, pp. 303-310. 205
Duchessi, P., Schaninger, C.M. and Hobbs, D.R. (1989), “Implementing a manufacturing planning
and control information-system”, California Management Review, Vol. 31 No. 3, pp. 75-90.
Eisenhardt, K. and Graebner, M. (2007), “Theory building from cases: opportunities and challenges”,
Academy of Management Journal, Vol. 50 No. 1, pp. 25-32.
Felsenstein, D. and Bar-El, R. (1989), “Measuring the technological intensity of the industrial sector:
a methodological and empirical approach”, Research Policy, Vol. 18 No. 4, pp. 239-252.
Forsman, H. (2011), “Innovation capacity and innovation development in small enterprises. a comparison
between the manufacturing and service sectors”, Research Policy, Vol. 40 No. 5, pp. 739-750.
Furtado, A.T. and Carvalho, R.Q. (2005), “Padrões de intensidade tecnologica da indústria brasileira: um
estudo comparativo com os países centrais”, São Paulo Em Perspectiva, Vol. 19 No. 1, pp. 70-84.
Griffin, A. (1997), “PDMA research on new product development practices: updating trends and benchmarking
best practices”, Journal of Product Innovation Management, Vol. 14 No. 6, pp. 429-458.
Griliches, Z. and Mairesse, J. (1984), “Productivity and R&D at the firm level”, in Griliches, Z. (Ed.),
R&D, Patents, and Productivity, University of Chicago Press, Chicago, pp. 339-374.
Guan, J. and Ma, N. (2003), “Innovative capability and export performance of Chinese firms”,
Technovation, Vol. 23 No. 9, pp. 737-747.
Hatzichronoglou, T. (1997), “Revision of the high-technology sector and product classification”,
[Working Paper no 1997/2], OECD Publishing, Paris.
Hirsch-Kreinsen, H., Jacobson, D., Laestadius, S. and Smith, K. (2005), “Low and medium
technology industries in the knowledge economy: the analytical issues”, in Hirsch-Kreinsen,
H., Jacobson, D. and Laestadius, S. (Eds), Low-Tech Innovation in the Knowledge Economy,
Peter Lang GmbH, Frankfurt am Main, pp. 11-30.
Huergo, E. (2006), “The role of technological management as a source of innovation: evidence from
Spanish manufacturing firms”, Research Policy, Vol. 35 No. 9, pp. 1377-1388.
Kafouros, M. (2005), “R&D and productivity growth: evidence from the UK”, Economics of
Innovation and New Technology, Vol. 14 No. 6, pp. 479-497.
Katz, J. (1984), “Domestic technological innovations and dynamic comparative advantage”, Journal
of Development Economics, Vol. 16 Nos 1/2, pp. 13-37.
Lall, S. (1992), “Technological capabilities and industrialization”, World Development, Vol. 20 No. 2,
pp. 165-186.
Markwald, R. (2004), “Intensidade tecnologica e dinamismo das exportações brasileiras”, Revista
Brasileira De Comércio Exterior, Vol. 79 No. 2, pp. 3-11.
Mendonça, S. (2009), “Brave old world: accounting for ‘high-tech’ knowledge in ‘low-tech’
industries”, Research Policy, Vol. 38 No. 3, pp. 470-482.
Núcleo de Gestão da Inovação Tecnologica (NITEC) (2015), “Caminhos da inovação na indústria
gaúcha”, available at: www.ufrgs.br/nitec/wp-content/uploads/2015/12/revista_inova.pdf
Organization for Economic Cooperation and Development (OECD) (2003), Science, Technology and
Industry Scoreboard, OECD, Paris.
Organization for Economic Cooperation and Development (OECD) (2007), Science, Technology and
Industry Scoreboard, OECD, Paris.
INMR Palda, K. (1986), “Technological intensity: concept and measurement”, Research Policy, Vol. 15 No.
4, pp. 187-198.
15,2 Patel, P. and Pavitt, K. (1997), “The technological competencies of the world’s largest firms: complex
and path-dependent, but not much variety”, Research Policy, Vol. 26 No. 2, pp. 41-156.
Penrose, E. (1959), The Theory of the Growth of the Firm, John Wiley & Sons, New York.
Reichert, F., Torgas, N., Zawislak, P. and Arundel, A. (2016), “Exploring innovation success recipes in
206 low-technology firms using fuzzy-set QCA”, Journal of Business Research, Vol. 69 No. 11, pp. 1-5.
Reinartz, W., Krafft, M. and Hoyer, W.D. (2004), “The customer relationship management process: its
measurement and impact on performance”, Journal of Marketing Research, Vol. 41 No. 3, pp. 293-305.
Roth, A. and Miller, J. (1992), “Success factors in manufacturing”, Business Horizons, Vol. 35 No. 4, pp. 73-81.
Rush, H., Bessant, J. and Hobday, M. (2007), “Assessing the technological capabilities of firms:
developing a policy tool”, British Journal Management, Vol. 37 No. 3, pp. 221-236.
Santamaría, L., Nieto, M.J. and Barge-Gil, A. (2009), “Beyond formal R&D: taking advantage of other sources
of innovation in low- and medium-technology industries”, Research Policy, Vol. 38 No. 3, pp. 507-517.
Segarra-Ciprés, M., Bou-Llusar, C. and Roca-Puig, V. (2012), “Exploring and exploiting external knowledge:
the effect of sector and firm technological intensity”, Innovation, Vol. 14 No. 2, pp. 203-217.
Shafia, M., Shavvalpour, S., Hosseini, M. and Hosseini, R. (2016), “Mediating effect of technological
innovation capabilities between dynamic capabilities and competitiveness of research and
technology organisations”, Technology Analysis & Strategic Management, Vol. 28 No. 7, pp. 1-16.
Srholec, M. (2007), “High-tech exports from developing countries: a symptom of technology spurts or
statistical illusion?”, Review of World Economics, Vol. 143 No. 2, pp. 227-255.
Teece, D. (1986), “Profiting from technological innovation”, Research Policy, Vol. 15 No. 6, pp. 285-305.
Tello-Gamarra, J. and Zawislak, P.A. (2013), “Transactional capability: innovation’s missing link”,
Journal of Economics, Finance, and Administrative Science, Vol. 18 No. 34, pp. 2-8.
Terjesen, S., Patelb, P.C. and Covina, J. (2011), “Alliance diversity, environmental context and the value
of manufacturing capabilities among new high technology ventures”, Journal of Operations
Management, Vol. 29 Nos 1/2, pp. 105-115.
Verhoef, P.C. (2003), “Understanding the effect of customer relationship management efforts on customer
retention and customer share development”, Journal of Marketing, Vol. 67 No. 4, pp. 30-45.
Von, T.N. and Acha, V. (2005), “Innovation in low-tech industries”, in Fagerberg, J., Mowery, D.C. and Nelson,
R.R. (Eds), The Oxford Handbook of Innovation, Oxford University Press, Oxford, pp. 407-432.
Wang, C.H., Lu, I.Y. and Chen, C.B. (2008), “Evaluating firm technological innovation capability under
uncertainty”, Technovation, Vol. 28 No. 6, pp. 349-363.
Wang, J. and Tsai, K. (2003), “Productivity growth and R&D expenditure in Taiwan’s manufacturing
firms [Working Paper no. 9724]”, National Bureau of Economic Research, Cambridge, MA.
Williamson, O. (1999), “Strategic research: governance and competence”, Strategic Managerial Journal,
Vol. 20 No. 12, pp. 1087-1999.
Williamson, O. (1985), The Economic Institutions of Capitalism, Free Press, New York, NY.
Wong, V., Shaw, V. and Sher, J. (1998), “Effective organization and management of technology
assimilation: the case of Taiwanese information technology firms”, Industrial Marketing
Wonglimpiyarat, J. (2010), “Innovation index and the innovative capacity of nations”, Futures, Vol. 42
No. 3, pp. 247-253.
Yam, R., Guan, J., Pun, K. and Tang, E. (2004), “An audit of technological innovation capabilities in Chinese
firms: some empirical findings in Beijing, China”, Research Policy, Vol. 33 No. 8, pp. 1123-1140.
Yin, R. (2003), Case Study Research: Design and Methods, Sage Publications, Newbury Park.
Zawislak, P.A., Alves, A., Tello-Gamarra, J., Barbieux, D. and Reichert, F.M. (2012), “Innovation
capability: from technology development to transaction capability”, Journal of Technology
Management & Innovation, Vol. 7 No. 2, pp. 14-27.
Zawislak, P.A., Alves, A., Tello-Gamarra, J., Barbieux, D. and Reichert, F.M. (2013), “Influences of the internal
capabilities of firms on their innovation performance: a case study investigation in Brazil”, International
Journal of Management, Vol. 30 No. 1, pp. 329-348.
Appendix
Research instrument
Briefly describe the most important facts in the company’s history Where
does the company’s knowledge come from?
How did the company develop knowledge and the techniques to do what it does? How would
you compare the company’s level of knowledge to its competitors?
Briefly describe the company’s commercial strategy
Briefly describe the relationship with suppliers and of purchases Briefly
describe the relationship with customers and of sales
What makes customers buy from you? How do
you define your prices?
What is the company’s commercial position compared to its competitors? Briefly
describe the company’s strategy
Briefly describe the company’s administrative processes
How are the company’s costs compared to its competitors? Briefly
describe the company’s productive strategy
Briefly describe the company’s productive process
How do you compare the company’s level of productive efficiency to the competitors? Briefly
describe the development strategy and decision-making
Briefly describe the technology development process
How do you compare the level of the company’s development activities to its competitors?
Give three examples of changes in the company
Give three examples of company’s innovations, and if they were new for the company, for the
sector, for the country or for the world
What kind of outcomes did the innovations cause for the company?
What is the company’s differential advantage to keep it competitive in the market?
What are the legal/institutional incentives or constraints for the company to innovate?
Rank in order of importance for innovation the following areas of the company: Technology,
Operations, Management and Commercial. Justify your choice.
Jorge Tello-Gamarra can be contacted at: jorgetellogamarra@gmail.com
Innovation capability in firms
207
J. Eng. Technol. Manage. 33 (2014) 32–46
Contents lists available at ScienceDirect
Journal of Engineering and

Technology Management
journal homepage: www.elsevier.com/locate/jengtecman
Intangible aspects of innovation capability in SMEs:

Impacts of size and industry
Minna Saunila *, Juhani Ukko
School of Industrial Engineering and Management, Lappeenranta University of Technology, Lahti
School of Innovation, Saimaankatu 11, FI-15140 Lahti, Finland
ARTICLE INFO ABSTRACT
Keywords: In the current literature, developing innovation capability in an organization is

Innovation capability increasingly important. The approach of this study is quantitative. The data for
SME the study were gathered using a web-based questionnaire targeting Finnish SMEs
Size
employing 11–249 persons and with less than 50 Meuro in revenue. A sample of
Industry
2400 SMEs was randomly selected. The response rate was 7.68 percent. This
paper contributes to current understanding by presenting a construct for
categorizing the intangible aspects of organizational innovation capability. The
study also reveals that a firm’s size or industry does not have a remarkable effect
on the firm’s innovation capability.
2014 Elsevier B.V. All rights reserved.
Introduction
The successful operation of firms in almost all industries is becoming highly dependent on the firms’ abilities to
produce innovations. Innovation is a process of turning opportunities into new ideas and turning these ideas into
widely used practices (Tidd et al., 2005). Innovation is more than just a great idea; it is the opportunity to solve a
problem that matters. The key is executing an idea in useful practice. Scholars have suggested that innovation
capability is a multi-faceted construct. The categories used for innovation capability often adopt a certain type of
innovation, such as product innovation, instead of the overall innovation capability (Ibrahim et al., 2009). In addition,
innovation capability has been divided into radical and incremental innovation capabilities (Sen and Egelhoff, 2000).
xxxii) Corresponding author. Tel.: +358 40 518 7841.

E-mail addresses: minna.saunila@lut.fi (M. Saunila), juhani.ukko@lut.fi (J. Ukko).
http://dx.doi.org/10.1016/j.jengtecman.2014.02.002
0923-4748/ 2014 Elsevier B.V. All rights reserved.
M. Saunila, J. Ukko / Journal of Engineering and Technology Management 33 (2014) 32–46 33
Moreover, the current literature has concentrated on evaluating a firm’s innovation capability by defining the types of
capabilities that form the overall innovation capability. For example, product innovation capability, process innovation
capability, market innovation capability, strategic innovation capability, organizational capability, manufacturing
capability, networking capability, entrepreneurial capability, and R&D capability have been studied (see Christensen,
1995; Guan and Ma, 2003; Wang and Ahmed, 2004; Forsman, 2009). However, no dominant theoretical perspective
integrates the individual sections of innovation research (Drazin and Schoonhoven, 1996). Therefore, innovation
capability needs to be defined through an empirical study using existing innovation research.
The concept of innovation capability presented in this paper is intangible because innovation capability refers to the
potential to create innovations. Due to the special features of SMEs, the potential for innovation is observed as more
important than the commercial end. The scarcity of resources, including human resources (both management and
personnel), financial capital, time, and security, has been considered one of the features of SMEs (e.g., Singh et al.,
2008; Ates and Bititci, 2011). Although size represents a weakness in terms of available resources, it favors a flat
organizational structure with a lack of bureaucracy. This size dependence allows for flexibility, adaptability and speed
in responding to the changing environment (Garengo et al., 2005). For this reason, SMEs usually have a high potential
for innovation, which may result in types of innovations other than just commercial products. Innovation capability,
similar to intangibles in general, is hard to specify directly, but it can be specified by defining closely related aspects.
These aspects of innovation capability are also innovation activity inputs. According to Davila et al. (2006), the inputs
are the resources dedicated to creating innovations. These inputs may be tangible, such as people, money, time,
equipment, or intangible, such as motivation, knowledge and firm culture. Many studies have presented intangible
aspects related to innovation capability (Lawson and Samson, 2001; Martensen et al., 2007; Skarzynski and Gibson,
2008; Tura et al., 2008; Paalanen et al., 2009) but often from a theoretical perspective. The results have seldom been
based on empirical studies, as is the case in this study. In addition, this study caters to various aspects of innovation
capability, departing from the majority of existing empirical studies that focus only on one or two intangible aspects of
innovation capability (e.g., culture, structures).
The objective of this study is to define the intangible aspect of a firm’s innovation capability using a questionnaire
targeting Finnish SMEs. The paper contributes to current understanding by presenting a construct for categorizing the
intangible aspects of a firm’s innovation capability in practice. In this respect, the nature of the study is explorative.
Another purpose of the paper is to clarify whether innovation capability differs with the size and industry of the firm.
The results contribute to the existing discussion on innovation capability by diminishing the gap between theory and
practice and by building requisites for further research.
Literature review
Definition of innovation capability
According to Lawson and Samson (2001), innovation capability is a theoretical framework aimed at describing the
actions that can be taken to improve the success of innovation activities. Innovation capability consists mainly of the
firm’s intangibles. Intangibles are the non-physical characteristics of a firm, which will produce value in the future
(Kannan and Aulbur, 2004). Intangibles, which are referred to as potential in this paper, are not assets as such.
Exploiting intangibles generates results (Bontis, 2001; Marr, 2007). Innovation capability is composed of the main
processes within the firm and cannot be separated from the main practices because innovation capability is the potential
to carry out the practices. Neely et al. (2001) also suggest that a firm’s innovation capability is the potential to generate
innovative outputs. Similarly, Lawson and Samson (2001, p. 384) define innovation capability as ‘‘the ability to
continuously transform knowledge and ideas into new products, processes and systems for the benefit of the firm and
its stakeholders.’’ Saénz et al., 2009 consider innovation a dynamic capability with multiple aspects (i.e., a capability
that allows the firm to integrate, build, and reconfigure internal and external competences to address rapidly changing
Jurnal Dinamika Manajemen, 9 (1) 2018, 114-125
http://jdm.unnes.ac.id
Nationally Accredited based on the Decree of the Minister of Research,
Technology and Higher Education, Number 36a/E/KPT/2016
How Does Knowledge Absorption Foster Performance?

The Mediating Effect of Innovation Capability
Heru Sulistyo1 , Sri Ayuni2 
Faculty of Economics, Universitas Islam Sultan Agung, Semarang, Indonesia
Info Article Abstract

History Article: This research aims to examine the impact of knowledge sharing and knowledge absorption on the
Received December 2017
innovation capability and the performance of Sharia banks. The rapid competition in banking
Approved January 2018
industry among conventional and Sharia banks requires innovation capability in order to enhance
Published March 2018
the performance and achieve the sustainable competitive advantage. The develop-ment and
Keywords:
performance of Sharia banks at the present still tend to be low. Innovation on product and service
Knowledge Sharing; Knowledge
Absorption; Innovation Capabil- is one of the success keys of Sharia banks in enhancing the performance and com-petitiveness.
ity; Employee Performance. Innovation capability can be developed if the employees within the organization are able to
perform knowledge sharing or knowledge absorption. The sample of this research includes the
employees of Sharia Banks in Semarang as many as 102 respondents. The sampling technique
uses purposive sampling with criteria; permanent employees who have been work-ing no less than
5 years. The data collection method is performed using questionnaire, while the data analysis
technique uses SEM. The finding concludes that knowledge sharing and knowl-edge absorption
significantly impact the innovation capability and performance.
Bagaimana Knowledge Absorption dapat Meningkatkan

Kinerja? Efek Mediasi dari Innovation Capability
Abstrak
Tujuan penelitian ini adalah untuk menguji dan menganalisis pengaruh knowledge sharing dan
penyerapan pengetahuan terhadap innovation capability dan kinerja Bank Islam.Persaingan in-
dustry perbankan antar Bank konventional maupun bank islam yang tinggi menuntut kemamp-
uan inovasi agar dapat meningkatkan kinerja dan mencapai keunggulan bersaing yang berkelan-
jutan. Perkembangan dan kinerja Bank Islam saat ini masih relative rendah bila dibandingkan
dengan bank konvensional. Inovasi produk dan jasa merupakan salah satu kunci keberhasilan
Bank Islam dalam meningkatkan kinerja dan daya saing. Kemampuan inovasi dapat ditingkat-
kan bila karyawan dalam organisasi mampu memiliki knowledge sharing maupunpenyerapan
pengetahuan. Sampel dalam penelitian ini adalah karyawan Bank Mandiri syariah, Bank BNI
Syariah, dan Bank BRI syariah Di Semarang sebanyak 102 responden. Teknik sampling meng-
gunakan purposive sampling dengan kriteria karyawan tetap yang memiliki masa kerja lebih dari
lima tahun. Metode pengumpulan data dengan menggunakan kuesioner dan teknik analisis data
menggunakan SEM. Hasil penelitian menyimpulkan bahwa knowledge sharing dan peny-erapan
pengetahuan berpengaruh signifikan terhadap kapabilitas inovasi dan kinerja.
JEL Classification: M5, M51

Correspondence Address ISSN
FE Unisula Jl. Raya Kaligawe km 4 Semarang 2086-0668 (print) 2337-5434 (online)
Email: heru@unissula.ac.id DOI: 10.15294/jdm.v9i1.14657
Heru Sulistyo & Sri Ayuni/ How Does Knowledge Absorption Foster Performance?...
INTRODUCTION management (Battor & Battor, 2010); informal

social interaction (Liu et al., 2015). One of the
The growth of sharia banking industry in important factors in promoting innovation ca-
Indonesia is still facing several obstacles, such as the pability in sharia banking industry is the know-
low quality of human resources who under-stand the ledge of human resources. Knowledge as a part
sharia economics, perform less opti-mal service, and of an organizational asset has a positive impact
have low integrity and product innovation. on innovation, superior performance and com-
Although the financial assets of sha-ria banks in petitive advantage (Andreeva & Kianto, 2011).
Indonesia reach IDR 897.1 trillion A successful company should have the
in the financial services authority (OJK), the ability to collect, store and distribute specific
market share is only able to reach 5.18%. Based knowledge to create and sustain a competitive
on OJK’s data for the Quarter III 2016, the advantage. Knowledge will not be able to en-
fi-nancing grows at around 14 each year while courage the creation of innovation if it cannot
the third-party funds (DPK) grow above 20%. be absorbed or shared among employees. This
Human resources who understand banking are is also happening in the sharia banking industry,
21.80%. This indicates that sharia banking where the majority of human resources do not
needs to improve the innovation capability of have enough knowledge of Sharia products.
human resources in improving the performance Knowledge interaction among members of the
and competitive advantage. Creation of inno- organization always requires knowledge, mas-
vation can facilitate and improve job or service. tery, and understanding of the concept, hence to
Liao et al. (2009) stated that innovation mostly facilitate the work of the organization. Kno-
focuses on development or technology upgra- wledge sharing is considered to be the most
ding, or information and product development. important aspect of knowledge management
On the other side, Al. Othman and Sohaib (Almudallal et al., 2016; Elianto & Wulansari,
(2016) stated that innovation is a new idea of 2016).
recombination of old ideas, schemes, formulas The study of knowledge sharing has done
or unique approaches which are then conside- by Pai and Chang (2013); who found that kno-
red to be new by individuals involved and can wledge absorption influences knowledge sha-
be developed internally. ring activities. Knowledge sharing can drive
Innovation capability is the ability to de-velop improvements in innovation and organizational
new products that meet market needs, develop and performance (Lin, 2007; Al-Hayaly & Alnajjar,
adopt new products and process technologies, as 2016) The results of research conducted by Liao
well as respond to unexpected activities performed (2006) and Jaberi (2016) concluded that
by competitors (Tsai & Tsai, 2010). Innovation knowledge sharing activities positively affect
capability eases the improvement of work and the improvement of organizational innovati-on.
service. A number of previous studies on factors The study results of Andreeva and Kianto (2011)
affecting the capabi-lities of innovation have been found a positive relationship between knowledge
conducted as fol-lows; marketing and sharing and innovation. Akram (2017) also found
entrepreneurship capabili-ties (Sulistyo & that knowledge sharing has a positive effect on
Siyamtinah, 2016); knowledge sharing (Lin, 2007); innovation. However, the-re are differences in
customer relationship management (Lin et al., 2010); the results of studies on the effect of
organizational culture and empowerment (Cakar & knowledge sharing on innovation. Results of
Ertürk, 2010); Islamic Work Ethic (Kumar & Rose, research conducted by Malkawi and Rumman
(2016) found that knowledge sharing does not
2012); relationship management (Garrido-Mo- affect innovation. Therefore, this study aims
rino et al., 2015); psychological empowerment to examine the effect of knowledge sharing
(Ertürk, 2012); performance measurement and knowledge absorption on the capability and
(Saunila & Ukko, 2012); customer relationship performance of sharia banking industry. Many
115
studies have been conducted (Fındıklı et ding that widely spread. The process of know-
al., 2015) on manufacturing and bureaucratic ledge sharing is conceptualized in two dimen-
sec-tors. However, it is still rarely used in sions: sharing the knowledge and collecting the
service industries, such as sharia banks. knowledge (Hooff & Ridder, 2004). Knowled-
ge sharing is defined as the individual
Hypothesis Development process of communicating personal intellectual
Innovation Capability capital to others, while knowledge collecting is
Innovation capability is the implemen- defin-ed as the process of peer-to-peer interaction
tation and creation of technology applied to to encourage them to share their intellectual capi-
new systems, policies, programs, products, tal (Hooff & Ridder, 2004; Lin, 2007).
processes and services to an organization (Idd- Knowledge sharing is considered to be an
ris, 2016). Innovation capability is also an abi- important process of knowledge manage-ment.
lity to absorb and use external information for Knowledge sharing is defined as an act of
then transfer it into new knowledge (Cohen transferring or disseminating knowledge from one
s Levinthal, 1990). Innovation capability is a person, group, and organization to anot-her. In an
comprehensive set of characteristics of the individual perspective, Knowledge sharing means
organization that facilitates and drive innova- individual willingness to share what they have in
tion strategies. Innovation as a modification an organization. Knowledge sharing has
of products, processes, services, organizational benefits both in individual and or-ganizational
systems, and marketing systems to create cus- levels. In individual level, opportu-nities for
tomer value (O’Cass & Weerawardena, 2009). employees to develop their skills and performance
Innovation capability consists of technical in- will be provided by the knowled-ge management
novation and administrative innovation (Ni-sula when these employees work together and share
& Kianto, 2013). their knowledge. Knowled-ge sharing is one of
Innovation capability as a company’s the main processes of KMS (Knowledge
performance through various types of innova- Management System) of the or-ganization (Yeşil
tion to achieve an overall improvement in in- & Dereli, 2013). Knowledge sharing allows
novation capability (Liao et al., 2009). Innova- both parties not only to store information but also
tion must use the production and marketing of to strengthen and expand through the exchange
technology to produce new products or servi- process.
ces to customers, or attribute new products to The research conducted by (Yeşil & De-
customers. Kashan and Mohannak (2015) sug- reli, 2013) found that knowledge collecting
gests that organizational innovation involves significantly has the positive relationship with
generating or adopting new ideas or behaviors innovation capability, while knowledge dona-ting
to become new products or services, new pro- does not affect innovation capability. The
duction technologies, operating procedures or results of a study conducted by (Liao, 2006)
new strategies or new management strategies. concluded that knowledge sharing activities have
a positive effect on the improvement of
Knowledge Sharing and Innovation organizational innovation. The results of the study
Capability Knowledge sharing (KS) is an by Andreeva and Kianto (2011) found a positive
integral component of knowledge management relationship between knowledge sha-ring and
that helps in transforming individual knowledge innovation. Akram (2017) also found that
into organizational knowledge and improving knowledge sharing has a positive effect on
company performance (Foss et al., 2010). KS innovation. Knowledge sharing has a signifi-
can improve the organization’s knowledge- cant effect on innovation capability (Kumar
based resources/ and lead to the improvement and Rose, 2012; Byukusenge et al., 2016).
of work result through the use of information, H1: Knowledge sharing significantly affects the
experience, practice, insight, and understan- innovation capability.
116
Knowledge Absorption and that the capacity of knowledge absorption is

Innovation Capabi-lity very important in predicting organizational
Knowledge absorption is one of the ca-pability (Henttonen et al., 2016).
important issues in knowledge management which H2: Knowledge absorption significantly affects
plays role in building knowledge-based system. the innovation capability.
Absorption refers to the process of gai-ning
knowledge into an organization from ex-ternal Knowledge Sharing and Performance
sources through any possible way. So it can be Employees in organizations willing to
stated that knowledge absorption is an activity of share information and knowledge with collea-gues
seeking and gaining knowledge into knowledge- and work together will contribute to the
based resources Radovanić and Matović improvement of organizational knowledge and
(2016), knowledge absorption is a way to organizational performance. Donation of know-
acquire, change, and transfer expertise from ledge among individuals within the organization
knowledge sources to knowledge-based sys-tems. generates group and organizational knowledge,
Knowledge absorption and creation are first steps thereby increasing the inventory of knowledge
in knowledge development process (Liao et available to companies. Organizations that pro-
al., 2009). mote employees to contribute knowledge in
Knowledge absorption is the first groups and organizations tend to generate new
activity in series of activity of receiving ideas and develop new business opportunities,
knowledge from the external environment and thus facilitating innovation activities and impac-
transforming it into a representation that can be ting performance improvements. Research con-
internalized and/or be used in an organization. ducted by Henttonen et al. (2016) concluded that
Knowled-ge absorption means transferring knowledge sharing significantly affects the
knowledge resources among firms with the employee performance. Research conducted by
aim of acquiring knowledge for learning. Kwahk & Park (2016) found that knowled-ge
knowledge absorption technique is a skilled sharing activities significantly affect the job
method captured from knowledge sources and the performance The research of Wang and Yang
source can be hu-man, book, journal, database, (2016); Arizqi (2017) concluded that innovati-on
report or compu-ter system. The ability of capability initiates the influence of knowled-ge
knowledge absorption own by employees in the sharing on performance.
organization will facili-tate the interaction and H3: Knowledge sharing significantly
exchange of knowledge within the organization affects the performance.
which then impact on improving innovation
capability. Liao (2008) found that knowledge Knowledge Absorption and Performance
absorption significantly affects innovation. The capacity for knowledge absorption is
The study by Liao (2008) suggests that the company’s ability to utilize externally held
knowledge absorption capacity has a significant knowledge through three consecutive processes
and positive influence on innovation capabili-ty. of recognizing and understanding potentially
Thus, the capacity of knowledge absorption plays an valuable new knowledge outside the company
important role in improving the capabi-lities of through exploratory learning. The second pro-
innovation. The higher capacity of kno-wledge cess involves the assimilation of valuable new
absorption will increase the company’s innovation knowledge through transformative learning as
capability. Research conducted by Cohen and well as and using assimilation of knowledge to
Levinthal (1990) found that inc-reased absorption create new knowledge and commercial out-
capacity (Nodari et al., 2016) of knowledge will comes through exploitative learning. The higher
have an impact on improving innovation. Similar the level of absorption of employee knowledge
research results also perfor-med Muskat & Deery in the organization, the higher the achievement
(2017) which concluded of employee performance in the organization.
117
Figure 1. Research Model
The results of the study found that the within the company will improve business per-
higher absorption capacity of employees in formance. The research conducted by Sulistyo
the organi-zation to knowledge will and Siyamtinah (2016) found that innovation
increase the ideas, new ideas of employees capability affects the improvement of emplo-
to product innovations and processes that yee performance. Battor and Battor (2010) also
impact on performance improve-ment. found the same results, the higher the company’s
The results of research conducted Li et al. innovation will improve performance.
(2011) found that higher knowledge absorp-tion H5: Innovation capability affects the perfor-
capacity will improve innovation perfor-mance. mance.
Liu et al. (2015) examine the effect of the
decreased absorption capacity of knowledge on METHOD
performance degradation. The results show that
the higher the rate of the decreased absorp-tion Sample
capacity of knowledge will increase the per- The sample of this research is employees of
formance degradation. Nodari et al. (2016) also three Bank Mandiri sharia, Bank BNI Sharia, and
found that high knowledge absorption capacity Bank BRI sharia In Semarang as many as 102
will improve performance. respondents. The sampling technique uses
H4: Knowledge absorption significantly affects purposive sampling with criteria of permanent
the performance. employees who have the working period of more
than five years. The selection of purpo-sive
Innovation Capability and Performance sampling techniques to ensure the selected sample
The ability of innovation is an important is a permanent employee who has a working
driving force in improving performance. Com- period of more than five years has suf-ficient
panies that want to improve innovation perfor- experience in gaining various knowledge
mance, it must have innovation capability. The from both external and internal sourced organi-
capability of innovation is the ability to develop zations, so that knowledge sharing process will
and respond to the development of new pro- take place within the organization and impact on
ducts in accordance with market demand and capability innovation and performance will be
respond to technological activities undertaken visible. Data collection using questionnaires
by competitors. Thus the capabilities of inno- distributed with the help of human resources staff
vation as an important driving force in impro- in each Bank for 1 month. Subsequently, the
ving performance. The study by Tsai and Tsai collected questionnaires have verified the
(2010) found that high innovation capabilities eligibility for data analysis.
118
Measurement Data analysis

Knowledge sharing was measured by four Data analysis is done by using descriptive
item adapted from Liao et al. (2007). A Samp-le analysis and partial least squares analysis (PLS)
item is “I often share with my colleagues the is used to test the structural equation of
new information”. Knowledge absorption was influen-ce among variables (inner model) either
measured by three item adapted from Liao et al. know-ledge sharing management, knowledge
(2009). A Sample item is “I actively use various absorp-tion, innovation or performance and to
sources of information to support the comple- test the validity and reliability of research
tion of work. Innovation capabilities were me- instrument (Outer model).
asured by four item adapted from Andreeva and
Kianto (2011). A sample item is “I often use RESULT AND DISCUSSION
new ideas to get things done”. The Perfor-
mance was measured by four item adapted from Based on the respondents’ answers, it is
Janssen (2001). A Sample item is “I am able to acquiring the mean value and standard devia-tion
complete the quantity of work assigned by the elaborated in Table 1. Based on the desc-riptive
company.” Participants provided their respon- data (Table 1), it can be concluded that all
ses on five-point Likert-type scales (1 = variables are included in the category of high,
Strong-ly disagree; 5 = Strongly agree). which means that knowledge sharing, knowled-
Table 1. Mean Value and Standard Deviation
Variable Mean Standard Deviation

Knowledge Sharing 3.81 0.48
Knowledge Absorption 3.82 0.41
Innovation Capability 3.91 0.45
Performance 3.87 0.52
Table 2. Validity and Reliability
Variable/indicator Original Sample Standard T-Statistic Composite

estimate deviation Reliability
KNOWLEDGE SHARING 0.818
KS1 0.534 0.094 5.670
KS2 0.795 0.050 15.944
KS3 0.818 0.065 12.536
KS4 0.743 0.094 7.909
KNOWLEDGE ABSORPTION 0.805
KA1 0.707 0.086 8.243
KA2 0.832 0.049 16.859
KA3 0.741 0.092 8.100
INNOVATION CAPABILITY 0.765
IC1 0.488 0.209 2.340
IC2 0.726 0.113 6.406
IC3 0.710 0.062 11.527
IC4 0.741 0.057 13.119
PERFORMANCE 0.873
P1 0.696 0.104 6.688
P2 0.854 0.070 12.198
P3 0.843 0.071 11.882
P4 0.781 0.055 14.071
119
Structural Equation (Inner Model)
Figure 2. The Result of Structural Equation Model
Tabel 3. The Result of an Inner Model
Original Sample Mean of Standard

Estimate Subsamples Deviation T-Statistic
Knowledge -> Innovation 0.485 0.508 0.111 4.386
Absorption -> Innovation 0.302 0.306 0.114 2.643
Knowledge -> Performance 0.438 0.383 0.110 3.994
Absorption -> Performance 0.273 0.324 0.178 1.535
Innovation -> Performance 0.235 0.255 0.119 1.970
ge absorption, innovation capability and capability, so H2 is supported, T statistic value

perfor-mance have been well implemented 2.643 > 1.96. The value of original sample esti-
in sharia banks. mate is 0.302 which means that the higher the
knowledge absorption in the organization, the
Validity and Reliability higher the innovation capability will be.
All indicators of each variable are valid The results also show that knowledge
with T statistic > T table (1.96). Reliability test sharing significantly affects the performance, so
results also show that all variables are reliable that H3 is supported, with T statistic value 3.994
which is shown with composite reliability value (Z) 1.96. The value of original sample
that is higher than 0.70 (Table 2). estimate is
Table 3 shown the result of testing the 0.438 which means that the higher the know-
relationship among variables using Partial Least ledge sharing in the organization, the higher the
Square shows that knowledge sharing signifi- performance will be. Knowledge absorption has
cantly affects the innovation capability, hence no effect on performance, so H4 is rejected,
H1 is supported, with T statistic value 4.386 this is shown by T statistic value 1.535 < 1.96.
xxxiii) 1.96. The value of original sample Inno-vation capability has the significant
estimate is 0.485 which means that the higher the effect on the performance, so H5 is supported,
kno-wledge sharing in the organization, the higher with a T sta-tistic value of 1.97 > 1.96. The
the innovation capability will be. Knowledge original value of sample estimate is 0.235 which
absorption has a significant effect on innovation means that the higher the innovation capability,
the higher the performance will be.
120
The R Square value of innovation ca- knowledge donating does not affect innovation
pability is 0.52, which means 52% variation capability. Usman and Mat (2017) also found
of innovation capability can be explained by that knowledge sharing has a positive effect
the variation of knowledge sharing and on innovation. Research conducted Kumar and
knowledge absorption. 48% is explained by Rose (2012) found that knowledge sharing has a
other variables which are not included in the significant effect on innovation capability.
model. Based on R square performance, 70% The absorption of knowledge has a sig-
performance va-riation can be explained by nificant effect on innovation capability. A good
knowledge sharing, knowledge absorption, knowledge absorption by employees as a capital for
and innovation capabi-lity. sharing knowledge among them will enhan-ce their
The results showed that knowledge sha- innovation capability. Employees who are always
ring had a significant effect on the improve- active using a variety of information sources, new
ment of innovation capability. Employees who knowledge and new information will easily absorb
are always active share their knowledge gained new knowledge which impacts on innovation
from various training, volunteering experien-ce capability either product innova-tion, process,
to co-workers will encourage the process of service, and administration. The results support
dissemination and absorption of knowledge in the findings of Muskat and Dee-ry (2017) that
the organization. The knowledge that has been knowledge absorption has a sig-nificant effect on
absorbed will encourage employees to impro-ve innovation. However, the re-search results are
the capabilities of innovation, both product not in line with the findings of Liao et al. (2011)
innovation, process. In a knowledge-based that knowledge absorption is insignificant to
economy, the ability of sharia banks to compe- innovation.
te and achieve competitive advantage depends The argument of Liao et al. (2011) is pro-
on sustainable product and service innovation. found knowledge, especially when knowledge is
Innovation capability can be created if the bank gained through interaction with external orga-
has knowledge of both explicit knowledge and nizations, thinking ability will increase. Know-
tacit knowledge. The results show that the more ledge can be obtained from inter-organization or
the employees of sharia bank share knowledge intra-organization, the existence of the mutu-al
through training, work experience, and sha-ring, trust, interaction among members, and from the
the more the innovation capability will be. mental capital that is needed to achieve
Through interaction and transfer of knowledge innovation. The research results on knowled-ge
(knowledge donating and knowledge collec- sharing behavior also cannot affect human
ting) among employees in the organization, it resource innovation by finding creative ideas,
will develop knowledge to create ideas or imp- developing creative ideas and new technologies.
rove working method, so it will impact on per- Knowledge sharing activities with dimensions of
formance improvement. ability to share knowledge and motivation to share
The results support the findings of Liao knowledge can only be acquired from knowledge
(2006) that knowledge sharing activities have sharing activity through motivati-on which
the positive effect on the improvement of orga- significantly affects innovation, while
nizational innovation. The results also support knowledge sharing activity through knowledge
the findings of Andreeva and Kianto (2011) and sharing ability does not affect innovation.
Akram (2017) who found that knowledge sharing An organization needs a rapid develop-
positively affects innovation. The re-sults of the ment of new knowledge that employees can
study also support the findings of the research absorb well, making it difficult to imitate by
conducted by Yeşil and Dereli (2013) found that competitors. The high ability of knowledge
knowledge collecting has a positive relationship with absorption will encourage the emergence of
innovation capability, while new ideas, unique ideas that have an impact on
121
performance improvement. Nevertheless, the the completion of work more efficiently and
results show that a good knowledge absorpti-on ef-fectively, the quantity of work can be
will increase the innovation capability, yet not completed according to the standard that has been
significant on performance. This is either set and can be completed in a timely manner.
because the knowledge of absorption can pro- Organi-zations need to encourage and motivate
duce innovation or not. When employees are able the creation of knowledge dissemination process
to improve the absorption, it will impact the between works in organizations, both in formal
performance. Without taking action to ac-quire and informal situations.
knowledge and information, the ability to absorb The results showed that innovation ca-
knowledge does not facilitate inno-vation, pability has a significant effect on performan-
resulting in no significant performance. The ce. Organizations that always find and use
results indicate that the ability to use new new ideas, new methods, new ways and new
information and knowledge has no impact on techni-ques will completion of work on new
performance improvement if the innovation ca- processes and products that impact on employee
pability does not increase. This means that high perfor-mance and organizational performance. In
knowledge absorption capability has the oppor- the context of individual employees, the higher
tunity to increase innovation capability, as long as the capability to generate new ideas, new methods
the knowledge possesses generate ideas, new and techniques will produce quantities of work
ideas to improve performance. If the knowledge according to established standards, more
mastered does not support the completion of work effi-cient and effective, achieving the
with new ideas and ideas, then performan-ce will standard and timely quality. The results
not increase. support the findings of Sulistyo and Siyamtinah
The results of the study contradict the fin- (2016) and Battor and Battor (2010) found that
dings of Nodari et al. (2016), that the capability of the innovation capability affects the
knowledge absorption has a significant effect on improvement of employee performance. The
Performance. Knowledge sharing has a sig-nificant findings do not support the findings of
effect on employee performance. The results Byukusenge (2016), that finding in-novation
also support the findings of Henttonen et al. has no effect on performance impro-vements.
(2016) where knowledge sharing has a sig-nificant
effect on employee performance. The research CONCLUSION AND RECOMMENDATION
findings also support the research of Kwah and
Park (2016) that knowledge sharing activities The success of organizations to compe-te
significantly affect job performance and in the globalization era, especially in sharia
significantly affect innovation. banking industry, innovation is one of the key
Research conducted by Sulistyo and Siy- success factors in improving performance and
amtinah (2016) found that innovation capabi-lity competitive advantage. Through the improve-
affects the improvement of employee per- ment of innovation capability, employees will
formance. The process of sharing knowledge be easier to improve their performance and
involves two dimensions that help disseminate organizational performance. The results show
information and knowledge in the organizati-on, that high innovation capability and performan-
namely knowledge donating and knowledge ce will be able to be implemented properly if
collecting. Knowledge of employees from ex- employees within the organization do knowled-
ternal and internal sources is a knowledge asset ge sharing both from knowledge donating and
for the organization. When knowledge is shared knowledge collecting. Therefore, the ability to
with everyone in the organization, knowledge and share knowledge, interaction among employees
knowledge acquisition will occur and will result in knowledge sharing will be effective when
in increased knowledge that is useful for the knowledge absorption is also better through
122
searching for various information and Arizqi, A. 2017. Performance Enhancement

know-ledge. Good knowledge management Model of Human Resources through
capability will drive innovation capability Knowledge Sharing. Jurnal Dinamika
and performan-ce capability. Manajemen. 8 (1): 134-142.
Almudallal, A. W., Bakri, N., Muktar, S. N & El-Far-ra,
Knowledge is the main key for an orga-
M. M. 2016. Implementing Knowledge
nization to improve innovation capability and Management in the Palestian Public Sector
performance, especially in sharia banking in- Institutions: Empirical Study on the Presi-dency
dustry. Thus, sharia bank organizations need of the Palestinian Government. Inter-national
to improve and facilitate various knowledge Review of Management and Market-ing. 6 (S4):
activities through various training to share 101-107.
with other employees, regularly hold Al. Othman, F. A & Sohaib, O. 2016. Enhancing
meetings with employees either once a week Innovative Capability and Sustainability of
or once a month, invite experts to gain the Saudi Firms. Sustainability. 8 (1229): 1-16.
latest information and create an atmosphere in Andreeva, T & Kianto, A. 2011. Knowledge Pro-
the organization to support employee’s cesses, Knowledge-Intensity and Innovation:
a Moderated Mediation Analysis. Journal of
interaction in sharing the knowledge. Sharia
Knowledge Management. 15 (3): 1016-1034.
banking organizations need to prepare good
Battor, M. M & Battor, M. M. 2010. The Impact
information technology to pro-vide access for
of Customer Relationship Management
employees in obtaining new in-formation and Capa-bility on Innovation and
knowledge, so better knowledge absorption Performance Ad-vantages: Testing a
can ultimately help knowledge sha-ring Mediated Model. Journal of Marketing
activities and enhance innovation capabili-ty. Management. 26 (9-10): 842-857.
This study has limitations in the number Byukusenge, E., Munene, J. C & Orabia, L.
of samples that relatively small, the scope is 2016. Does Knowledge Management
still narrow and less varied, because it only Lead to In-novation? an Empirical Study
focuses on sharia banks, while banks that on SMEs in Rwanda. International
Journal of Management Science and
have status as sharia business units and sharia
Business Administration. 2 (9): 7-19.
credit banks have not been represented in this Cakar, N. D & Ertürk, A. 2010. Comparing Innova-
study. There-fore, the future research agenda tion Capability of Small and Medium-Sized
needs to expand a larger and more varied Enterprises: Examining the Effects of
sample of sharia bank categories in order to Orga-nizational Culture. Journal of Small
generalize the research to be well-dressed. Business Management. 48 (3): 325-359.
Cohen, W. M & Levinthal, D. A. 1990. Absorptive
References Capacity: A New Perspective on Learning
and Innovation. Administrative Science
Akram, M. 2017. The Effect of Knowledge Quar-terly. 35(1): 128-152.
Sharing Behavior on the Relationship of Elianto, W & Wulansari, N. A. 2016. Building
Psychologi-cal Contract, Creativity and Knowledge Sharing Intention with Interna-
Innovative Be-haviors. Journal of Basic tional Trust as a Mediating Variable. Jurnal
and Applied Scientific Research. 7 (9): 10- Manajemen Teknologi. 15 (1): 67-76.
17. Ertürk, A. 2012. Linking Psychological Empower-
Al-Hayaly, M. A. M. S & Alnajjar, F. J. 2016. ment to Innovation Capability: Investigating
Knowl-edge Management Processes and their the Moderating Effect of Supervisory
Im-pact on Organizational Performance, the Trust. International Journal of Business and
Adoption Balanced Scorecard: the Moderat- Social Science. 3 (14): 153-166.
ing Role of Quality Assurance Standards-An Fındıklı, M. A., Yozgat, U & Rofcanin, Y. 2015.
Applied Study on Private Jordanian Univer- Examining Organizational Innovation and
sities. International Journal of Business and Knowledge Management Capacity the Cen-
Management. 11 (6): 70-86.
123
tral Role of Strategic Human Resources work Sharing on Knowledge-Sharing Activi-

Prac-tices. Procedia-Social and ties and Job Performance in Enterprise Social
Behavioral Sciences. 181: 377-387. Media Environments. Computers in Human
Foss, N. J., Husted, K & Michailova, S. 2010. Gov- Behavior. 55: 826-839.
erning Knowledge Sharing in Organizations: Li, Y., Wang, L & Liu, Y. 2011. Organisational
Levels of Analysis, Governance Mechanisms Learning, Product Quality and
and Research Directions. Journal of Manage- Performance: the Moderating Effect of
ment Studies. 47 (3): 455-482. Social Ties in Chi-nese Cross-Border
Garrido-Morino, A., Lockett, N & Garcia-Morales, Outsourcing. Interna-tional Journal of
V. 2015. Exploring the Role of Knowledge Production Research. 49 (1): 159-182.
Management Practices in Fostering Custom- Liao, L-F. 2006. A Learning Organization
er Relationship Management as a Catalyst of Perspec-tive on Knowledge-Sharing
Marketing Innovation. Baltic Journal of Behavior and Firm Innovation. Human
Man-agement. 10 (4): 393-412. System Manage-ment. 25 (4): 227-236.
Henttonen, K., Kianto, A & Ritala, P. 2016. Liao, S. H., Wu, C. F & Chih, C. C. 2007. Knowledge
Knowl-edge Sharing and Individual Work Sharing, Absorptive Capacity and Innovation
Perfor-mance: an Empirical Study of a Capability: An Empirical Study of Taiwan’s
Public Sector Organisation. Journal of Knowledge Intensive Industries. Journal of
Knowledge Manage-ment. 20 (4): 749-768. Information Science. 33 (3): 1-20.
Hooff, B. Van Den & Ridder, J. A. De. 2004. Knowl- Liao, L-F. 2008. Knowledge-Sharing in R & D
edge Sharing in Context : the Influence of De-partments: a Social Power and Social
Organizational Commitment, Communica-tion Ex-change Theory Perspective. The
Climate and CMC Use on Knowledge Sharing. Internation-al Journal of Human Resource
Journal of Knowledge Management. 8 Management. 19 (10): 1881-1895.
(6): 117-130. Liao, S., Fei, W & Chen, C. 2009. Knowledge Shar-
Iddris, F. 2016. Innovation Capability: A ing, Absorptive Capacity and Innovation
Systematic Review and Research Agenda. Capability : an Empirical Study of Taiwan’s
Interdisciplin-ary Journal of Information, Knowledge-Intensive Industries. Journal of
Knowledge and Management. 11: 235-260. Information Science. 33 (3): 160-167.
Jaberi, E. 2016. The Effect of Knowledge Sharing Lin, H-F. 2007. Knowledge Sharing and Firm
on Innovative Behavior among Employee of In-novation Capability: an Empirical
Besat Hospital in City of Hamedan. Interna- Study. International Journal of
tional Academic Journal of Accounting and Fi- Manpower. 28 (3/4): 315-332.
nancial Management. 3 (4): 41-47. Lin, R-J., Chen, R-H & Chiu, K. K-S. 2010.
Janssen, O. 2001. Fairness Perceptions as a Modera- Custom-er Relationship Management and
tor in the Curvilinear Relationships between Innovation Capability: an Empirical
Job Demands and Job Performance and Job Study. Industrial Management & Data
Satisfaction. The Academy of Management Systems. 110 (1): 111-133.
Journal. 44 (5): 1039-1050. Liu, X., Huang, Q., Dou, J & Zhao, X. 2015. The
Kashan, A. J & Mohannak, K. 2015. An Empirical Impact of Informal Social Interaction on
Study of Capability Development within In-novation Capability in the Context of
Product Innovation Project. Journal of Tech- Buyer-Supplier Dyads. Journal of Business
nology Management and Innovation. 10 (1): Research. 78: 314-322.
70-82. Malkawi, M. S & Rumman, A. H. A. 2016. Knowl-
Kumar, N & Rose, R. C. 2012. The Impact of edge Management Capabilities and Its Im-
Knowledge Sharing and Islamic Work Ethic pact on Product Innovation in SMEs. Inter-
on Innovation Capability. Cross Cultural national Business Research. 9 (5): 76-86.
Management: An International Journal. 19 Muskat, B & Deery, M. 2017. Knowledge Transfer and
(2): 142-165. Organizational Memory: An Events Per-
Kwahk, K. Y & Park, D. H. 2016. The Effects of Net- spective. Event Management. 21 (4): 431-447.
124
Nisula, A-M & Kianto, A. 2013. Evaluating and pability and its Effects. Baltic Journal
Developing Innovation Capabilities with of Man-agement. 7 (4): 355-375.
a Structured Method. Interdisciplinary Sulistyo, H & Siyamtinah. 2016. Innovation Capa-
Journal of Information, Knowledge and bility of SMEs through Entrepreneurship,
Management. 8: 59-82. Marketing Capability, Relational Capital
Nodari, F., Oliveira, M & Maçada, A. C. G. 2016. and Empowerment. Asia Pacific
Organizational Performance through the Management Re-view. 21(4): 196-203.
Donation and Collection of Interorganiza- Tsai, M. T & Tsai, T. L. 2010. Innovation
tional Knowledge. VINE Journal of Informa- Capabil-ity and Performance in
tion and Knowledge Management Systems. 46 Taiwanese Science Parks: Exploring the
(1): 85-103. Moderating Effects of Industrial Cluster
O’Cass, A & Weerawardena, J. 2009. Examining Fabric. International Jour-nal of
the role of International Entrepreneurship, Organizational Innovation. 2 (4): 80-104.
Innovation and International Market Per- Usman, M & Mat, N. 2017. The Emergence of
formance in SME Internationalisation. Inno-vation, Knowledge Sharing Behavior,
Euro-pean Journal of Marketing. 43 Islamic Work Ethic and Entrepreneurial
(11/12): 1325-1348. Orienta-tion: A Conceptual Framework for
Pai, Fan-Yun & Chang, Hung-Fan. 2013. The the Pub-lic Sector. International Business
Effect of Knowledge Sharing and Absorption Management. 11 (6): 1234-1239.
on Organizational Innovation Performance-A Wang, M-H & Yang, T-Y. 2016. Investigating the Success
Dynamic Capabilities Perspectives. Interdis- of Knowledge Management: An Em-pirical Study
ciplinary Journal of Information, Knowledge of Small and Medium-Sized En-terprises. Asia
and Management. 8:83-97. Pacific Management Review. 21
Radovanić, N & Matović, D. 2016. (2): 79-91.
Yeşil, S & Dereli, S. F. 2013. An Empirical Investiga-
Organizational Form and Knowledge
Absorption. Industrija. 44 (3): 175-190. tion of the Organisational Justice, Knowledge
Saunila, M & Ukko, J. 2012. A Conceptual Frame- Sharing and Innovation Capability. Procedia-
work for the Measurement of Innovation Ca- Social and Behavioral Sciences. 75: 199-208.
125
Volume 1 No. 7, November 2011 ISSN-2223-4985
International Journal of Information and Communication Technology Research
©2010-11 IJICT Journal. All rights reserved
http://www.esjournals.org
Potential E-Commerce Adoption Strategies For Libyan Organization

*Abdulghader.A.Ahmed, Dalbir S, *Ibrahim M
*
School of Computer Sciences University Kebangsaan Malaysia 43000 UKM Bangi, Selangor, Malaysia.
ABSTRACT
The adoption of electronic commerce (e-commerce) has facilitated sound business relationships between manufacturers
and customers by customizing new functions that are unique to business transaction which provides reliable support that
improves its technological features. These advances earned from e-commerce adoption has paves way for numerous
technical enhancements for merchants with unique business needs and has provided supportive enablement to business
transaction with isolated location. These versed benefits has increased market expectations among variety of generic
consumer as well as the duality of manufacturers and customers. Base on the increasing need for e-commerce integration in
business network, this paper explores the role and relevance of policy intervention on the adoption of e-commerce among
developing countries base on Libyan business environment. In quantifying the managerial potential of e -commerce as a
business enabling technology, Resource Base View theory were used to assess e-commerce capability of e-commerce over
traditional mode of transaction.
Keywords: E-commerce, Policy intervention, E-commerce infrastructure, E-commerce adoption Business transaction network
1. INTRODUCTION government (B2G), consumer-to-consumer (C2C), and

mobile commerce (m-commerce) transaction. B2B e-
Electronic based transaction (E-commerce) refers commerce exists between organizations and among
to business transaction in which the parties involved businesses. However, about 80% of the e-commerce
interact electronically to exchanges information [1]. This transaction are of this type and have been assumed to grow
form of business transaction encompasses buying and faster than the B2C segment [4]. Business-to-consumer
selling over the internet through computer mediated (B2C) involves customers gathering information; purchasing
network [2] and encompasses wide range of products and physical goods (such as books and consumer products) or
services [3]. E-commerce entails the use of electronic information goods (such as software or e-books) through
communications and digital information processing electronically linked network [5]. Consumer-to-consumer
technology to create, transform, and redefine relationships (C2C) transaction exists between individuals, private or
for value creation between organizations and organizations consumers and is characterized by online auctions especially
as well as individuals. Depending on the type of in industries where firms bid what they want from multiple
technology involved and the extent of e-commerce suppliers [6].
integration into business processes, it may constitute part Consumer-to-business (C2B) transactions involve
or the entire business processes. reverse auctions which empowers consumer to influence
E-commerce adoption has proven to be a transaction. Typical instance are common among airlines
potential approach for enhancing the effectiveness of where travellers are given the best ticket offers in response
business practices. These advantages could boost to encourage traveller to fly from a particular post. C2C
operational efficiency of Libyan organization by widening sites such as eBay and Napster revealed that its market is
her marketing network. Owing to the increasing need of e- very large and produces millions of dollars in daily sales.
commerce transaction, this paper discusses the need and Mobile commerce (M-commerce) entails purchase and
relevance for policy intervention in e-commerce adoption sales of goods and services through wireless network such
as well as government role in e-commerce as handheld devices like cellular telephones and personal
implementation. This paper tends to explore the role and digital assistants (PDAs). Japan leads globally in m-
relevance of policy intervention on the adoption of e- commerce transactions [7].
commerce among Libyan business organization. E-
commerce capability of e-commerce over traditional mode 2.1 Policy Intervention on E-commerce
of transaction were assessed base on Resource Base View Processes
theory by quantifying its managerial potential as a
business enabling technology and further recommends
E-commerce as a vital component of business
potential adoption techniques.
strategy can potentially boosts the performance of Libyan
organization. The adoption of electronic strategy into
2. TYPES OF E-COMMERCE business transaction could enhance productivity, reduces
transaction cost, and enables mass customization and
E-commerce transaction exists as business-to- convenient participation of customer [8]. Almost all the
business (B2B), business-to-consumer (B2C), business-to-
321
developing countries currently participate in e-commerce open standards [13]. Internet-based transaction could
either as seller or buyer. To facilitate e-commerce growth facilitate the transaction of goods and services which may
in Libya, there is need to improve e-commerce have required interconnection by allowing for an
infrastructures. To achieve this, policy intervention has instantaneous flow of vital information through efficient
been recommended as a necessity to overrule [4]: and effective integration of the organizational processes
(AA) These benefits are viewed as an ideal strategy to
t High cost of internet access which includes cost boost the economic state and market network of Libya
of connection service, subscription fees, and organization.
charges for websites with sufficient bandwidth.
u Content restriction by national security under
public policy which affect information services *
3Role of Libyan Government to E-
such as media and entertainment sectors
v Limited availability of credit cards. Commerce Adoption
w Insecurity in network. Government policy and incentives plays vital role
x Development of transportation infrastructure to in encouraging the adoption of e-commerce practices [14].
avoid slow and uncertain delivery of goods and
services. For wider adoption of e-commerce, Libyan government
y Inadequate professional information technology should encourage firms and organizational participation by
workforce. adopting e-commerce into all government transaction
networks with policies that guarantees secure online
E-commerce has been noted as a strong tool for transaction and give incentives for e-commerce
sustainable organizational growth. Primary processes that infrastructure. This will enable firms and organizations to
are enhanced in e-commerce transaction network are tap the potential benefits of e-commerce. To ease the
production processes, customer-focused processes, and adoption and creation of favourable environment there is
internal management processes [9]. need for [14]:
xxxiv) Production processes include the *

The creation of favourable policy environment
procurement, for e-commerce and
*
processing of payments ordering and By becoming a leading user of e-commerce in
replenishment of stocks; production control their respective operations as well as encouraging
processes; electronic links with suppliers among its mass usage.
others;
xxxv) Customer-focused processes include Among the favorable policy environment for e-commerce
promotion, sales over the Internet, processing of adoption, Libyan governments should heed to include:
purchase order for customers, payments and
customer support.
- Promoting access to inexpensive and easy internet
xxxvi) Internal management processes access to information networks
encompasses employee services, training and - Legal recognition of e-commerce transactions
recruiting, video-conferencing and internal
information sharing. Electronic transaction - Protection of consumer from internet fraud
enhances information flow among production and
sales unit to improve productivity [10]. - Protecting of consumer’s right and privacy
- Legal protection against unauthorized access to

2.2 Relevance of the Internet to E-Commerce computer systems
Transaction in Libya - Protection of intellectual property
The internet is a vast network that allows people - Policy and legal framework for e-commerce in
from different geographical location to interact Libya should address the above issues. It is also
important for Libyan government to adopt laws,
inexpensively and reliably. It encompasses global policies that promote trust and confidence among
collection of networks connected to share information via e-commerce participants that is compatible with
common sets of protocol [11]. The emergence of the international norms on e-commerce [15].
internet has facilitated e-commerce transactions in Libya at
reduced cost. Before the use of the internet for commercial Ray et al., [16] proposed numerous ways in
transactions in Libya, organizations used private networks which government in the developing countries could
like electronic data interchange to transact businesses encourage the adoption of electronic-based transaction.
which are expensive to install and maintain [12]. However, Similarly these measures can be applied among Libyan
the integration of the internet into the transaction network organizations. Libyan government can also motivate
of Libya could rapidly spread e-commerce adoption organizations and individuals to adopt e-commerce
primarily because of its low costs and practices through:
322
E-procurement: Government agencies should trade chain management activities are used [18]. Effective
electronically with suppliers through e-procurement organizational criteria reported from previous literature are
information systems. as shown in table 1 and 2 respectively.
Successful adoption of e-commerce requires
Customs clearance: through the computerization of suitable environment to yield its advantages. Basis on
customs processes and operations covering electronic previous literatures various criteria of e-commerce
submission, processing, payment, and automated systems adoption have been identified at different organizational
for data entry to integrate customs tables, codes and pre- levels as summarized in the tables 1 and 2
assessment.
3.1 Recommendation on E-commerce
Tax administration: through electronic transmission and Adoption
processing of tax return information, permits, issuances of
tax clearances via online, licenses and electronic The importance of e-commerce is undisputed
registration processes of businesses and taxpayers. among developing nations [24]. Different reasons are
responsible for the asymmetry like positive attitude
However, the private sectors could be towards e-commerce and better qualifications for staff. E-
discouraged from participating in e-commerce transaction commerce activities concentrate on B2B although its
if Libyan government fails to engage in e-procurement, adoption could be determined on its efficiency. The
secure their records online or electronically remit customs adoption of e-commerce requires that the emerging market
fees, [17]. establishes its transaction strategies. Base on the need for
effective e-commerce implementation, this study
- CRITERIA FOR E-COMMERCE recommends potential steps necessary for e-commerce
ADOPTION IN THE DEVELOPING adoption in Libya.
COUNTRIES Twelve Recommended steps for effective e-
commerce implementation aims promoting e-commerce
Based on Rao, et al., [18] e-commerce adoption among Libya organizations. The
recommendation includes:
development and implementation encompasses presence,
portals, transaction integration and enterprise integration.
Presence is the first step in which most e-commerce E-commerce features: website as tools for e-commerce
organizations displays the product, brochure and offer on a tool tailors global markets such as Europe, America and
Website. The portals stage comprises of order placing Asia and should comprise of features like corporate
through the use of profiles and cookies by customers and information, products and services, information,
suppliers transactions integration involves financial frequently asked questions (FAQs), useful information on
transactions between partners while enterprises integration non-product like weather, online enquiries, security
refers to full integration processes of businesses that statements, online reservations, reservation tracking and
makes old business route indistinguishable to online online payment strategy
business.
The use of model for the adoption of online Customer Relationship Management (CRM): website
transaction has proven to be subject of distinct and unique uses effective electronic CRM helps to encourage
phenomena [10, 17, 19-22]. However, the adoption of e- customers and to create impression that could captures
commerce for organization processes can be motivated their preferences, requirements however, their is need to
through the use of appropriate model. include contact details, e-mail, an online feedback form,
Effective organizational criteria are important an electronic newsletter, promotions, special offers and
variables that can influence the adoption changes other loyalty schemes, a call-back service, and
associated with e-commerce among organizations in the personalization through customized content.
developing countries [23]. Business organization varies in Website promotional campaigns: the Internet could be
propensity to deploy e-commerce and is influenced by top used to promote products and services at a reduce cost
management, information systems professional’s owing to the unprecedented level of connectivity and the
characteristics as well as the business structure of the ability to communicate efficiently and effectively and
organization [9]. The effectiveness these variables can be directly with customers.
used to evaluate e-commerce adoption in the developing
countries. To successfully utilize e-commerce, buyers and Knowledge transfer facilities: website provides
sellers must have sufficient information system interactive environment for the exchange of knowledge
infrastructures for maximum participation over every that cane used by organizations to provide products and
transaction. The readiness of a buyer could be traced from services to their customers through electronic guestbook
the extent at which information systems are used in or user review facility, instant messaging facilities,
purchasing, the extent at which internet purchasing electronic newsletters, bulletin boards and online tracking
activities are conducted and the extent at which supply of orders.
323
Increasing website traffic: Organizations should include view theory which emphasis on managerial skill and e-
websites facilities to attract more customers through commerce transaction network.
techniques such as doorway pages, customer accounts,
effective Meta and title tags, and registration of websites - RESOURCE-BASED VIEW (RBV)
on major search engines. The website used for transaction THEORY ATTRIBUTES TO E-
should be regularly updated.
COMMERCE CAPABILITIES
Web content accessibility: Website content should be
accessible and usable by all people, including those with Resource-based View (RBV) model developed
disabilities by following the Web Content Accessibility by Mata et al., [26] revealed that e-commerce resources,
Guidelines (World Wide Web Consortium) [25] when its managerial skills and business network are sources for
developing and designing their websites. competitive advantage because of their valuable, imperfect
immobility and heterogeneous distribution across business
Overcoming security and legal barriers: Sensitive data organizations. RBV added that durable e-commerce
capabilities are often involves long process of
of customers such as credit card details and personal
development between people and technology with respect
information should be protected during e-commerce
to the history of the organization. The development
transaction by using firewall, entrapment server, user
process of an organization involves thousands of
accounts, proxy server, virus email updates, encryption,
repertoires, routines and managerial decisions among
anti-spyware, audit logs, digital certificates, anti-virus
employees, suppliers, business partners and competitors.
software, Internet usage policy, secure electronic
The skills to coordination and integrate existing resources
transaction, secure socket layer, digital signatures,
into capabilities often requires organization-specific
prepayment systems, user accounts, disaster recovery
knowledge on e-commerce. These elements cannot easily
plans, trust seals, privacy and security statements, and a
be codified since they are imitable as a result of social
security payment link.
complexity and casual ambiguity. Base on this fact, e-
commerce capability is a strategic resource with durable
Overcoming the technological barriers: Organizations
advantages for business organizations.
should use appropriate technology to setup e-commerce
RBV model was postulated to test the capabilities
system taking into consideration the need for modern
payment infrastructure and a high speed, reliable Internet of e-commerce as well as its sustainable competitiveness.
connection. RBV is a promising contemporary theory which combines
strategic and organization insight and competitive
advantage on firm existence [27]. RBV applications
Overcoming organizational barriers: Organizational
involve studies base on information technology system
culture and should accommodate e-commerce-induced
resources with capabilities to yields sustainable and
changes. Owing to this, organizations carry out thorough
competitive advantage [28]. The following section
investigations before the implementation of e-commerce-
examines the three attributes of e-commerce capabilities
induced changes.
base on RBV model and theoretical conclusion for future
study such as e-commerce technology resources,
Overcoming behavioural barriers: The adoption and use
managerial e-commerce skills and business networks.
e-commerce should consider mutual relationship the
employer and employees so as not to resist change when it
comes. However, to reduce the risk of key staff leaving the 4.1 E-Commerce Technology Resources
organization after being trained, there is need to
investigate employee’s commitment and involve top Most studies focused on the economic values of
management in decision before training. e-commerce technology resources that improve
organizations operation and effectiveness as well as
Overcoming the financial barriers: Adequate finance is differentiating its respective products and services [29-31].
needed to run e-commerce system. In a situation where However, electronic devices are rare and immobile
fund is insufficient, the organization can borrow from resources because hardwires and software can be
financial institution using investment risk analysis, return purchased from the market. In the other hand, critics on
on investment, cost-benefit analysis to manage the RBV stressed that electronic mediums are not strategic
financial commitment starting with basic inexpensive e- resources because, they lack of inimitability [32]. E-
commerce facilities. commerce technology resources are made up of sets of
infrastructure, information strategy (IS), network
Training to prepare for e-commerce adoption: there is applications, and electronic architecture plan [33]. In
need to train key personnel on e-commerce strategy so as addition to its complexity, there development processes
to overcome most significant barriers such as the security evolves over long period through complex interaction
and technological associated barriers. A typical measure to between people, business processes, and technology which
overcome these is through the adoption of resource based are homogeneously valuable across firm. However,
managerial skills of e-commerce such as the use of e-
324
commerce infrastructures (such as the internet) facilitate however; the development of managerial e-commerce
its adoption among organizations. skills requires long time to accumulation learning
experiences as well as to overcome trial and error [37].
4.2 Managerial E-Commerce Skills
4.3 Business Networks
Managerial e-commerce skills encompasses (1)
the information technology (IT) manager’s ability to Business network sustains competitive advantage
understand the needs of a firm, stakeholders, employees, through e-commerce. Intensified competition makes firms
suppliers and customers, (2) the capability to adopt e- to depend on the network of supplier and customer to
commerce strategies to the firm’s business structure, create value. Accumulation of more customers in online
value-chain and its transaction networks (3) the capability business generates integrated knowledge and relationships
to work collectively by managers towards a common goal that considerably extends beyond their personal resources
with the aim to reduce cost, improve speed, productivity and control [38]. It has been reported that customer value
and to increase revenue. Researchers pointed out that the are improved through customers and suppliers networks
skill of managers provides sources for competitive [39 & 40]. A network-based community focuses on the
advantage [33 & 30]. This is not an exemption under RBV need for trust and greater collaboration to facilitate
theory where human skills have proven to play a vital role information sharing across the value chains [38].The
for competitive performance in various sectors like e- adoption of e-commerce technology into firms in the
commerce, manufacturing and retailing, [30, 34 & 35]. developing countries can break-off barriers associated
It has been reported that managerial skills and with distance and integrate suppliers and customers
level of firm performance has significant relationship [36] network in a domain.
Table 1: Effective organizational criteria for e-commerce adoption in the developing countries generated from
previous literatures
Criteria’s Sources
Leadership and management issues [19, 40-44]

Human resource management issues [42 & 45].
Size of the organization or organization [23, 41 & 46]
Organization culture [19 & 42]
Organizational structure
Products [20]
Financial structure [42]
Organizing [47]
Adapted from Rao, et at., [18]
Table 2: Criteria for e-commerce adoption among business organizations in the developing countries
Criteria’s Sources
High competition [10, 19-23 &41].
Customers pressure [17, 19-21].

Suppliers pressure [19-21]
Customers readiness [10, 48 & 49]
Demonstration of electronic purchasing [50]
Readiness of business organizations [10]
Lack of vital expansion [19, 41& 49]
Adapted from: Rao, et at., [18]
325
5. CONCLUSION Ltd. on sequences, IEEE Transactions on

Engineering Management 50: pp.8=26.
E-commerce is a smarter approach to conduct
business and the best approach is for organizations to [6] Shih, E., Kraemer, KL. & Dedrick, J. I.T.
manage their activity as an integral part of their approach 2008. diffusion in developing countries.
to developing and sustaining customer relationship. E- Commun ACM; 51(2):43– 8.
commerce technology is available and affordable, and is
an international uptake that presents ways for companies [7] Japan External Trade Organization. February
to demonstrate their capability through the use of IT 2002. Electronic Commerce in APEC
systems for business transaction. E-commerce-base Economies: Focusing on Electric/Electronic
transaction uses electronic and digital information Parts Procurements.
processing technology to create, transform, and redefine
relationships for value creation between organizations and
[8] Lederer, A. L., Mirchandani, D.A., and Sims, K.
organizations as well as individuals. The internet plays a
2001. "The Search for Strategic Advantage
central role in mediating transaction processes in
from the World Wide Web." International
implementation of e-commerce among Libyan
Journal of Electronic Commerce 5 ( 4):
organizations. Reports show that e-commerce adoption is
117-133.
enhanced through policy intervention on its processes and
provision of e-commerce infrastructures as well as the
provision of incentives. The RBV model revealed that e- 352 Tarafdara, M. V., & Sanjiv, D. 2006. Challenges
commerce technology provides competitive advantages in the adoption of E-Commerce technologies in
that are vital in enhancing imperfect market across firms. India: The role of organizational factors.
RBV further revealed that e-commerce development International Journal of Information
process involves numerous organisational routines, Management, (26), 428–441.
repertoires as well as managerial decision among business
partners, employees, suppliers, and competitors. [10] Zhu, K., Kracmer, K., & XU, S. 2002. A cross
country study of electronic business adoption
ACKNOWLEDGEMENT using the technology–organization– environment
framework. In Proceedings of the 32th
The research was funded by The General People's international conference on information systems.
Committee for Higher Education, Bane waleed University,
Libya.
[11] Tonderai, M., Janet, E., & Ray, D. 2010.
REFERENCES Recommendations for e commerce systems in
the tourism industry of sub-Saharan Africa.
[1] MK, Euro Info Correspondence Centre Telematics and Informatics 26 :12– 19.
(Belgrade, Serbia), ―E commerce-Factor of
Economic 2008. [12] Doing business and investment in Libya, 2009.
World business, investment and government
- Andrea Goldstein, & David O’Connor, 2004. E- library, published by International business
commerce for Development: Prospects and Policy publication Inc.
Issues, (OECD Development Centre,
September 2004); available from [13] Shaaban, E., Alireza, H., 2009. A framework
http//www.oecd.org/dev/publication/tp1a.hm; for evaluating electronic commerce adoption in
accessed 15January 2011. Iranian organizations International Journal of
Information Management 29: 27–36.
[3] Alsmadi, M., Khan, Z., & McTavish, A.
2011. Evaluating competitive advantage [14] Jalalia, A.A., Okhovvatb, R. M., Okhovvat, M.
priorities of SMEs in Jordan. International 2011. A new applicable model of Iran rural e-
Journal of Networking and Virtual Organisations commerce development. Procedia Computer
9 (1): 25 – 43. Science 3 (2011) 1157 1163.
[4] Johnson, M. 2010. "Barriers to innovation *

Yong B. M.,Sung-Ho J.& Jong S. L.
adoption: a study of e- markets." Industrial 2011.Ontology-BasedResource
Management Data Systems 110(2):157-174. Management for Cloud Computing.
ACIIDS 2: 343-352.
[5] Fox, W. & Bayat, M.S. 2007. A Guide to
Managing Research. Juta & Organization [16] Ray, G., Muhanna, W. A., & Barney, J. B.
2005. "Information Technology and the
326
Performance of the Customer Service 353 Kaynak, E., & Tatoglu, E. 2005. An analysis of
Process: A Resource-Based the factors affecting the adoption of electronic
Analysis,"MIS Quarterly, 29: 4. commerce by SMEs: Evidence from an emerging
market. International Marketing Review, 22(6):
[17] Daniel, E. M.,& Grimshaw, D. J. 2002. An 623 - 640.
exploratory comparison of electronic commerce
adoption in large and small enterprises. Journal 354 Barney, J. B. 2001. "Is the resource based "view"
of Information Technology, 17, 133–147. a useful perspective for strategic management
research? Yes." Academic of Management
[18] Rao, S. S., Truong, D., Senecal, S., & Le, Review Vol.26, (1): pp.41-56.
T. T. 2007. How buyers’ expected benefits,
perceived risks, and e business readiness 355 Truijens, O. 2003. "A Critical Review of the
influence their e marketplace usage. Industrial Resource-based View of the Firm," University of
Marketing Management, 36(8): 1035–1045. Amsterdam, Netherlands. Sprouts: Working
Papers on Information Systems, 3(6).
[19] Ling, C. Y. 2001. Model of factors influences on http://sprouts.aisnet.org/3-6
electronic commerce adoption diffusion in
small & medium sized enterprise. Curtin [30] Hulland, J., Wade, M., & Antia, K., 2007. ―The
University of Technology. Working Paper. Impact of Capabilities and Prior Investments on
Online Channel Commitment and
[20] Wang, J. C., & Tasi, K. H. 2002. Factor in Performance‖, the Journal of Management
Taiwanese firm’s decisions to adopt electronic Information Systems, Vol. 23 No. 4, pp. 109-142.
commerce: An empirical study. The Word
Economy, 25(8): 1145–1167. 359 Porter, M. E., & Kramer, M.R. 2002. "The
Competitive Advantage of Corporate
[21] Rashid, M. A., & Qirim, N. A. 2001. E- Philanthropy." Harvard Business Review
commerce technology adoption framework by Vol.80( No. 12).
New Zealand SME’s. In Proceedings of the 35th
Hawaii international conference on system 360 Carr, N. G. 2004. Does IT Matter? Information
science. Technology and the Corrosion of Competitive
Advantage. Boston, Massachusetts, Havard
353 Julta, D., Bodorick, P., & Dhaliwad, J. 2002. Business School Press. 56
Supporting the e-business readiness of small and
medium-sizedenterprises: Approaches and [33] Halawi, L. A., McCarthy, R. V., & Aronson, J.
metrics. Electronic Networking Applications and E. 2006. ―Knowledge Management and the
Policy, 12(2): 139–164. Competitive Strategy of the Firm‖. The
[23] Trouong, D. & Rao S.S. 2002. Development Organizational Learning Journal, Volume 13,
a contingency model for adoption of electronic Issue 4, p.384-397.
commerce. Decision Science Institute,
annual meeting proceedings. [34] Powell, T. C. 2003. "Research Notes and
Commentaries." Strategic Management Journal
[24] DECKER, R., HERMELBRACHT, A., & 24: 285-291.
KHOROUNJAIA, T. 2008. Ecommerce
als Instrument zur Bearbeitung des [37] Aljifi, H. A., Pons,A. & Collis, D. 2003. "Global
russischen Marktes, Status Quo und e-Commerce: A Framework for Understanding
Entwicklungsperspektiven. Journal fur and Overcoming the Trust Barrier." Information
Betriebswirtschaft, 53, 5-6, 190- 207. Management and Computer Security 11(3): 130-
138.
[25] W32.Bugbear 2005, Symantec Security
Response,http://securityresponse.smantec.com/ a [38] Brown, J. S. & Duguid, P. 2002. The Social Life
vcenter/venc/data/pf/32.bugbear@m.html of Information. Boston, Massachusetts, Harvard
Business School Press.
[26] Mata, F. J., Fuerst, W.L., & Barney, J.B.
1995. "Information Technology and Sustained [39] Powell, T. H., Thomas, H., & McGee, J.
Competitive Advantage: A Resource-Based 2004.Dynamic knowledge creation. Paper
Analysis"." MIS Quarterly 19(4): 487- 505. presented at the Dynamics of Strategy, University
of Surrey.
327
[38] Srivastava, R.K., Fahey, L. & Christensen, [49] Internet Governance for Development 2006.
H.K. 2001. The Resource-based Viewand South Centre Analytical Note.
Marketing: The Role of Market-based www.southcentre.org
Assets in Gaining Competitive
Advantage. Journal of Management, 27, 777-802. [50] Sung, T. K. 2006. E-commerce critical success
factors: East v. West. Technological forecasting &
[40]Frels, J. K., Shervani, T.A., & Srivastava, R.K. social change, 73(9): 1161–1177.
2001. The Integrated Networks Model:
Explaining Resource Allocations in Networked [51] Turban, E. R., Jr., Kelly, R., & Porter, R. E. 2007.
Business Markets, University of Texas at Austin. Information technology (p. 615). New Jersey:
John Wiley & Sons.
[40] Mehrtens, J., Cragg, P. B., & Mills, M. A.
2001. A model of Internet adoption by SmEs. [50] Kare-Silver, M. D. 2002. Electronic shock 2000.
Information and Management, 39(3): 165–176. New York, McMillan Press.
[41] Jun, M. C. S. 2003. Key obstacles to EDI AUTHORS INFORMATION

success: From the US small
manufacturing organizations’
perspective. Industrial Management & Abdulghader.A.Ahmed.Moftah:
Data Systems, 103(3):192-203. He completed his undergraduate degree in
computer science at 7th October
[42] Grandon, E. P. & Michael, J. University Bani Walid, Libya in 2001. He
2004.Electroniccommerce adoption anempirical is a master candidate in computer science
study of small and medium US business. at faculty of computer science & Information Technology
Information and Management, 42, 197–216. University Kebangsaan Malaysia (UKM).
[46]Ruikar, K., Anumba, C. J., & Carrillo,P. M. 2006.

VERDICT—An E-readiness assessment Dr. Dalbir Singh received the degree in
application for construction organizations. Computer Science from the Universiti Sains
Automation in Construction, 15: 98–110. Malaysia, in 2002. He received the Ph.D.
degree in Computer Science from the
Universiti Malaya in 2009. Currently, he is a
[44] Lee, S., & Kim, K.-J. 2007. Factors affecting the
senior lecturer at National University of
implementation success of Internet based
information systems. Computers in Human Malaysia. His research interest includes Human Factor in
Information System.
Behavior, 23: 1853–1880.
[45] Standing, C., & Chad, L. 2007. Organizational

evaluation of the benefits, constraints, and Ibrahim Mohamed received the degree in
satisfaction of business to business electronic Accounting & Finance from the Liverpool
commerce. International Journal of electronic JM University, in 1996.He received the
commerce, 11(3): 107–134. Masters degree in Information Technology
from the National University of Malaysia
[46] Mukti, N. A. 2000. Barriers to putting business in1999. Currently he is a lecturer at National University of
on Internet in Malaysia. The Electronic Journal of Malaysia His research interests include business data
Information Systems in Developing Country, modelling.
2(6): 1–6.
328
RAM, Rev. Adm. Mackenzie, (Mackenzie Management Review), 18(6) • São Paulo, SP • Nov./Dec. 2017 • ISSN 1678-6971 (electronic version) • doi
10.1590/1678-69712017/administracao.v18n6p71-95 Submission: Apr. 22, 2017. Acceptance: Aug. 12, 2017. Evaluation system: double-blind review.
Universidade Presbiteriana Mackenzie (UPM). Silvio Popadiuk (Editor-in-chief), Nanno Mulder (Associate editor), p. 71-95.
INNOVATION IN THE BUSINESS MODEL

FROM THE PERSPECTIVE OF DYNAMIC
CAPABILITIES: BEMATECH’S CASE
ELIC VODOVOZ
Universidade Federal do Paraná (UFPR).
MÁRCIA RAMOS MAY

To cite this paper: Vodovoz, E. & May, M. R. (2017). Innovation in the Business Model from the
perspective of Dynamic Capabilities: Bematech’s case. Revista de Administração Mackenzie, 18(6), 71-95.
doi 10.1590/1678-69712017/administracao.v18n6p71-95
Este artigo pode ser copiado, distribuído, exibido, transmitido ou adaptado desde que citados, de forma clara e
explícita, o nome da revista, a edição, o ano e as páginas nas quais o artigo foi publicado originalmente, mas sem
sugerir que a RAM endosse a reutilização do artigo. Esse termo de licenciamento deve ser explicitado para os casos
de reutilização ou distribuição para terceiros. Não é permitido o uso para fins comerciais.
Elic Vodovoz e Márcia Ramos May
ABSTRACT
Purpose: The literature on business models deals with creation and capture of
the value obtained through innovation as elements necessary for the continuity
of the companies. The concepts inherent in business models demonstrate the
operation logics s for the creation of value. To verify how the value creation
process occurs, a case study has been carried out based on the change in the
business model of Bematech S.A, a Brazilian company from Information
Technology (IT) service area.
Originality/Value: The theory about business models and dynamic capabilities
indicates some connections between these concepts, but this connection lacks
practical observation, so the purpose of this article is to understand how
innovation occurred in the business model and search, which routines were
essential for the value creation in the company Bematech. The phenomenon
was observed from the perspective of dynamic capacities.
Design/methodology/approach: Based on the reviewed literature, it was

decided to conduct in-depth interviews with employees from the strategic
planning and IT areas. A graphical representation of the business model was
used to capture its structure, the business model canvas.
Findings: The results of this study show how dynamic capability and business
model connections occur and demonstrate which high-level routines enabled
innovations in the Bematech’s business model for value creation and value
capture. The study also shows the boundary between dynamic capabilities and
the business model.
KEYWORDS
Business model. Dynamic capabilities. Value creation. Value capture.
Routines.
72
Revista de Administração Mackenzie – RAM (Mackenzie Management Review), 18(6), 71-95 • SÃO PAULO, SP •
NOV./DEC. 2017 • ISSN 1678-6971 (electronic version) • doi 10.1590/1678-69712017/administracao.v18n6p71-95
Innovation in the business model from the perspective of dynamic capabilities: Bematech’s case
1. INTRODUCTION
In the scenario today, managers must be able to organize multiple distribution
channels, complex supply chains, rare technological resources, and yet remain
flexible enough to shape the changing market. In this situation of increasing
complexity, the business model theme has acquired importance as a way of
explaining the functioning of an organization considering the components of its
strategy, an accessible option to understand or manage a business by its main
processes and routines (Cavalcante, Kesting, & Ulhøi, 2011).
This understanding is possible through logical representation of how value is

delivered to the companies’ customers. This broad view of a business is called a
business model (OSTERWALDER; PIGNEUR, 2010). Its construction allows
identifying how a particular enterprise is implemented, how one can capture value
from it and the structure necessary for this.
By observing the influence of the business model in the strategic base of the
company flexibility in technology changing environments and market (Bock,
Opsahl, George, & Gann, 2012), it is desirable to analyze how this relationship has
been built and the changes designed over time. In this case, the concept of dynamic
capacities is adapted to the role of analysis lens, in order to understand the
behavior of the actors inherent in the process of changing business models
(TEECE, 2010).
In this context, Teece (2007) argues that through the development of dynamic
capabilities, it is possible to identify and reconfigure competencies that the
company needs to act in an environment of constant change.
Despite of the author’s assertion, some practical questions remain about the
connections between the concepts of dynamic capabilities and business models,
that is, the positioning of dynamic capacities in relation to business models in a
real case. In the theoretical scope, the answer to these questions can direct the
study of these concepts and help understanding the boundaries between them. In
the organizational sphere, this clarity can help in the process of innovation on
business models, that is, in the constitution of methodologies or technical tools
oriented to innovation in business models.
Therefore, the purpose of this article is to understand how innovation has

occurred in the business model and what routines have been essential in the search
for the creation of the Bematech’s values. The phenomenon is observed from the
perspective of dynamic capacities.
73
2. BUSINESS MODEL
The concept of business model is not unanimous among authors, as Zott, Amit,
and Massa (2011) affirms. However, points of convergence are observed between
the various definitions found in articles and books. The theorists understand the
business model as a concept directly related to creation, delivery and capture of
value (OSTERWALDER; PIGNEUR, 2010).
As a way to facilitating the understanding about organizations, Petrovic et al.
Kittl. (2001) perceive the business model as a description of a complex business
that enables studying the structure, the relationship between the structural elements
and how these elements respond in the real world. In this context, Stahler (2002)
states that a model is always the simplification of a complex reality and helps to
understand the fundamentals of a business or to plan how a business should behave
in the future.
In the same context, Magretta (2002) treats the business model as a story that
explains how a company works, that is, describes how the “parts” of a business are
interconnected.
The difficulty to represent all business model elements made Osterwalder
(2004) try to unify the concepts in building blocks of the business model, to
represent the value creation logics in a tool known as Business Model Canvas. To
do so, they used the existing literature as a basis to design the analysis tool that
contemplates the set of elements of the business model and their relations to
express the logic about how a company makes money. Therefore, the business
model canvas is seen as a visual representation of the relationships between the
business model elements to facilitate understanding of the value delivered to the
customer segments, the organizational architecture of the company and its network
of partners.
In addition to facilitating the understanding of the business, the visualization
of the business model with the help of representation by the tool, brought the
possibility to understand the business model in its structural aspect, that is, the
structural design of the company (Baden-Fuller & Morgan, 2010) for the
exploration of new business opportunities, indicating the possibilities of generating
value through a systemic analysis aimed at the innovation of organizations (ZOOT;
MASSA, 2011; OSTERWALDER; PIGNEUR, 2010; Cavalcante, S., Kesting, P., &
Ulhøi, J 2011).
2.1. Canvas Business Model Template
In the analysis performed in this work, the nine blocks idealized by

Osterwalder and Pigneur (2010) have been used as a graphical representation
74
of the business model. The business model (Canvas) is the analysis tool selected
for this study, since it is the one that presents the greatest theoretical
comprehension among the representations of the analyzed business models
(Vodovoz, 2015). Thus, the components of the Canvas business model are detailed:
z Customer segment: Individuals or entities that purchase the goods or services.
aa Value proposition: The value proposition is at the heart of the business model.
This positioning reflects “what” the company offers to solve the problem or
meet the customer’s need.
bb Channels: Channels describe how the company delivers products to customers,
how the value proposition reaches the customer.
cc Customer relationship: the customer relationship block describes what extent
of loyalty the company expects to maintain with their customers.
dd Revenue sources: are the ways revenues enter. This block describes how the
money paid by customers reaches the company.
ee Key Features: A pack of key features explains what are the main features that
keep the operation of the business model.
ff Key activities: are activities in which the company must be proficient to keep
the business model.
gg Key partnerships: are ways to maximize reach or even enable a business
model.
hh Cost Structure: The cost structure describes the value that is necessary for the
operation of the business model; in an objective way, costs are concentrated on
key activities, key resources and partnerships.
3. CONCEPT OF VALUE
In the field of strategic management, value has been used to explain the search
for competitive advantages of organizations in two chains of thought: Porter (1985)
and Barney (1991). However, there is still no common concept about value
(Barney & Arikan, 2001; Bowman & Ambrosini, 2000).
For Porter (1985), in his perspective known as the Strategic Positioning
Analysis (SPA), the essence of competitive advantage is the highest value
customers are willing to pay in relation to the company manufacturing cost.
According to Barney (1991), and based on the concept of resource-based
Vision (RBV), competitive advantage is achieved by a company when its value
creation strategy cannot be easily copied by competitors.
The work on dynamic capabilities commonly uses terms, such as value
creation and value capture. It is known that the definition of value in the view
75
of the dynamic capabilities is influenced by the Barney's current (1991), in which

the value is related to valuable resources, rare, difficult to imitate or replace.
However, it is perceived that the dynamic capabilities also relate the creation of
value to high-level routines based on the entrepreneurial activity of identifying
opportunities and resource mobilization, implementation of business models
(Augier & Teece, 2008), processes, leadership ability applied to these resources
(Teece, 2014).
However, RBV lacks studies about the creation and capture of value for
understanding the phenomenon in strategic management (Kraaijenbrink, Spender,
& Groen, 2010). In the search to fill these gaps, one option is to invoke other
theoretical lenses such as the dynamic capabilities.
3.1. Value creation and value capture
The ability to create value does not guarantee the persistent performance of the
company’s activity, since there are external to the company factors, such as
competition, for example. For a long-term activity, the organization must be
effective in creating and capturing value. This is because the created value may be
different from the captured value, and often so is the company can create value but
may fail to capture value.
The literature shows that several authors indicate elements and processes they
consider important in the creation of value in the organizational scope. For
Boulton, Libert and Samek (1997, 2000), value creation comes not from a single
enterprise resource, but from the integration of all organizational elements.
When facing the same perspective, Lindgreen, Hingley, Grant, and Morgan
(2012) point out that the relationships between organizational resources, whether
tangible or intangible, are the central elements of the value creation process.
According to Leih, Linden, and Teece (2014), the capture of value depends on
an architecture that combines strategies of entry, integration, cooperation and
diversification with the organizational design in harmony with these strategies, that
is, the creation of value from processes or routines of high-level capabilities that
enable organizations to review and develop capabilities for long-term value
creation and capture.
In this context, the creation of value handled in this work moves away from
mathematical exchanges insofar as the value created is influenced by the processes
of transformation, capacity to identify the opportunities and the clients' needs and
the combination and transformation of resources (Anderson, Nars 1998;
ZEITHAMAL, 1998, Penrose, 1995). These key
76
elements for creating and capturing value, as well as the role of the manager in
creating the essential routines in value creation and capture processes, are
addressed in the next section under the lens of dynamic capabilities.
3.2. Dynamic capabilities
The vision of the dynamic capacities takes care of the adaptive aptitude of the
firm in a dynamic environment (Teece & Pisano, 1994; TEECE; PISANO;
SHUEN, 1997). It is the organizational capacity to detect, integrate, learn and
reconfigure its internal or external resource base, whether knowledge, skills or
strategies, to adapt and respond to the demands of a constantly changing
environment (TEECE; 2010, 2012).
Teece and Pisano (1994) developed the theme based on RBV. According to
Barney (1991) and Grant (1996), organizational resources considered difficult to
reproduce by other competitors, when used for value generation, can generate
competitive advantages. The dynamic capabilities perspective has added the
external component and importance of strategic capacity management, proposing
greater flexibility to develop new capabilities for RBV, as well as renewing
existing ones with the purpose to generate competitive advantages. Even though,
the resources employed are scarce or difficult to replicate (TEECE; PISANO;
SHUEN, 1997; Eisenhardt & Martin, 2000; Augier & Teece, 2008).
In the same context, in a more recent approach, dynamic capabilities are

determinant for the speed and degree to which companies align and realign their
resources to meet the needs of the environment, as well as to seize the
opportunities generating sustainability and advantages over to competitors
(TEECE, 2007, 2012).
In order for this to happen, the companies' actions should be focused on: a)
processes of integration and alignment of resources (TEECE, 2012),
understanding internal processes in their congruence and com plementarities, that
is, integration between processes and organization with the help of logic; b)
learning – processes that are performed repeatedly, improve the efficiency of the
organization and enable identification of dysfunctions c) adaptation – running
well-organized learning processes, such as benchmarking, in order to gain
competitive advantage in dynamic environments and develop the ability to learn
from the organization to be a routine practice; thus, organizations with dynamic
capabilities are always observing the changes in the environment by evaluation of
the markets and competitors and consequently, they are able to adapt to the
changes more easily (Teece & Pisano, 1994).
77
The concepts advocated by dynamic capabilities approximate those found in

Teece’s (2010) business and value-creation models when they claim that the
generated competitive advantages depend on the increase in value resulting from
high-level routines that enable firms to reconfigure their resources, identify and
exploit opportunities, synchronize processes with business models to shape the
environment in their favor (LEIH; LINDEN; TEECE, 2014).
In summary, Leih, Linden, and Teece (2014) argue that understanding dynamic
capabilities is fundamental, since they represent the basis for creating and
capturing value through high-level routines.
The literature lists two types of capacities: Ordinary capacities and dynamic
capacities (Winter, 2003). Ordinary or operational capacities encompass
operational, administrative and governance activities that ensure the organization’s
ability to exist (production and sale of products and services) (Lehi and Linden,
Teeche, 2014). Dynamic capacities represent high-level routines; they differ from
ordinary capacities, but are interrelated in a hierarchical relationship because
dynamic capacities are those capable of reconfiguring and creating new
combinations of operational capacities to maintain the relevance of these ordinary
capacities in turbulent environments (PAVLOU; SAWY, 2011). An improvement in
ordinary capacities can bring greater efficiency of activities; however, dynamic
capacities are the only ones capable of creating value and competitive advantages
when combined with a solid strategy as well as a good execution and allow the
company to develop the ideal products in relation and are critical for entry into new
markets (TEECE, 2012).
xxxvii) From this perspective, the dynamic capabilities view points to the
existence of three groups of management processes that can be used to identify
opportunities for value capture and continuous renewal in accelerated change
environments. These are the types of high-level routines (TEECE, 2007,
2012):
xxxviii)Sensing – routines to identify and evaluate opportunities.
xxxix) Seizing – Routines for integration of resources aiming at coordination for
the capture of value from the new opportunity.
xl) Transforming – Routines for organizational renewal.
These stages are interconnected and can be verified in the innovation of

business models, as well as in their creation and implementation of new models
(Leih, Linden, & Teece, 2014).
78
Taking into account the concepts of creation and capture of value and
establishing a connection with the dynamic capabilities, according to Katkalo,
Pitelis, and Teece (2010), the activities incorporated in the high-level routines can
be grouped into two classes: Value, referring to the ability to use resources, such as
technology, for value creation and those that seek to capture value through
organizational design determined by diverse strategies, such as cooperation,
integration and diversification.
Therefore, the approach of the routine concepts, the role of the manager in the
development of dynamic capacities that can change the business model and the
capacities observed during the analysis of the process of change in the Business
model of the company.
3.3. Routines
One of the possible meanings for the word “routine” is that of an action that
repeats sequentially as a constant practice. In relation to organizations, routines are
based on sequential activities for tasks to be performed, but routines can be linked
to organizational change capabilities through learning (Fenderman and Penton,
2003). Although routinely bring the idea of rigidity, routines can present flexibility
under conditions of environmental turbulence and adapt (TEECE, 2012).
The dynamic capabilities are related to the high-level routines and present a
strategic character, different from the operational routines that represent the
ordinary capacities, responsible for execution of processes and not for
identification, prioritization and selection of processes. Ordinary capacities are
essentially responsible for the company’s operational activities, they maintain the
efficiency of activities (TEECE, 2012). The high-level routines resulting from
dynamic capabilities allow the company to prepare to manufacture products with
the best value proposition for customers and take advantage of future technological
opportunities (TEECE, 2012), they are responsible for the ways how information
and experiences will be harnessed, the exchange of experience between clients and
the organization or coordination of resources (Teece & Pisano, 1994). In this
context, the common capabilities are directly linked to the resources that will be
reconfigured by dynamic capabilities (Helfat et al., 2009).
Therefore, ordinary capacities are at a lower level if compared to the dynamic

capabilities of producing routines that integrate resources and reconfigures them
through the organizational learning process to generate competitive advantages
(Henderson & Cockburn, 1994; TEEPE; PISANO; SHUEN, 1997).
79
3.4. Role of the manager in dynamic capacities
Although business model changes occur due to the dynamic capabilities found
at various levels of the organization (LEIH, LINDEN, TEECE, 2014), in most
cases, certain dynamic capabilities may be the result of the experience of a selected
group within the organization. Usually, these people are the managers or the
entrepreneurs, whose role is to implement routines that enable alignment of
resources to achieve goals and take advantage of opportunities in the environment,
where the company is inserted, that is, to integrate, to build and to reconfigure
internal or external competences or resources on rapidly changing environments
(TEECE, 2012).
In this context, the managers are the ones, who verify opportunities for
changes in the business model, such as the moment of establishing new
partnerships with the help of high-level routines resulting from the development of
dynamic capacities to probe new scenarios, verify possibilities and enable
opportunities for resources at their disposal (TEECE, 2012).
Therefore, one can emphasize the participation of leaders, entrepreneurs,
managers and their groups responsible for the development of business models of
sensing, seizing and transforming activities to sustain dynamic capabilities,
drawing the routines, coordinating the resources (TEECE, 2012).
4. METHODOLOGY
The question that motivated this study was the identification of the high-level
routines that act in the creation of value in the business model of the investigated
company. For this purpose, exploratory qualitative approach with a single case
study was used. Two cuts were made. An initial cut before the change in the
business model, and another one after the change in the business model.
Based on the reviewed literature, it was decided to carry out in-depth

interviews with company employees, who had the necessary information to study
in the areas of information technology, Strategy and operations.
In order to capture the structure of the organization’s initial business model
and the post-change model, we used the canvas model structured together with
employees and managers from the company. Information for the development of
business model representations was complemented by analyzing documents,
telephone interviews, e-mail exchange with managers in the procurement, human
resources and investor relations sectors.
Based on the collection of information about the business model and its
changes, the dynamic capacities were identified using the Pavlou and
80
El Sawy (2011) model, submitting the questionnaire used by these authors to the
key agents in the company acting in the high-level strategic decisions, as adoption
or acquisition and use of technology aiming at finding high-level routines for the
identification, learning, integration, coordination, reconfiguration dynamic
capabilities.
In summary, the survey of the information of the high-level routines was done
through the interviews, research in the corporate website and documents presented
by the investor relations sector of the analyzed company. To this end, Bematech
was selected because it experienced singular technological and strategic changes in
its business model.
5. RESULTS AND ANALYSIS

The results are presented starting from analysis and comparison of the initial
business model and the one verified until the closing of the data collection. In
addition, the routines developed by the company are examined. At the end, a
parallel is made between the business model and the dynamic capabilities of the
company.
5.1. Bematech’s Business Model
The analysis of Bematech’s business models starts with the presentation of

aspects of the initial model of the organization, which can be observed in the
canvas model in Figure 1 filled in black. The company specialized in dot-matrix
printers. In its value proposition, the model contemplated banking printers, fiscal
printers and other hardware, such as keyboards, computers, monitors, sold together
or separately to retail customers. During this period, banks were the company’s
main customers. Due to the fiscal requirements, the printer started to be used at
points of sale, a great opportunity for growth in the retail market.
The main sales channels were represented by resellers, but there were still
direct sales to customers, such as financial institutions.
The relationship was made mainly indirectly by the resellers because since the
beginning, Bematech outsourced its service channels, except for the most potential
clients that established service contracts without inter-mediaries. Assistance
services were outsourced in the same way as resellers. The direct channel with
Bematech was customer service over the phone.
As for activities, you could check the search for the company’s growth based
on innovation activities and developing partnerships with resellers,
81
hardware vendors and representatives of service providers, in order to increase

sales.
Over the years, there have been changes in the company’s organizational
capabilities as a result of innovation in the business model.
Figure 1
Current business model Bematech

Key Key Value Customer Customer
Partners Activities Proposition Relationships Segments
• Hardware • Contract for • Printers • Technical assistance • Retail
suppliers services • Other hardwares • Internet • Corporate
• Business • Technical • Modular • Relationship center • Food service
assistance development automation • Internet • Ground
• Software • Acquisition of systems • Continuous transportation
house technology • ERP in the follow-up • Hospitality
• Software • Develop and cloud • Contract for services
companies maintain • Cloud services • Technical assistance
• Internal partnerships • Advantage program
developers • Incorporations
• Sales
• Innovation
• Software
development

Key Channels
Resources
• Employees • Sales force
• Physical • Reseller
structure • Internet
• Acquired • Developers
experience
• IT equipment
• Factory
• Developers
• Software
ownership
Cost Structure Revenue Streams
• Hardware sales
• Software sales
• User training
• Software licence
• Cloud services
Source: Elaborated by the authors.

82
After the changes perceived in Figure 1, in red, there was an increase in the
complexity of the business model, as a result of the increase of the new operational
capabilities. In the new model there are new value propositions represented in a
concept of one-stop-shop, meaning the supply of complete solutions for business
automation. Therefore, Bematech started selling complete packages of software
and hardware. The most significant changes to the initial model took place in the
value proposition and customer segment: software offers and cloud solutions
focused on customers by certain vertical clients, rather than center on products.
Thus, the teams are distributed among verticals: Corporate, food service and retail.
In this new model, the activities standing out are those related to acquisitions that
add value to the business model, developing partnerships with resellers,
distributors and software houses in order to increase the reach and meet the
company’s growth strategy.
One innovation of the new model is the existence of new digital channels for
relationship with customers and partners, such as “Bematech Plus”, a relationship
program and Bematech University, an initiative to empower partners and bring
greater synergy to the processes.
It is a partner-based model to sell products and services, provide support and
develop solutions for customers. All partners powers are highly dependent on
partners model, especially resales and software houses.
According to the director of investor relations, Bematech’s main strategies are
focused on expanding the customer base, always focusing on small and medium
retailers, increasing recurring revenues, making new acquisitions in strategic areas
of retail, expanding in Latin America and the world and offering products and
services to meet the customers' needs in commercial automation.
In this context, we highlight the new routines and operational capabilities,

such as those related to technology acquisition and incorporation of companies,
which bring new possibilities for the generation of value for investors and
customers. These activities are the basis for continued innovation, as the acquired
companies bring new technologies to be used in new value propositions and
development of solutions for new verticals. Partnerships tied to investments in
technology and the assimilation of these technologies become a powerful
combination in creating and capturing value in the business model.
5.2. Bematech’s dynamic capabilities
From the analysis of the business model, the results of the application of the
adapted Pavlou and El Sawy model (2011) could be presented to
83
identify the dynamic capacities translated into high-level routines, as well as to

point out how these high-level routines modified the ordinary or operational
capacities of Bematech; and then, the operational routines, to show the relationship
between them and the dynamic capabilities.
5.3. Bematech high-level routines
A feature found in the company that may be the basis for its growth is a
strategy focused on expansion. The content of the interviews and the analysis of
the business model revealed that the entire Bematech structure is focused on
expansion in markets, number of customers, products, services offered and above
all, that there are internal and external forces that generate the need for change, but
in this case, it could be said that the strategic direction and analysis of the scenarios
is developed in the analysis of the information produced by the various sectors and
teams, analyzed managers and the actions resulting from the analyses are decided
at meetings. In this context, the company’s high-level routines are developed,
separated according to the Pavlou and El Sawy (2011) analysis model.
5.4. Sensing
In order to fulfill its strategic goals, Bematech constantly checks new

acquisition opportunities in verticals of operation or in new market niches, such as
the recent road transport market, vertical in which it starts acting after the
acquisition of RJ, one of the leading companies in software for ticketing and
controls of these companies: the routine is based on a team for analysis of such
opportunities (Merger and acquisition team – M&A), whose main function is
identification and analysis of the acquisition opportunities. This team is divided in
people who prospect companies, others who understand the acquisition candidate's
business (degree of synergy with Bematech’s business), and a team to analyze the
impacts of the acquisition.
Another identified routine is the documentation of feedback from resellers and

customers who call the call center, but in this case, it was not possible to identify
the specific routine to organize this data and use it to bring future benefits.
According to the information technology (IT) director of Bematech, it is very

important to check the compatibility of the acquired systems with the Bematech
systems; there is a high cost of integration to be considered if the systems and the
databases are completely incompatible.
84
5.5. Learning routines
Bematech has routines to review the operational routines. The technical

training programs represented by Bematech University are reviewed constantly.
Bematech University was an initiative to build the qualification of partners, such as
resellers, in the company’s products and services and its main goal was the
partners' loyalty. In addition to the technical area, the trainings include business
representative training, such as training in marketing, finance and strategic
management. Currently, several courses are offered at distance on the University
site.
Procurement processes also reflect the learning routines. Such processes have
influence throughout the company and are considered by the outreach director as
the ones that bring more value to the company, as the efforts to integrate new
technologies are fuel for innovation and incremental improvements in products and
services. In this context, the company has professionals in the field of research and
development in software and hardware, approximately two hundred professionals,
who use routines to develop software, such as SCRUM methodology, an agile
development methodology software.
5.6. Seizing routines
The importance given to the partners' loyalty and training can be perceived in
a newer program: Bematech plus. The program is an example of review of
operational capacity and aims at restructuring the relationship channels with
customers through actions directed to equipment resellers and software partners.
According to the director of investor relations (IR), one of the main processes
within the purchasing activity is the integration process. The challenge of creating
a resource integration process between the acquired companies and Bematech
made it create a position of board responsible for this part. This director’s main
goals are to integrate information technology resources, people, building
infrastructure and processes of companies.
This integration team also acts in identifying new opportunities within the
acquisition process, how to meet new verticals, new products and services, by
developing the mix of products.
The committees are composed of directors and responsible for vertical
operations, and they also play an important role in the integration of resources.
These committees discuss problem solutions and action plans based on the
company’s strategies.
85
5.7. Coordination routines
The “Bematech Mais” (Bematech plus) plays a role of reseller coordinator.

Using the requests for information provided by retailers on site, you can perform a
check on resales to encourage the sales mix of specific products or perform actions
on specific niches of customers.
Another routine that must be aligned to the strategic goals of the company’s
annual activity review and direct the budget for the development of products
targeted to the vertical target, for example, in recent years the company has
invested in support services structures online, such as storage, processing and
software cloud in management. This practice is aligned with the strategy of
increasing recurring revenue from monthly payments or for certain periods of time
using a software or rental space and processing in the cloud.
5.8. Transforming
The processing capacity is required when the organization has to deal with
threats or exploit opportunities (Leih, Linden, & Teece, 2014). It is the result of
identification, learning, integration and coordination dynamic capabilities.
In this context, it can be said that the new business model is the result of this
process. According to the director of investor relations, new develop-ment in goods
and services each year is related to innovation routines, acquisition and exploration
of market opportunities.
5.9. Value creation and capture
From the business model theory, we found value creation. According to

Osterwalder and Pigneur (2010), value creation occurs from the progress of the
proposed value or creating new value propositions. In this context, the addition of
new value propositions and their acceptance by customers already demonstrates
that there has been value creation and value capture possibility for the company.
This increase business model about the value proposition and its ability to capture
value for new channels that bring opportunities to win more customers and thus,
create other sources of income, can be seen in Figure 2.
86
Figure 2
Value creation in the Bematech’s business model
Value Proposition Customer Relationships Customer Segments
• Printers • Technical assistance • Retail

• Other hardwares • Internet • Corporate
• Modular automation systems • Relationship center • Food service
• ERP in the cloud • Internet • Ground transportation
• Cloud services • Continuous follow-up • Hospitality
• Contract for services
• Technical assistance
• Advantage program

Channels
• Sales force
• Reseller
• Internet
• Developers
Revenue Streams
• Hardware sales
• Software sales
• User training
• Software licence
• Cloud services
From the perspective of dynamic capabilities, high-level routines are related to

the creation and value capture in business model and are responsible for the
reconfiguration of operational capabilities and routines to meet new customer
niches, offer new products and develop service channels.
The vision of dynamic capabilities brought the possibility of relating the
routines of high-level Sensing, Seizing and Transforming (Figure 3), a task that
would not be possible only from the analysis of the high-level routines in business
model that create and capture value in business model.
87
Figure 3
High-level routines that create and capture

value in business model
Bematech Sensing Seizing Transforming
• Constant verification of • Programs of technical. • Software partner
acquisition opportunities and commercial training management routines,
UNUM (hotel industry), of channels (Bematech Bematech more.
RJ (road transport). University). • Budget for product
• Constant verification of • Research and improvement
product creation development (200 (investment in recurrent
opportunity. professionals). revenue products).
• Committees. • Reseller management
Value • Rethink the business routines.
creation through new channel
strategies with
well-defined rules and
policies (Bematech Plus).
• Analysis of feedback
from resellers and
customers.
• Product review for
improvement.
• Merge & acquisition • Verification of the • Development of new
Merge & acquisition opportunities that the products and services
team. acquisitions bring during annually (Multi-POS,
• Routines for feedback the process of promotions engine and
from resellers and adjustments between Bematech Shop Mobile)
Value customers. companies. – FOOD SERVICE GDI
Capture • Product review for • Opportunities for new (indicator manager).
improvement. products and services. • Bematech PDV, rear web
• Bematech more. chef for sale SaaS.
• Committees. • Changes in the business
• Merge & acquisition model.
team.
After checking the high-level routines of the company, we can carry out the
analysis of routines that have brought changes or created the perfect environment
for such changes in the business model. Connections can be made to the extent that
most of the changes in the business model are connected to one or more routines.
The analysis allows one to check
88
which business model elements the company is more focused on, in what the
business model element has a greater connection with the high-level routines of the
company.
5.10. Connection between routines and

changes in the business model
Figure 4
Business model and high-level routines

Business model High-level routines
New value propositions: modular • Constant verification of acquisition opportunities.
automation systems, ERP in the • Constant selection of product creation opportunity.
cloud, cloud services • Research and development (200 professionals).
• Analysis of feedback from customers and resellers.
• Reviewing products for improvements.
• Verification of the opportunities the acquisitions bring
during the process of adjustments between companies.
• Budget to improve products.
• Development of new products and services annually.
New customers: corporate, food • Verification of the opportunities the acquisitions bring
service, hospitality, ground during the process of adjustments between companies.
transportation • Constant verification of procurement opportunities.
• Merge & acquisition team.
New channel: agent developers • Rethinking the business through new channel strategies
with well-defined rules and policies (bematech mais).
• Technical and commercial training programs for channels
(bematech university).
• Software management routine partners, bematech more.
• reseller management routines.
New key activities: the acquisition of • Rethinking the business through new channel strategies
technology, maintaining partnerships, with well-defined rules and policies (bematech mais).
software development • Committees.
• Merge & acquisition team.
New partners: software houses,

• To rethink the business through new channel strategies
software companies, internal
with well-defined rules and policies (bematech more).
developers
New key resources: developers, • To rethink the business through new channel strategies
software with well-defined rules and policies (bematech more).
New customer relationships: benefits • Budget for product improvements (investment in
program for customers recurring revenue products).

89
Figure 4 demonstrates that the company Bematech maintains its focus on

creating new value propositions, since most of its high-level routines identified in
the study are focused on this element of the business model. Consequently, the
company will have the potential to serve more customer niches and develop new
sources of revenue.
6. DISCUSSION
An indication of the higher capacity of the new model for the increase in
revenues was obtained by checking the economic results recorded in the financial
statements of the company Bematech. There was an increase in revenues, mainly in
recurring revenues – those from licensing software and systems to use software as
service (SAS), such as Enterprise resource planning (ERP) in the cloud. The
increase in revenues alone would not be a robust evidence that the new business
model is more efficient to generate financial returns; however, the accounting
information, documents and interviews of people responsible for the financial
statements indicated a growth in the proportion of recurring revenues compared to
the total revenues of the company. The survey of this information in recent years
demonstrates an upward trend in these revenues, that is, evidence that the new
model has shown an improvement in the ability to generate revenues and better
economic results.
Dynamic capabilities enabled to understand the value creation process

mediated by operational capabilities, and the case studies reaffirmed the trend of
the dynamic capabilities theory in representing, with quality, a theoretical
framework for the analysis of aspects related to the creation of value and business
models.
In this context, if the analysis was limited to the business model and did not
bring the perspective of dynamic capabilities, it would be possible to obtain a list
of what has changed in the business model; however, the actions that caused such
changes could not be understood in the analysis of the business model components.
It could be seen, especially in interviews with managers, that the high-level

routines were developed by the senior management and that stocks of high-level
routines require the manager to be engaged effectively. This information confirms
the Leih, Linden, and Teece's (2014) statement that the manager’s role is essential
both in the creation of high-level routines, as the harvesting phase, and the results
from these routines, that is, if the
90
information obtained from the capabilities dynamics, such as identifying an

opportunity, is not exploited or understood to implement a change in the business
model, such routines will be useless.
As for the process of change and value creation, it can be assumed that current
business models would not exist if there were no technological changes. Business
models have been modified by checking the value creation potential of new
technological resources or those existing and already available in companies, i.e.,
the business model was designed after checking the resources within the company.
According to the company's managers, the combination of these features, together
with the creation of different business models, enabled the company to create a
differentiated value proposition related to competitors. These new proposals,
implemented and accepted by customers, meant the value created by the company
and the ability to capture value.
Some routines have played an important role in changing the business model,
such as those in surveys to identify new acquisitions, learning routines to employ
the technologies acquired in mergers, integration routines of acquired companies
and product review routines to detect opportunities for improvement.
The acquisitions include two moments of investments: the acquisition of the

company’s control and the cost of integrating the new company to Bematech.
Therefore, Bematech has the feature of acquiring the technologies it needs

from financial transactions involving large amounts. The development of new
products and services is a consequence of the technologies built into the business
model.
It was noticed that Bematech used the technological infrastructure to bring the
actors of the business model. Evidence of this are the joint efforts in information
technology to benefit resellers and software partners in a complex channel strategy.
This approach reveals the importance attributed to the sales channels and the reach
in the market. Another important part of the use of information technology is the
development of products and services, these result from the routines of acquisition
and integration of new technologies and innovation processes in the areas of
software and hardware.
In short, the connection between the business model and dynamic capabilities
in this study occurred multilaterally, and the high-level routines were fundamental
to the innovation of the business model and impacted various elements of the
business model. Such routines acted as true sources of change, providing the ideal
environment for change.
91
7. CONCLUSIONS
Theories related to the business model, value creation and dynamic
capabilities demonstrate the existence of overlaps and important relation-ships,
when addressing topics, such as strategy and organizational structure. In this
context, the vision of dynamic capabilities allowed the process of creating value in
business models driven by changes in the technological structure of the companies,
object of this study, to be explored, which made the achievement of the overall
goals of the work possible.
The theory of business models proved to be multidisciplinary, a view that was
formed under the influence of other theories and practical applica-tion in any
organization. The analysis of business models enabled bringing an analytical
overview of the information technology and cleared its role in creating value for
the company.
Understanding the process of creating value in the proposed goal of the work
could have been addressed from other perspectives, such as resource-based view
(RBV), the theory of transaction costs, contingency theory, among others.
In the demonstration of the business model, it was not possible to verify the
dynamic capacity as part of the business model; however, it can be a confirmation
in this particular case, since it is a single case study. Therefore, in the case under
analysis, each dynamic capacity could be considered as an external factor to the
business model. The key activities observed in the representation of the business
model could be those closer to a high-level routine; however, it was observed that
the high-level routines that altered the key activities are not confused. Some
ordinary routines might be observed in key business model activities, for example,
acquiring partnership, maintaining partnerships and software development. These
activities are necessary to the business model, since they are part of the logic that
explains how the company creates and captures value, but they cannot be
considered high-level routines, since they do not modify other routines. Therefore,
it was possible to observe that the dynamic capacities change business models;
however, it was not possible to observe whether business models participate in the
development of dynamic capacities.
REFERENCES
Anderson, J. C., & Narus, J. A. (1998). Business marketing: understand what
customers value. Harvard business review, 76, 53-67.
92
Augier, M., & Teece, D. J. (2008). Strategy as evolution with design: The
foundations of dynamic capabilities and the role of managers in the economic
system. Organization studies, 29(8-9), 1187-1208. doi 10.1177/017084060
8094776
Baden-Fuller, C., & Morgan, M. S. (2010). Business models as models. Long
range planning, 43(2), 156-171. doi 10.1016/j.lrp.2010.02.005
Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of
management, 17(1), 99-120. doi 10.1177/014920639101700108
Barney, J. B., & Arikan, A. M. (2001). The resource-based view: Origins and
implications. Handbook of strategic management, 124188. doi 10.1111/b.97806
31218616.2006.00009.x
Bock, A. J., Opsahl, T., George, G., & Gann, D. M. (2012). The effects of culture
and structure on strategic flexibility during business model inno-
vation.JournalofManagementStudies,49(2),279-305.doi10.1111/j.1467-6486.
2011.01030.x
Bowman, C., & Ambrosini, V. (2000). Value creation versus value capture: towards
a coherent definition of value in strategy. British Journal of Manage-ment, 11(1),
1-15.
Boulton, R., Libert, B., & Samek, S. M. (2000). Cracking the Value Code: How
successful businesses are creating wealth in the New Economy. New York, NY:
HarperBusiness. doi 10.1111/1467-8551.00147
Boulton, R. E., Libert, B. D., & Samek, S. M. (2000). A business model for the new
economy. Journal of Business Strategy, 21(4), 29-35. doi 10.1108/ eb040102
Cavalcante, S., Kesting, P., & Ulhøi, J. (2011). Business model dynamics and
innovation: (re) establishing the missing linkages. Management Decision, 49(8),
1327-1342. doi: 10.1108/00251741111163142
Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capabilities: what are they?
Strategic management journal, 1105-1121. doi 10.1002/1097-0266
(200010/11)21:10/ 11<1105::AID-SMJ133>3.0.CO;2-E
Grant, R. M. (1996). Toward a knowledge-based theory of the firm. Strategic
management journal, 17(S2), 109-122. doi 10.1002/smj.4250171110
Helfat, C. E., Finkelstein, S., Mitchell, W., Peteraf, M., Singh, H., Teece, D., &
Winter, S. G. (2009). Dynamic capabilities: Understanding strategic change in
organizations.
John Wiley & Sons.Henderson, R., & Cockburn, I. (1994). Measuring competence?
Exploring firm effects in pharmaceutical research.Strategic management journal,
15(S1), 63-84. doi 10.1002/smj.4250150906
93
Katkalo, V. S., Pitelis, C. N., & Teece, D. J. (2010). Introduction: On the nature and
scope of dynamic capabilities. Industrial and Corporate Change, 19(4), 1175-
1186. doi 10.1093/icc/dtq026
Kraaijenbrink, J., Spender, J. C., & Groen, A. J. (2010). The resource-based view: a
review and assessment of its critiques. Journal of management, 36(1), 349-372.
doi 10.1177/0149206309350775
Leih, S., Linden, G., & Teece, D. (2014). Business model innovation and
organizational design: A dynamic capabilities perspective. Oxford: Oxford
University Press. Retrieved from https://papers.ssrn.com/sol3/papers. cfm?
abstract_id=2423191
Lindgreen, A., Hingley, M. K., Grant, D. B., & Morgan, R. E. (2012). Value in
business and industrial marketing: Past, present, and future. Industrial Marketing
Management, 41(1), 207-214. doi 10.1016/j.indmarman.2011. 11.025
Magretta, J. (2002). Why business models matter. Harvard Business Review, 80(5),
p. 86-92. Retrieved from https://hbr.org/2002/05/why-business-models-matter
Osterwalder, A., & Pigneur, Y. (2010). Business model generation: a handbook for
visionaries, game changers, and challengers. Journal of Product Innovation
Management, 26(6). doi 10.1111/j.1540-5885.2012.00977_2
Osterwalder, A. (2004). The business model ontology: A proposition in a design
science approach. Master these, Ecole des Hautes Etudes Commerciales,
Université de Lausanne, Lausanne, Switzerland.
Pavlou, P. A., & El Sawy, O. A. (2011). Understanding the elusive black box
ofdynamiccapabilities.DecisionSciences,42(1),239-273.doi10.1111/j.1540-
5915.2010.00287.x
Penrose, E. T. (1995). The Theory of the Growth of the Firm. Oxford University
Press, USA. doi: 10.1093/0198289774.001.0001
Porter, M. E. (1985). Competitive advantage: creating and sustaining superior
performance. Revista de Administração de Empresas, 25(2). doi 10.1590/ S0034-
75901985000200009
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic
management. Strategic management journal, 509-533. doi 10.1002/ (SICI)1097-
0266(199708)18:7<509::AID-SMJ882>3.0.CO;2-Z
Teece, D. J. (2010). Business models, business strategy and innovation. Long
range planning, 43(2), 172-194. doi 10.1016/j.lrp.2009.07.003
94
Teece, D. J. (2007). Explicating dynamic capabilities: the nature and micro-

foundations of (sustainable) enterprise performance. Strategic management
journal, 28(13), 1319-1350. doi 10.1002/smj.640
Teece, D. J. (2012). Dynamic capabilities: Routines versus entrepreneurial action.
Journal of Management Studies, 49(8), 1395-1401. doi 10.1111/j.1467-
6486.2012.01080.x
Teece, D. J., & Pisano, G. (1994). The dynamic capabilities of firms: an
introduction. Industrial and corporate change, 3(3), 537-556. doi 10.1093/
icc/3.3.537-a
Vodovoz, E. (2015). As mudanças na tecnologia da informação e a criação de valor
no modelo de negócio. http://hdl.handle.net/1884/39878
Winter, S. G. (2003). Understanding dynamic capabilities. Strategic manage-ment
journal, 24(10), 991-995. doi: 10.1002/smj.318
Zott, C., Amit, R., & Massa, L. (2011). The business model: recent develop-ments
and future research. Journal of management, 37(4), 1019-1042. doi
10.1177/0149206311406265
ABOUT THE AUTHORS

ELIC VODOVOZ
Master degree in Business,
Professor, ESIC Business & Marketing School.
Rua Curupaitis, 1963, – Portão – PR – Brasil – CEP 80330-030
E-mail: elicvz@me.com
MARCIA RAMOS MAY

PhD in Business,
Universidade de São Paulo (USP).
Professor, Universidade Federal do Paraná (UFPR).
Rua prefeito Lothário Meissner 632, Jardim Botânico – Curitiba – PR – CEP 80210-170 E-mail:
marciamay@gmail.com
95
Revista de Administração Mackenzie – RAM (Mackenzie Management Review), 18(6), 71-95 • SÃO
PAULO, SP • NOV./DEC. 2017 • ISSN 1678-6971 (electronic version) • doi 10.1590/1678-
69712017/administracao.v18n6p71-95
Journal of Vocational Behavior 62 (2003) 341–356
www.elsevier.com/locate/jvb
Job demands and job resources as predictors of

q
absence duration and frequency
Arnold B. Bakker, Evangelia Demerouti,c,d
a a
Elpine de Boer, and Wilmar B. Schaufeli
ii Department of Social and Organizational Psychology, Research Institute Psychology and Health,
Utrecht University, Heidelberglaan 1, P.O. Box 80.140, 3508 TC Utrecht, The Netherlands
xli) HCMG Human Capital Management Group, Utrecht, The Netherlands
(BB) Department of Work and Organizational Psychology, University of Nijmegen, Nijmegen, The
Netherlands
*
Department of Work and Organizational Psychology, Carl von Ossietzky University Oldenburg,
Oldenburg, Germany
Received 20 June 2001
Abstract
This study among 214 nutrition production employees uses the Job Demands–Resources (JD–
R) model to predict future company registered absenteeism. According to this model, job demands
are primarily responsible for health impairment, whereas job resources lead pri-marily to increased
motivation and attachment to work and the organization. Consistent with hypotheses derived from
the JD–R model and the absenteeism literature, results of structural equation modeling analyses
show that job demands are unique predictors of burnout (i.e., ex-haustion and cynicism) and
indirectly of absence duration, whereas job resources are unique predictors of organizational
commitment, and indirectly of absence spells. These findings have implications for individual and
organizational interventions aimed at reducing absenteeism.
2003 Elsevier Science (USA). All rights reserved.
Keywords: Job demands; Job resources; Burnout; Commitment; Absenteeism
- This research is part of the concerted research action on ‘‘Fatigue at Work’’ granted by the
*
Netherlands Organization for Scientific Research (NWO) (#580-02-202).
Corresponding author. Fax: +31-30-2537584.
E-mail address: a.bakker@fss.uu.nl (A.B. Bakker).
0001-8791/03/$ - see front matter 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0001-8791(02)00030-1
- A.B. Bakker et al. / Journal of Vocational Behavior 62 (2003) 341–356
1. Introduction
There exists basic agreement that job stress relates to physical and behavioral out-
comes such as health complaints, burnout, and absenteeism (Kahn & Byosiere, 1992).
However, in contrast to burnout and health complaints, absenteeism, while detrimental to
the organization, is not necessarily harmful to those employees who are absent from
work. Quite the contrary, their absence may be instrumental for re-cuperating from
experienced job stress. In the current study, we use the Job De-mands–Resources model
(Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) to examine how di fferent categories
of work characteristics influence future absenteeism among Dutch production workers,
through their relationships with burnout and or-ganizational commitment.
1.1. Absenteeism
Generally, two different absence measures are distinguished: absence frequency and
duration (Hensing, Alexanderson, Alleback, & Bjurulf, 1998). Absence frequency is the
number of spells or times an individual has been absent during a particular pe-riod,
regardless of the length of each of those spells. Usually, absence frequency is considered
to be an indicator of ‘‘voluntary absenteeism’’ and a function of employ-ees motivation.
In contrast, absence duration is the total length of time an individual has been absent over
a specified period regardless of the number of absence spells. Absence duration is
generally considered to be an indicator of ‘‘involuntary absen-teeism’’ that results from
the inability rather than the unwillingness to come to work, for example as a result of ill
health. The correlation between absence frequency and duration ranges between a low
).05 and a moderately high .60 (see Farrell & Stamm, 1988).
Since absenteeism includes different components (i.e., frequency and duration), there
seem to exist different processes that lead to frequent or long absenteeism (Kohler &
Mathieu, 1993). Indeed, most empirical studies that focus on individual experiences at
work as precursors of absenteeism can be classified along two main explanations for
employees decision to report themselves sick (Johns, 1997). First, employees may be
absent because they want to withdraw from aversive work circum-stances. Using a
general ‘withdrawal’ hypothesis, it has been found that employees who are low in job
satisfaction and organizational commitment are more frequently absent than those high in
job satisfaction and commitment (e.g., Cohen, 1991; Far-rell & Stamm, 1988; Mathieu &
Kohler, 1990; Sagie, 1998). In these studies, absen-teeism is usually interpreted as an
escape from, compensation for, or even protest against aversive or demoralizing work
circumstances (cf. Chadwick-Jones, Nichol-son, & Brown, 1982). This agrees with the
notion of voluntary absenteeism.
A second explanation for absenteeism is that absence behavior is a reaction to job
stress, where stress is conceived as a failure to cope with job demands. This explana-tion
stipulates that absenteeism may be used as a coping mechanism to deal with job strain
and that it is not simply a behavioral reaction to dissatisfaction (Johns, 1997). Several
stressors (i.e., job related factors thought to cause negative psychological
A.B. Bakker et al. / Journal of Vocational Behavior 62 (2003) 341–356 343
reactions like tension, anxiety, and fatigue) such as workload (Dwyer & Ganster, 1991),
monotony (Melamed, Ben-Avi, Luz, & Green, 1995), and role problems (Ja-mal, 1984)
have indeed been associated with higher absence rates. However, Johns (1997) has
observed that while various studies have reported relationships between stressors and
absenteeism, tests of mediation models (stressor ! stress reac-tions ! absence), including
the mediating role of stress reactions, are rare.
1.2. Organizational commitment and absenteeism
Meyer and Allen (1991) consider commitment as a multidimensional concept in-

cluding three components: affective, normative and continuance commitment. A ffec-tive
commitment refers to employees emotional attachment to, identification with and
involvement in the organization, whereas normative commitment refers to em-ployees
attachment to the organization and to its goals because of ideology or felt obligation.
Finally, continuance commitment refers to a general awareness of the costs of leaving the
organization or to the perceived number of employment alterna-tives and degree of
sacrifice.
Most absenteeism studies have examined the correlates of affective and continu-ance
commitment (Gellatly, 1995). Affective commitment, that is expected to in-crease when
work experiences are personally rewarding, has consistently been found to relate
negatively to absenteeism (see Johns, 1997, for a review). In contrast, continuance
commitment is expected to encourage absence behavior. As Brehm (1966) noted, feeling
locked in might provoke reactance expressed in short episodes of escape. This positive
relationship between continuance commitment and absentee-ism (frequency) has indeed
been confirmed in some studies (e.g., Gellatly, 1995), al-though other studies found no
relationship (e.g., Allen & Meyer, 1990; Somers, 1995). Normative commitment is
expected to stimulate attendance due to the feeling of obligation. However, there is
hardly any empirical support for the relationship be-tween normative commitment and
absenteeism (Gellatly, 1995).
Following the withdrawal paradigm, people will be more likely to withdraw from
organizations to which they lack commitment. Indeed, Farrell and Stamm (1988) found a
corrected mean correlation of ).12 between commitment and absence dura-tion in their
meta-analysis including 11 samples. Interestingly, they found a higher corrected mean
correlation ().23) when they restricted themselves to the six samples that measured
absence frequency, which agrees with our previous reasoning that ab-sence frequency
primarily reflects voluntary absence. Furthermore, Cohen (1991) re-ports a corrected
mean correlation of ).11 between commitment and absence on the basis of 11 studies.
Thus, in general, the relationship between organizational com-mitment and absenteeism
is rather low with absence frequency being more strongly related than absence duration.
1.3. Burnout and absenteeism
Burnout can be defined in general terms as a syndrome of exhaustion, cynicism and

reduced professional efficacy (Maslach, Jackson, & Leiter, 1996). Whereas
emotional exhaustion and cynicism (or depersonalization) have been considered as the
core dimensions of burnout, feelings of reduced efficacy seem to play a di fferent role.
For instance, reduced efficacy may also be interpreted as a possible consequence of
burnout (Koeske & Koeske, 1989; Shirom, 1989). Furthermore, there is accumu-lating
evidence that personal accomplishment largely develops in parallel with the two other
burnout dimensions (Lee & Ashforth, 1996; Schaufeli & Enzmann, 1998). These findings
support the notion that emotional exhaustion and depersonal-ization (or cynicism)
constitute a syndrome, which is loosely related to professional e fficacy. Therefore,
professional efficacy is excluded from our research model.
Absenteeism is generally considered as an important consequence of burnout at the
organizational level. However, emotional exhaustion, depersonalization and re-duced
efficacy explain on average not more than 2% of the variance in absenteeism (e.g.,
Lawson & O Brien, 1994; Price & Spence, 1994). After reviewing the literature,
Schaufeli and Enzmann (1998) conclude therefore that: ‘‘. . . despite the popular as-
sumption that burnout causes absenteeism, the effect is rather small and is most re-lated
to emotional exhaustion’’ (p. 91). Indeed, several meta-analytic studies on absenteeism
show that work-related stress is but one of many variables accounting for employee
absence behavior, so we should not expect job stress and absenteeism to be strongly
correlated (Beehr, 1995; Nicholson, 1993). Non-work variables ac-counting for
absenteeism include a wide range of factors, such as personal character-istics, sport
injuries, smoking, alcohol consumption, psychological disorders, and physical pain (see
Johns, 1997; Youngblood, 1984). These non-work variables may also interact with work-
related variables, and show complex relationships with ab-senteeism. For example, in
their study among 211 employed, married parents, Erick-son, Nichols, and Ritter (2000)
found that family demands moderated the effect of job burnout on absence frequency.
Experiencing a high level of burnout was associ-ated with increased absenteeism if
employees had children under 6 living at home, or reported having di fficulty with their
child care arrangements.
1.4. The Job Demands–Resources model
At the heart of the Job Demands–Resources (JD–R) model (Demerouti et al., 2001)
lies the assumption that whereas employees in different organizations may be confronted
with different working environments, the characteristics of these envi-ronments can be
always classified in two general categories—job demands and job resources—thus
constituting an overarching model that may be applied to various occupational settings,
irrespective of the particular demands and resources involved. Job demands refer to those
physical, psychological, social, or organizational aspects of the job that require sustained
physical and/or psychological (cognitive and emo-tional) effort and are therefore
associated with certain physiological and/or psycho-logical costs. Examples are a high
work pressure, role overload, poor environmental conditions and problems related to
reorganization. Job resources refer to those phys-ical, psychological, social, or
organizational aspects of the job that are either/or: (1) functional in achieving work
goals; (2) reduce job demands and the associated phys-iological and psychological costs;
(3) stimulate personal growth and development.
Resources may be located at the level of the organization at large (e.g., pay, career
opportunities, job security), at the interpersonal level (e.g., supervisor and co-worker
support, team climate), at the level of the organization of work (e.g., role clarity, par-
ticipation in decision-making), and at the task level (e.g., performance feedback, skill
variety, task significance, task identity, autonomy; see also Hackman & Oldham, 1976).
A second proposition in the JD–R model is that work characteristics may evoke two
different processes. First, high job demands (i.e., work overload) may exhaust employees
mental and physical resources and may therefore lead to health problems or burnout
(e.g., Demerouti, Bakker, Nachreiner, & Schaufeli, 2000, 2001; Lee & Ashforth, 1996;
Leiter, 1993). Second, poor or lacking job resources preclude actual goal
accomplishment, which is likely to cause failure and frustration. In its turn this may lead
to withdrawal from work, and reduced motivation or commitment. When the external
environment lacks resources, individuals cannot reduce the potentially negative influence
of high job demands and they cannot achieve their work goals. In such a situation,
reducing commitment can be an important self-protection mech-anism that may prevent
the future frustration of not obtaining work-related goals (cf. Antonovski, 1987;
Hackman & Oldham, 1976, 1980).
1.5. The present study
This study uses the JD–R model to examine how job demands and job resources
influence absence duration and frequency among Dutch production employees, through
their relationship with burnout and organizational commitment. On the ba-sis of this
model, we hypothesize that the work environment influences employees absence
behavior in two different ways. First, we expect that demanding aspects of work (e.g.,
extreme job demands) lead to impaired health (i.e., burnout). Therefore, and in
accordance with the stress explanation for absenteeism, we predict that job demands (and
not job resources) will have a positive impact on absence duration, through the
experience of burnout (Hypothesis 1). In other words, we expect that burnout will play a
mediating role in the relationship between job demands and total number of days absent,
being an indicator of strain-related absence from work.
Second, we expect that job resources facilitate actual goal accomplishment (in-
cluding dealing with demands), which provokes feelings of success, which further
enhance organizational commitment. Therefore, and in line with the withdrawal
explanation for absenteeism, we predict that job resources (and not job demands) will
have a negative effect on absence frequency, through their (positive) influence on
organizational commitment (Hypothesis 2). Thus, we expect that commitment will play a
mediating role in the relationship between job resources and absence frequency, being an
indicator of voluntary absence from work. Both hypotheses are graphically depicted in
Fig. 1. Finally, the two absence measures were as-sumed to influence each other, since
the frequency measure may also include ab-sence due to involuntary factors such as
illness, and the duration measure may also include voluntary, or avoidable, absence (cf.
Thomson, Griffiths, & Davison, 2000).
Fig. 1. The Job Demands–Resources model applied to absenteeism.
2. Method
2.1. Participants and procedure
In March 1998, a questionnaire was distributed among all 330 employees of a nu-
trition production company in The Netherlands. Employees were kindly requested to fill
out the questionnaire, and to post it in a special box at their workplace. The con-
fidentiality of the answers was emphasized. By means of a unique code, question-naire
data could be linked with absenteeism data that were retrieved from the company s
computerized registration system. A total of 214 employees filled out and returned the
questionnaire (response rate ¼ 65%). The sample included 147 males (69%) and 67
females (31%). Their mean age was 41 years (SD ¼ 9) and mean organizational tenure
was 13 years (SD ¼ 9). Most employees worked full-time (91%). Meetings with
management and employees representatives facilitated the identification of the most
important and relevant job demands (workload and prob-lems with reorganization) and
job resources (job control and participation in deci-sion making) at the time of the study.
2.2. Measures
Job demands. Workload was assessed with five items from the Dutch version (Fur-da,
1995) of Karasek s (1985) Job Content questionnaire. The scale includes items
that refer to quantitative, demanding aspects of the job (e.g., working hard, having too
much work to do). A sample item is: ‘‘My work requires working very hard.’’ Items are
scored on a four-point Likert-scale, ranging from (1) ‘‘never’’ to (4) ‘‘al-ways.’’ The
internal consistency of the job demands scale was satisfactory: Cron-bach s a was .74.
Problems with reorganization were assessed with a four-item scale based on Van
Veldhoven & Meijman (1994), including ‘‘Do changes in your tasks pose di fficulties to
you? (1 ¼ never, 4 ¼ always). Cronbach s a was .76.
Job resources. Job Control was assessed with six items from Van Veldhoven &
Meijman s (1994) questionnaire (see also De Jonge, Bosma, Peter, & Siegrist, 2000). The
items refer to employees control regarding job content and the timing of work tasks. An
example item is: ‘‘Are you allowed to decide by yourself how to perform your work?’’
Participants could use the same answer categories as for job demands (Cronbach s a ¼ .
91). Participation in decision-making was measured with a six-item scale based on
Karasek (1985), including ‘‘I have influence over de-cisions that are made by my
supervisor’’ (1 ¼ never, 4 ¼ always). Cronbach s a was
.86. All responses were coded such that higher scores referred to higher job demands and
more job resources, respectively.
Burnout was assessed using the Dutch version (Schaufeli & Van Dierendonck, 2000)
of the Maslach Burnout Inventory-General Survey (MBI-GS; Schaufeli, Leiter, Maslach,
& Jackson, 1996). Two sub-scales from this inventory were used in the current study,
namely exhaustion and cynicism. Exhaustion was measured with five items, such as: ‘‘I
feel emotionally drained from my work.’’ The internal consis-tency of the exhaustion
scale was high: Cronbach s a was .88. The second burnout dimension, cynicism, was also
assessed with five items, including ‘‘I have become less enthusiastic about my work’’
(Cronbach s a ¼ .72). All items were scored on a seven-point rating scale, ranging from
(0) ‘‘never’’ to (6) ‘‘every day.’’ High scores on ex-haustion and cynicism are indicative
for burnout.
Organizational Commitment was measured using a reliable and validated Dutch
version (De Gilder, Van den Heuvel, & Ellemers, 1997) of Meyer & Allen s (1991) three-
dimensional questionnaire. Exemplary items are: ‘‘I feel emotionally attached to the
organization’’ (affective commitment; Cronbach s a ¼ .87); ‘‘I believe that peo-ple
should be loyal to their organization’’ (normative commitment; Cronbach s a ¼ .90); and
‘‘Right now, I have the feeling that I have too few alternatives to quit from my job’’
(continuance commitment; Cronbach s a ¼ .89). All items were scored on a five-point
rating scale, ranging from (1) ‘‘not at all’’ to (5) ‘‘to a large extent.’’ High scores on
affective, normative and continuance commitment are indicative for organizational
commitment.
Absenteeism data were retrieved from the company s computerized registration
system. We included absence frequency (i.e., amount of absence spells) and ab-sence
duration measures (i.e., total days lost). A period of one year was chosen to increase
stability in the absence measures (Hammer & Landau, 1981). Ab-sence data for each
participant were collected during the one-year period follow-
ing the administration of the questionnaire. The mean absence frequency was 1.18 (SD
¼ 1:31) times and the mean absence duration was 15.92 days (SD ¼ 34:94).
353 A.B. Bakker et al. / Journal of Vocational Behavior 62 (2003) 341–356
2.3. Analyses
The model was tested with structural equation modeling (SEM) analyses using the
AMOS computer program (Arbuckle, 1997). Maximum likelihood estimation meth-ods
were used and the covariance matrix of the items was the input for the analysis. The
goodness-of-fit of the model was evaluated using absolute and relative indices. The
absolute goodness-of-fit indices calculated were the v2 goodness-of-fit statistic and the
root mean square error of approximation (RMSEA; Browne & Cudeck, 1993). Non-
significant v2-values indicate that the hypothesized model fits the data, and RMSEA-
values smaller than or equal to .08 are indicative of an acceptable fit (Browne & Cudeck,
1993).
However, the v2 goodness-of-fit statistic is sensitive to sample size, so that the
probability of rejecting the hypothesized model increases with increasing sample size. To
overcome this problem, the computation of relative goodness-of-fit indices is strongly
recommended (Bentler, 1990). As suggested by Marsh, Balla, & Hau (1996), we used the
Non-Normed Fit Index (NNFI), the Incremental Fit Index (IFI), and the Comparative Fit
Index (CFI). For these relative fit-indices, as a rule of thumb, values of .90 or higher are
considered as indicating a good fit (Hoyle, 1995).
The latent exogenous factors, job demands and job resources, were both opera-
tionalized by two exogenous observed variables (see Fig. 1). The manifest indicators of
job demands were workload and problems with reorganization. Job resources were
indicated by job control and participation in decision-making. In addition, the structural
model includes two types of endogenous variables: (1) burnout and or-ganizational
commitment as latent (mediator) variables, and (2) absence duration and frequency as
observed variables. The latent endogenous factor burnout was as-sessed by two observed
variables, namely exhaustion and cynicism, whereas the la-tent endogenous factor
organizational commitment was indicated by affective, normative and continuance
commitment.
3. Results
3.1. Descriptive statistics
Means, standard deviations and correlations among all study variables are pre-sented
in Table 1. Both job demands show weak to moderate negative correlations with the two
job resources. The two absenteeism measures show a moderate positive correlation (r ¼ :
45), and they both correlate significantly with the specific job de-mands and job
resources (except reorganization–absence frequency). In our study, only the total duration
of absenteeism has considerable skewness (i.e., 3.34) and kur-tosis (i.e., 11.34), since the
skewness index is larger than 2 and the kurtosis index is larger than 5 (cf. Kendall &
Stuart, 1958). For absence frequency, the skewness and kurtosis values were 1.59 and
3.10, respectively. Therefore, we decided to utilize the log-transformed scores of absence
duration and the raw scores of absence frequency in the following analyses.
A.B. Bakker et al. / Journal of Vocational Behavior 62 (2003) 341–356
Table 1
Means, standard deviations and correlations of the variables included in the SEM-analyses, N ¼ 214
M SD 1 2 3 4 5 6 7 8 9 10
1. Workload 1.88 .51
2. Reorganization 1.87 .42 .38
3. Job control 2.52 .65 ).36 ).19
4. Participation 1.63 .76 ).20 ).31 .41
5. Exhaustion 1.69 1.13 .46 .36 ).29 ).19
6. Cynicism 1.58 1.04 .31 .32 ).23 ).38 .62
7. Affective commitment 3.24 .83 ).16 ).18 .26 .46 ).25 ).32
8. Continuance commitment 3.12 1.18 .14 .18 ).28 ).13 .16 .14 .15
9. Normative commitment 2.66 1.07 .03 .02 ).19 .09 ).03 ).09 .37 .46
10. Absence duration 1.18 1.31 .19 .16 ).23 ).15 .16 .13 ).07 .20 .13
11. Absence frequency 15.92 34.94 .24 .04 ).19 ).15 .06 .10 ).06 .24 .08 .45
*
p < :05.
**
p < :01.
354 349A.B. Bakker et al. / Journal of Vocational Behavior 62 (2003) 341–356
3.2. Model testing
The proposed model including all hypothesized relationships (see Fig. 1) was tested
with SEM-analyses. Results indicated that the model did not fit adequately to the data,
v2ð41Þ ¼ 201:36; GFI ¼ :84; RMSEA ¼ :14; CFI ¼ :76; IFI ¼ :76, and NNFI ¼ :68.
Inspection of the modification indices revealed that this lack of fit be-tween the model
and the data was mainly due to covariations between the mea-surement errors of both
burnout dimensions, and between the measurement errors of the three commitment
dimensions (i.e., between the errors of the indicators of each of the two mediator
variables). The existence of an additional variable that is not included in the model might
be responsible for such error-correlations (De Jonge et al., 2001), and is necessary in
order to explain the outcome variables more fully (Long, 1983; MacCallum, Wegener,
Uchino, & Fabrigar, 1993). In addition, it is important to note that items with identical
rating scales often have measurement errors that are correlated (Byrne, 1989). This
means that the fit of the proposed model can be improved if the measurement errors
among the items of the subscales are considered. Indeed, the revised model—including
the four covariations—shows a reasonable fit to the data, with only the NNFI slightly
below the criterion level of .90, v 2ð37Þ ¼ 95:55; GFI ¼ :92; RMSEA ¼ :08; CFI ¼ :91;
IFI ¼ :91, and NNFI ¼:87. In this revised model, all manifest variables load
significantly on the in-tended latent factors, except normative commitment (k ¼ :07;
n:s:)
As can be seen in Fig. 2, the paths from job demands to burnout and from burn-out to
absence duration were positive and significant (p < :05). This means that the
Fig. 2. Maximum likelihood estimates for the JD–R model, N ¼ 214. Note. All factor loadings and path
coefficients are significant at the p < :05 level, except the factor loading marked with an asterisk.
higher the job demands reported by the employees, the higher their level of burnout, and
the longer their absence duration (cf. Hypothesis 1). In addition, the path-coef-ficient
from job resources to commitment was highly positive and significant, whereas the
coefficient of the path from commitment to absence frequency was neg-ative and
significant. This means that more job resources coincided with higher levels of
commitment, which in turn resulted in less absence spells (cf. Hypothesis 2). Fur-
thermore, Fig. 2 shows that absence frequency and duration influence each other. More
specifically, our findings suggest that absence frequency influences absence du-ration (b
¼ :60; p < :01; R2 ¼ :35) and vice versa (b ¼ :15; p < :05; R2 ¼ :02). Even after
controlling for these effects, the JD–R model explained 4% of the variance in absence
duration and 4% of the variance in absence frequency.
In order to test the alternative hypothesis that job demands might also show a re-
lationship with commitment, and job resources with burnout, we included both di-agonal
paths in the model. However, adding both paths did not result in a significant
improvement of the fit between model and data, Dv 2ð2Þ ¼ :87, n.s. More-over, the
coefficients of both paths were non-significant (job demands ! commit-ment: b ¼ :04; t
¼ :23, n.s.; job resources ! burnout: b ¼ :28; t ¼ 1:72, n.s.). In a second alternative
model, we included the paths from burnout to absence fre-quency and from commitment
to absence duration in order to test the hypothesis that burnout and job commitment
influence both absence measures. These two ad-ditional paths also did not increase
model fit (Dv2ð2Þ ¼ 4:02, n.s.) and their coeffi-
cients failed to reach significance as well (burnout ! absence frequency: b ¼
:07; t ¼ :61, n.s.; commitment ! absence duration: b ¼ :14; t ¼ 1:49, n.s.). The third
alternative hypothesis to be tested was whether the working conditions would have direct
effects on the absenteeism measures. Unfortunately, adding direct paths from job
demands and job resources to the absenteeism measures, led to identification problems
so that no model parameters could be estimated. Generally speaking, identification
problems occur when the specified model is very poor (Bentler & Chou, 1987).
Finally, we examined the correlations between the demographics and the model
variables. It turned out that only organizational tenure and age showed significant re-
lationships. Specifically, organizational tenure and age were positively related to the
latent variable job demands (.24 and .18, respectively) and negatively to the latent var-
iable job resources ().22 and ).31). This suggests that those with more work experience
and higher age reported more job demands and less resources. In addition, age was
positively related to absence duration, and organizational tenure was positively related to
each of the three commitment indicators (affective .26; continuance .47; normative
.23). Thus, older employees were longer absent, and those with most work experience
showed strongest commitment to the organization. Inclusion of these relationships in the
final model did not significantly affect the structural relationships. The fit of the model—
including the demographics—to the data was satisfactory, and even slightly better than
the model without the control variables, v 2ð51Þ ¼ 114:18; GFI ¼ :92; RMSEA ¼ :08;
CFI ¼ :93; IFI ¼ :93, and NNFI ¼ :89.
In sum, SEM-analyses supported the hypothesized mediating role of burnout in the
relationship between job demands and the total number of days absent, as well
as the mediating role of organizational commitment in the relationship between job

resources and frequency of absence spells. None of the alternative relationships be-tween
the variables included in the model were significant when added to the model, and the
structural relationships did not change when demographics were included in the model as
control variables.
4. Discussion
This study used the JD–R model (Demerouti et al., 2000, 2001) to examine how
different categories of working conditions—job demands and job resources—are re-lated
to future absence duration and frequency. The central hypothesis was that job demands
would be unique predictors of absence duration, through their impact on burnout, and
that job resources would be unique predictors of absence frequency, through their impact
on organizational commitment. Using absence data collected during the year following
the assessment of job demands and resources, results pro-vided strong support for the
hypothesized pattern of relations. Job demands (i.e. workload and problems with
reorganization) were indeed unique predictors of pro-duction workers levels of
exhaustion and cynicism (i.e. the core dimensions of burn-out), and indirectly of absence
duration (positive relationships). In contrast, job resources (i.e. job control and
participation in decision-making) were unique predic-tors of commitment (positive
relationship), and indirectly of absence frequency (neg-ative relationship). Alternative
models, including direct paths from job demands and job resources to the two di fferent
absenteeism measures, or including paths from job demands to commitment and from job
resources to burnout did not fit better to the data than the proposed model. Moreover,
none of the coefficients of the alternative paths was significant. Because the correlational
analysis revealed that all working characteristics (except one) were significantly related
to the two measures of absen-teeism, our findings suggest that burnout and commitment
mediate the relationship between job demands and resources on the one hand, and
absence duration and frequency on the other hand, respectively.
Thus, our theoretical framework (Demerouti et al., 2000, 2001) was successful in in-
tegrating two different processes responsible for two types of absenteeism. The first
process can best be described as a health impairment process starting with high job de-
mands, which lead to burnout and longer periods of absence, respectively. The second
process is motivational in nature, and starts with job resources. Employees who can
draw upon job resources such as job control and participation in decision-making might
be more motivated to do their job, feel stronger commitment to their organiza-tion, and
report themselves less often sick than their counterparts. These findings inte-grate and
expand previous studies, in which moderate support was found for the idea that
employees who experience job stress are longer absent (e.g., Firth & Britton, 1989;
Saxton, Phillips, & Blakeney, 1991) and for the notion that employees low in job sat-
isfaction and organizational commitment are absent more frequently (e.g., Cohen, 1991;
Farrell & Stamm, 1988). Our findings show a unique pattern of relationships that is
consistent with Johns s (1997) observation that employees may be absent because
they temporarily do not want to work due to demoralizing or aversive work circum-
stances ( withdrawal hypothesis), or because they are unable to work as they are stressed
by their work situation ( stress hypothesis), or both. Furthermore, the model of
absenteeism tested in our study is one of the few that incorporates individually as-sessed
job characteristics, stress-reactions and work-related attitudes for the explana-tion of
different forms of absenteeism. Results support the notion that absenteeism is not a
unitary concept since different processes seem to cause different aspects of absen-
teeism. Our findings are all the more convincing since our frequency measure may also
have included absence due to involuntary factors such as sickness, and our duration
measure may have included voluntary, or avoidable, absence. Indeed, our correlational
analysis showed that the two absence measures share 20% of their variance. Neverthe-
less, we agree with Thomson et al. (2000) that future studies may profit from a better
discrimination between absence frequency and duration, for example by examining
certified and non-certified absence.
One unexpected finding was that, in contrast to affective and continuance commit-
ment, normative commitment did not load significantly on the latent factor organi-
zational commitment. Because this latent commitment factor had a negative e ffect on
absence frequency, this means that employees more often reported themselves sick in
case they felt more affective attachment to the organization, and perceived relatively few
alternatives to their job (cf. Gellatly, 1995; Johns & Nicholson, 1982). Employ-ees
loyalty to the organization (i.e. normative commitment) did not help to explain the
frequency of absence spells. It is unclear why this type of commitment did not perform
as predicted. Additional confirmatory factor analyses produced a clear three-factor
solution, and showed that—at the item level—all commitment items loaded on the
intended factors, and explained together 71% of the variance. Most probably, normative
commitment has different predictors and outcomes than the two other commitment
dimensions (see Johns, 1997).
Our study was restricted to the examination of two specific job demands and two
specific resources. At the heart of Demerouti et al. s (2001) JD–R model lies the as-
sumption that whereas every organization may have its own specific characteristics,
these factors can still be classified in two general categories (i.e., job demands and job
resources). Future studies should examine a broader range of demands and re-sources,
potentially related to absenteeism in a similar way.
The present study has several limitations. First, the measurement of the work char-
acteristics was based solely on self-reports, which increases the possibility that the re-
lationships between job demands and resources on the one hand, and burnout and
commitment on the other hand might be due to common method variance. However, the
differentiated pattern of relationships and the consistency of our findings with the-ory-
rooted hypotheses suggest that the single method bias is not a major drawback of this
study. A second limitation is that we tested our model among a specific group of
professionals, namely employees from a nutrition production company who work within
a specific constellation of working conditions. This calls under question the ex-ternal
validity of our findings. However, we believe that the current findings are not un-ique for
production workers, since evidence for each of the relationships in our model has been
found in studies among employees in a wide range of occupations.
Despite these limitations, the present findings may have important implications for
organizational practice and for individuals vocational behavior. First and fore-most, our
study shows that different absence measures are the result of two di fferent processes.
This underlines the importance of a systematical distinction between ab-sence duration
and absence frequency by human resource managers. Results clearly suggest that, in
order to decrease the duration or frequency of absenteeism, specific countermeasures
have to be taken regarding the working environment. Specifically, in order to reduce or
prevent burnout and consequently absence duration, specific job demands (in the present
study: workload and problems with reorganization) should be reduced or optimized. In
addition, in order to increase commitment and lower ab-sence frequency, the availability
of job resources (in this study: job control and par-ticipation in decision-making) should
be considered. Schaufeli & Enzmann (1998) have described several interventions at the
organizational level that can be used to attain this, including job redesign, job coaching,
and organizational development programs. However, it may be easier to influence
absence frequency with manage-ment tools than absence duration, since reducing
workload or avoiding reorganiza-tion (e.g., downsizing) processes might be di fficult to
realize in some cases, whereas the provision of job control and the increase of employee
participation might be easier to achieve through job (re) design approaches.
References
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of a ffective, continuance, and normative
commitment to the organization. Journal of Occupational Psychology, 63, 1–18.
Antonovski, A. (1987). Unraveling the mystery of health. How people manage stress and stay well. San
Francisco: Jossey-Bass.
Arbuckle, J. L. (1997). Amos users guide version 3.6. Chicago: Smallwaters.
Beehr, T. A. (1995). Psychological stress in the workplace. New York: Routledge.
Bentler, P. M. (1990). Comparative fit indexes in structural equation models. Psychological Bulletin, 107, 238–
246.
Bentler, P. M., & Chou, C. P. (1987). Practical issues in structural equation modeling. Sociological Methods and
Research, 16, 78–117.
Brehm, J. W. (1966). A psychological theory of reactance. New York: Academic Press.
Browne, M. W., & Cudeck, R. (1993). Single sample cross-validation indices for covariance structures.
Multivariate Behavioral Research, 24, 445–455.
Byrne, B. M. (1989). A primer of LISREL: Basic applications and programming for confirmatory factor
analysis models. New York: Springer-Verlag.
Chadwick-Jones, J. K., Nicholson, N., & Brown, C. (1982). Social psychology of absenteeism. New York:
Praeger.
Cohen, A. (1991). Career stage as a moderator of the relationships between organizational commitment and its
outcomes: A meta-analysis. Journal of Occupational Psychology, 64, 253–268.
De Gilder, D., Van den Heuvel, H., & Ellemers, N. (1997). Het 3-componenten model van commitment [The
three component model of organizational commitment]. Gedrag & Organisatie, 10, 95–105.
De Jonge, J., Bosma, H., Peter, R., & Siegrist, J. (2000). Job strain, e ffort-reward imbalance and employee
well-being: A large-scale cross-sectional study. Social Science and Medicine, 50, 1317–1327.
De Jonge, J., Dormann, C., Janssen, P. P. M., Dollard, M. F., Landeweerd, J. A., & Nijhuis, F. J. N. (2001).
Testing reciprocal relationships between job characteristics and psychological well-being: A cross-lagged
structural equation model. Journal of Occupational and Organizational Psychology, 74, 29–46.
Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2000). A model of burnout and life
satisfaction among nurses. Journal of Advanced Nursing, 32, 454–464.
Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands–resources model of
burnout. Journal of Applied Psychology, 86, 499–512.
Dwyer, D. J., & Ganster, D. C. (1991). The e ffects of job demands and control on employee attendance and
satisfaction. Journal of Organizational Behavior, 12, 595–608.
Erickson, R. J., Nichols, L., & Ritter, C. (2000). Family influences on absenteeism: Testing an expanded
process model. Journal of Vocational Behavior, 57, 246–272.
Farrell, D., & Stamm, C. L. (1988). Meta-analysis of the correlates of employee absence. Human Relations, 41,
211–227.
Firth, H., & Britton, P. (1989). Burnout , absence and turnover amongst British nursing sta ff. Journal of
Occupational Psychology, 62, 55–59.
Furda, J. (1995). Werk, persoon en welzijn: Een toets van het JD-C model [Work, personality, and well-being:
A test of the JD-C model]. Unpublished doctoral dissertation, Utrecht University, The Netherlands.
Gellatly, I. R. (1995). Individual and group determinants of employee absenteeism: Test of a causal model.
Journal of Organizational Behavior, 16, 469–485.
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory.
Organizational Behavior and Human Performance, 16, 250–279.
Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley.
Hammer, T. V., & Landau, J. (1981). Methodological issues in the use of absence data. Journal of Applied
Psychology, 66, 574–581.
Hensing, G., Alexanderson, K., Alleback, P., & Bjurulf, P. (1998). How to measure sickness absence?
Literature review and suggestion of five basic measures. Scandinavian Journal of Social Medicine, 26, 133–
144.
Hoyle, R. H. (1995). The structural equation modeling approach: Basic concepts and fundamental issues.
In R. H. Hoyle (Ed.), Structural equation modeling, concepts, issues, and applications (pp. 1–15).
Thousand Oaks, CA: Sage.
Jamal, M. (1984). Job stress and job performance controversy: An empirical assessment. Organizational
Behavior and Human Performance, 33, 1–21.
Johns, G. (1997). Contemporary research on absence from work: Correlates, causes and consequences.
International Review of Industrial and Organizational Psychology, 12, 115–173.
Johns, G., & Nicholson, N. (1982). The meanings of absence: New strategies for theory and research.
Research in Organizational Behavior, 4, 127–173.
Kahn, R. L., & Byosiere, P. (1992). Stress in organizations. In M. D. Dunnete & L. M. Hough (Eds.),
Handbook of industrial and organizational psychology (Vol. 3, (2nd ed., pp. 537–650). Palo Alto, CA:
Consulting Psychological Press.
Karasek, R. A. (1985). Job content questionnaire. Los Angeles, CA: University of Southern California.
Kendall, M. G., & Stuart, A. (1958). The advanced theory of statistics. New York: Hafner.
Koeske, G. F., & Koeske, R. D. (1989). Construct validity of the Maslach Burnout Inventory: A critical review
and reconceptualization. Journal of Applied Behavioral Science, 25, 131–144.
Kohler, S. S., & Mathieu, J. E. (1993). Individual characteristics, work perceptions, and a ffective reactions
influences on differentiated absence criteria. Journal of Organizational Behavior, 14, 515–530.
Lawson, D. A., & O Brien, R. M. (1994). Behavioral and self-report measures of sta ff burnout in development
disabilities. Journal of Organizational Behavior Management, 14, 37–54.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the three dimensions of
job burnout. Journal of Applied Psychology, 81, 123–133.
Leiter, M. P. (1993). Burnout as a developmental process: Consideration of models. In W. B. Schaufeli, C.
Maslach, & T. Marek (Eds.), Professional burnout: Recent developments in theory and research (pp.
237–250). Washington, DC: Taylor & Francis.
Long, J. S. (1983). Covariance structure models: An introduction to LISREL. Newbury Park, CA: Sage.
MacCallum, R. C., Wegener, D. T., Uchino, B. N., & Fabrigar, L. R. (1993). The problem of equivalent
models in applications of covariance structure analysis. Psychological Bulletin, 114, 185–199.
Marsh, H. W., Balla, J. R., & Hau, K. T. (1996). An evaluation of Incremental Fit Indices: A clarification of mathematical and empirical properties. In G. A.
Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modeling, issues and techniques (pp. 315–353). Mahwah, NJ: Erlbaum.
Maslach, C., Jackson, S. E., & Leiter, M. P. (1996). Maslach Burnout Inventory. Manual (3rd ed). Palo Alto, CA: Consulting Psychologists Press.
Mathieu, J. E., & Kohler, S. S. (1990). A test of the interactive e ffects of organizational commitment and job involvement on various types of absence.
Journal of Vocational Behavior, 36, 33–44.
Melamed, S., Ben-Avi, I., Luz, J., & Green, M. S. (1995). Objective and subjective work monotony: E ffects of job satisfaction, psychological distress, and
absenteeism in blue-collar workers. Journal of Applied Psychology, 80, 29–42.
Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualization of organizational commitment.
Human Resource Management Review, 1, 61–98.
Nicholson, N. (1993). Absence—there and back again. Journal of Organizational Behavior, 14, 288–290. Price, L., & Spence, S. H. (1994). Burnout
symptoms amongst drug and alcohol service employees: Gender differences in the interaction between work and home stressors. Anxiety, Stress and
Coping, 7, 67–84. Sagie, A. (1998). Employee absenteeism, organizational commitment, and job satisfaction: Another look.
Journal of Vocational Behavior, 52, 156–171.
Saxton, M. J., Phillips, J. S., & Blakeney, R. N. (1991). Antecedents and consequences of emotional exhaustion in the airline reservations service sector.
Human Relationships, 44, 583–595.
Schaufeli, W. B., & Enzmann, D. (1998). The burnout companion to study and practice: A critical analysis.
London: Taylor & Francis.
Schaufeli, W. B., Leiter, M. P., Maslach, C., & Jackson, S. E. (1996). The Maslach Burnout Inventory-General Survey. In C. Maslach, S. E. Jackson, & M.
P. Leiter (Eds.), Maslach Burnout Inventory, Manual (3rd ed). Palo Alto, CA: Consulting Psychologists Press.
Schaufeli, W. B., & Van Dierendonck, D. (2000). Maslach Burnout Inventory, Nederlandse versie: Handleiding [Maslach Burnout Inventory, Dutch version:
Manual]. Lisse, The Netherlands: Swets & Zeitlinger.
Shirom, A. (1989). Burnout in work organizations. In C. L. Cooper & I. Robertson (Eds.), International review of industrial and organizational psychology
(pp. 25–48). New York: Wiley.
Somers, M. J. (1995). Organizational commitment, turnover and absenteeism: An examination of direct and interaction e ffects. Journal of Organizational
Behavior, 16, 49–58.
Thomson, L., Griffiths, A., & Davison, S. (2000). Employee absence, age and tenure: a study of nonlinear e ffects and trivariate models. Work & Stress, 14,
16–34.
Van Veldhoven, M., & Meijman, T. (1994). Het meten van psychosociale arbeidsbelasting met een vragenlijst: de Vragenlijst Beleving en Beoordeling van
de Arbeid (VVBA). [Measurement of psychosocial job demands with a questionnaire: The questionnaire experience and evaluation of work (VBBA)].
Amsterdam: NIA.
Youngblood, S. A. (1984). Work, nonwork, and withdrawal. Journal of Applied Psychology, 69, 106–117.
STRATEGIC RESOURCES FOR SUSTAINABLE COMPETITIVE ADVANTAGE.
Article in International Journal of Advanced Research · March 2017

DOI: 10.21474/IJAR01/3507
CITATIONS READS
3 2,995
2 authors, including:
Jumadi Ridwan
Universitas Widya Mataram
6 PUBLICATIONS 5 CITATIONS
SEE PROFILE
Integrasi Pemasaran Pariwisata Dalam Mewujudkan Kunjungan Kembali Wisatawan Ke Daerah Istimewa Yogyakarta View project
All content following this page was uploaded by Jumadi Ridwan on 18 April 2017.

ISSN: 2320-5407 Int. J. Adv. Res. 5(3), 237-241
Journal Homepage: - www.journalijar.com
Article DOI: 10.21474/IJAR01/3507

DOI URL: http://dx.doi.org/10.21474/IJAR01/3507
RESEARCH ARTICLE
STRATEGIC RESOURCES FOR SUSTAINABLE COMPETITIVE ADVANTAGE.
*
Jumadi and Samsul Bakri.
Lecturer at the Economics Faculty of Widya Mataram Yogyakarta University.
……………………………………………………………………………………………………...
.
Manuscript Info Abstract
…………………….
………………………………………………………………
Manuscript History Companies are trying to position ourselves and try to have a sustainable
competitive advantage, by leveraging the strength to exploit
Received: 15 January 2017 opportunities and neutralize threats, and can avoid or fix vulnerabilities
Final Accepted: 13 February 2017
that exist in the company. The gain the competitive advantage is the
Published: March 2017
influenced by two factors, there are Internal environment and the
external environment. Internal environment are important elements in
Key words:- the company. The Analysis of internal environment beneficial for the
SCA (sustainable competitive company, especially the resource analysis. This analysis is done for
advantage), RBV (resources based internal resource can be strength of the company, but also can be a
view), Valuable, Rare, Not Inimitable, source of weaknesses. Terms of resources can be a force if these
Not Substitute, firm, EXSTRIM
resources are managed and empowered well, but on the contrary, if
internal resources are not managed and controlled properly it will be
the weaknesses of the company. Companies can achieve competitive
advantage through resource, resources the company has four criterion
the namely of Valuable, Rare, Low Inimitable Non- substitute (VRIN).
However, based on this study Authors to identified resource from VRIN
into EXSTRIM. The resources EXSTRIM that means of: Excellent,
Not Substitute, Not Trade, Rare, Not Inimitable, Mover & Modification
can be the strategic resource configuration in the company to creating a
sustainable competitive advantage (SCA).
Copy Right, IJAR, 2017, All rights reserved.
……………………………………………………………………………………………………...
.
Introduction:-
Strategies management become indispensable in managing fluctuating environmental conditions, the environment
can affect to the performance and company's goals. The company's strategy is formulated at three levels: operational
level, functional level and strategic level, in the context of industry competition, the main concern is the strategic
level. Each level of the strategy in general in order to create a competitive advantage that includes; cost leadership,
focus and strategy that differentiate it from other companies Porter (1991). A business manager should be able to
analyze the environment to understand opportunities and threats facing. Organizations need to build strength and
improve weaknesses; this process is not just a single step, but a process that includes various steps. One of them is
the scanning, the business situation, which, according to Albright (2004) scanning is an internal communication to
get information relevant external information on the Issues of potential influence on corporate decision-making
process in order to face the competition. The company may face competition when companies that have a
sustainable competitive advantage by harnessing the power companies to exploit opportunities and neutralize the
threat while avoiding or fixing a weakness or so-called analysis 'SWOT'. Porter's Five Force Model, Porter
conducted an analysis of the competitiveness of the industry, which is affected by the five forces that emphasizes
analysis of existing industrial structures. The Five forces such as; bargaining power of suppliers, bargaining power
with customers, the strength of the threat of new entrants, as well as the strength of the threat from competitors as
Corresponding Author:- Jumadi. 237
Address:- Lecturer at the Economics Faculty of Widya Mataram Yogyakarta University.
ISSN: 2320-5407 Int. J. Adv. Res. 5(3), 237-241
well as existing industry. While Porter framework has provided many useful insights both practitioners and
researchers to concentrate on external analysis 'OT' nevertheless still experiencing some significant problems. The
same thing Porter (1991), Valentin (2001) the SWOT analysis is an analytical tool for the environment that identify
strengths weakness opportunities and threat for the company. The business strategy is a way to master the wide
expertise and managing a company's profit with its surroundings. To realize that all companies should determine the
factors which might be in the face of competition once expected to be used to win the competition? One factor that
can be used to win the competition is a strategic resource company. Determine the combination of the company's
strategic resources is not easy, but if it is done well and seriously then the combination can be achieved with either.
The company's strategic resource that can be used to win the competition must be resources that have a competitive
advantage compared to the resources owned by other companies.
Understanding Resources:-
Many definitions of resources raised by the experts which is according to Barney (1991) Barney define resource
company covers all assets, capabilities, organizational processes, firm attributes, information, knowledge, which is
controlled by the company. According to Anderson (2007) the resource can be divided into internal and external
resources, external resources consist of relations with actors outside the company; customers and suppliers as well as
its reputation. Javad & Carlos (2010) introduced the idea that the company should be analyzed in terms of resources,
not only in terms of its products alone. According to resources based view (RBV), resources are defined as: anything
that can be regarded as a strength or weakness of a particular company. Resource companies as a factor in the
company and are owned or controlled by the company. The resource is generally defined as "all assets, capabilities,
organizational processes, firm attributes, information, and other knowledge that can be controlled by the company,
resource as an asset or an input to production tangible or intangible owned by an organization and can be controlled,
or in access to a semi-permanent.
Resources Advantage:-
According to Anderson (2007) special ability to create such varied resources management, entrepreneurial and
dynamic capability has been used to describe the ability to generate strategic resources. Furrer at.al (2008) indicates
that the resource-based grouping took the example of a bank that explains differences in corporate performance
better than the market grouping based products. However, market-based group still provides explanations of the
significance of the difference in performance in two of the three performance measures used. Meanwhile, according
to Grant (1991) by using the difference between the resources and capabilities, resources are tangible or intangible
assets, and capabilities are the ability generated from these assets. Anderson (2007) the resources can lead
management capabilities such as the ability of the market, production capabilities, and the ability of innovation. To
obtain optimal composition, the researchers suggest that the fit between the available resources of the company and
the requirements of the product-market activity in terms of concepts such as generic strategies should be carried out
it is appropriate revelation (Furrer at.al 2008). Resources Strategic Change, and Organizational Performance The
concept of strategic Resources by Kraatz & Zajac (2001) the resource that has an advantage can influence strategic
changes to the company, the company can emphasize how resources can hinder or may lead to organizational
learning in order to adapt. According to the opinion Banrey (1991) the latest research is a strategic resource is a
resource company that is difficult to replicate. Meanwhile, according to Chaharbaghi & Lynch; (1999) indicates that
strategic resources are not the only applications that efficiently and effectively to provide the level of high
competitiveness and ability to generate profits, but more important than that is how to make resources the mutual
synergy so really widened superior resources and strategies for the company in realizing a sustainable competitive
advantage. One way to the competitive advantage if the company has the dynamic strategic resource. The dynamic
the resources within the company the combination of resources within the organization based on the time dimension
will always be changing its composition, therefore, according to the authors' opinion that in order to realize the
resource made resources that strategic managers must be able to combine existing resources in the company.
According to Gro¨ßler (2007) that resources are located will affect the performance capability of the organization /
company (if not, they will not be located) there is a reciprocal process: the company's performance also affects
resource and capability. Gro¨ßler (2007) exemplifies the relationship between the ability to produce products with
high quality and sales revenue as a performance indicator. Producing high quality (among other factors) affects the
sales, which determines the sales revenue. Indicator of the understanding of strategic resources and its relation with
the ability to improve the performance of the company. As Gro¨ßler (2007) Barney (1991) found a strategic resource
must have at least four major characteristics, namely: Durable, has a positive long-term effect on the company's
strategic position; Non-tradable, it is impossible for competitors to easily afford; Non-applicable, it is impossible for
238
ISSN: 2320-5407 Int. J. Adv. Res. 5(3), 237-241
competitors to easily duplicate or copy them; Non substitutable, it is impossible for competitors to easily find a
replacement.
Meanwhile Barney (1991) that a strategic resource that can create competitive advantage and sustainability
performance is a resource that has the following criteria: Valuable (V): Resource-value, the resource company that
has a value that is able to set up the company in creating market opportunities and can reduce the challenges that
exist in the market. Rare (R): The resources are scarce (unique), whose existence is difficult to find the resources
and these resources are unique and not shared with other companies. Imperfect imitation (I): The resource that is
difficult to imitate, this resource is difficult to imitate by other companies. Non-substitutability (N): The resources
irreplaceable, resources in the other companies are difficult replacement to match.
Based on several definitions of strategic resources there is a difference between an expert opinions with another
expert opinion, the difference is due to the perspective and different educational backgrounds. Therefore, the author
tries to back describe strategic resources to get a thread or a new definition of the strategic resources. But before
describing the author or authors redefine about strategic Resources will provide an overview of some of the
definitions neighbor Strategic resources being addressed by some experts that will be displayed in the following
table.
Table :- Definition of Strategic Resources

No Author Definition
1 Banrey (1991) Business resources and difficult to imitate
2 Black and Boal (1994) Identify how the performance of the company and its
sustainability depends on the uniqueness, rarity.
3 Chaharbaghi & Lynch (1999) Strategic resource not only the application of efficient and
effective to provide the level of high competitiveness and
ability to generate profits, but more important than that is
how to make resources the mutual synergy that really
widened resource excellence and strategic for the company
in realizing SCA
4 Kraatz & Zajac (2001) Resources that have the advantage may affect strategic
change in the company, the company can emphasize how
resources can hinder or may lead to organizational learning
in order to adaptation
5 Anderson (2007) Special ability to create such varied resources management,
entrepreneurial and dynamic capability has been used to
describe the ability to generate strategic resources.
Based on Banrey (1991) the identification of the resources are: V: Valuable, R: Rare, I: Not Inimitable N: Not
Substitute, Black and Boal (1994) identification of the resources are VRIO that mean are V: Valuable, R: Rare, I:
Not Inimitable, O; Organization Orientation, they are differences existing definition. So the Author can the
conclusion of the strategic resources is "The unique resource obtained through a process that is adaptive and
proactive to spearhead (guns) as a means to win the competition and excellence continuous competitiveness. This
resource should have the first criteria; excellent namely resource superior or resources are: Excellent, Second: Not
Substitute are resources that not be replaced, Third: Not Trade resources that are not easily assessed by monetary or
not be traded, Fourth: rare is a scarce resource that is difficult to obtain, Fifth: Not Inimitable that the resources that
are difficult to imitate by other companies, Sixth: Mover & Modification is easy to be adapted to the dynamically
changing environment, means that in the past, now becoming a strategic resource, but not necessarily in the future
come can be a strategic resource back. Therefore, if the company to gain a sustainable competitive advantage the
company must make resource adaptive, proactive and reactive to the environment. Based on the above, the Authors
refer to as strategic resources are: EXSTRIM or the means of: Excellent, Not Substitute, Not Trade, Rare, Not
Inimitable, Mover & Modification.
Anderson (2007) The ways to realize the strategic resource contribution-based view of research taking the difference
in the configuration of resources between companies, if there is a description of how these strategic resources are
developed, the discussion usually focuses on direct investments as a way to obtain strategic resources. However,
differences in the configuration of resources between companies can also be explained by factors other than the
difference in making investment decisions. Usually the activity that takes place within the company in terms of
239
ISSN: 2320-5407 Int. J. Adv. Res. 5(3), 237-241
organizational processes and activities carried launching products in the market and can also develop resources in
order to obtain strategic resources. Thus, a strategic resource can be obtained in three ways:
jj Through direct investment;
kk Through the process of organization, and
ll Through the activities undertaken in the product markets.
Barney (1991) identifies the barriers that used to barrier process of imitation, the things that will be used when
analyzing barriers to imitation are:
xlii) The social complexity;
xliii) Ambiguity causal,
xliv) Conditions unique history
Social complexity refers to the intangible factors such as social and cultural relations, while the causal ambiguity
exists when the relationship between resources and competitive advantage is unknown. Conditions unique history
confirms that each company is a unique entity and that the possibility of acquiring the resources depending on the
time point, where and when to get those resources.
Strategic Concept advantage:-

According to Barney (2002) the company will gain a competitive advantage if the company's strategy and values
which not be done by the company's competitors. Strategic advantage, on the other hand, covers a broader
perspective in which this matter, so Chaharbaghi & Lynch (1999) include the following:
(CC) Provide the organization with a unique ability to develop strategic directions that create new opportunities
and future competitive environment.
(DD) Is a source of leadership resources in one or more of the products on the market.
Sustainable competitive advantage (SCA) is a process that needs to compete today without compromising the ability
of organizations to meet the requirements to compete in the future. Chaharbaghi & Lynch (1999) the process of
developing such an organization as a dynamic concept, which covers the basics of the following:
*
Sustainable Competitive Advantage is a conservation-oriented as the organization does not have unlimited
resources. Given these limitations, conscious management of existing resources is needed.
*
Sustainable Competitive Advantage is based on the needs of the organization in which economic activities
should be concerned with the needs of the market for goods and services resulting from economic activity.
*
Sustainable Competitive Advantage is future-oriented, placing itself outside day-to-day management of long-
term development resources to bring strategic advantages
*
A Strategic Advantage is the standard for corporate renewal, it is very dependent and influenced by, the
development and timely introduction to strategic resources.
Based on the results from the description above one of the factors that can be used to create a sustainable
competitive advantage SCA is a resources firm. Configuration was good between resources is expected to be
decisive to the sustainable competitive advantage.
Conclusion:-
Based on the discussions that have been submitted at the beginning to the end that was found disagreement among
experts on the categorizing of corporate resources is generally grouped into three categories, namely: physical,
tangible resources, intangible resources and human resources. A good combination between the three elements
contained in these resources is expected to establish a sustainable competitive advantage SCA. For the competitive
advantage for the company to be created in earnest with how to configure the use of the existing resources within the
company to the optimum. Appropriate combination or selection of the resources for the company is the main capital
in determining the level of a sustainable competitive advantage. The resources based view RBV has contributed
much to the formation of a strategic resource in promoting a sustainable competitive advantage. Resources that have
a sustainable competitive advantage if they meet the criterion VRIN namely; Valuable, Rare, Non Inimitable and
Non Substitute. Meanwhile, according to the Authors that strategic Resources should have the criterion: first
Excellent is the resource superior or resources excellent, Second: not Substitute is the resources that not be replaced,
Third: Not trade is the resources that are not easily assessed by monetary or not be traded, Fourth: Rare is the
resource that is difficult to obtain, Fifth: Not Inimitable is the resources that is difficult to imitate by other
companies, Sixth: Mover & Modification is easy to be adapted to the dynamically changing environment, means
240
ISSN: 2320-5407 Int. J. Adv. Res. 5(3), 237-241
that in the past, is now becoming a strategic resource, but not necessarily in the past that will come can be a strategic
resource back. Based on the above, the Authors refer to as a resource to resource strategic are the EXSTRIM
resources criterion.
Recommendation:-
The first performed by the expert company strategy is the identification of all the factors that may affect its business.
In this process, that important to know and to do is the related to internal activities involved in the business, this
includes all the systems of internal structure, strategies used, and the culture of the organization. The company's
activities will interact with the environment components outside the company such as customers, competitors, and
supplier; in addition it includes all parties (stakeholders) such as trade unions, media, and government. Competition
has entered a new era with the constant changes. The infrastructure problems that disrupt the organization can’t be
solved effectively without resource-based strategy is therefore necessary to incorporate the following principles:
*
Strategic resources must be regenerated more quickly than those used by other companies
*
The resources must be used more quickly excel
*
Basis of resources should be considered to be invested faster than the rate of acceptance of the environment
(adaptive) and proactive to realize resource from VRIN criterion to E X S T R I M to targeting in the sustainable
competitive advantage (SCA).
References:-
- Anderson Jim, 2007, ”A Holistic Approach to Acquisition of Strategic Resources”, Journal of European
Industrial Training Vol. 31 No. 8, 2007 pp 660-677
- Albright Kendra S., 2004,”Environmental Scanning: Radar for Success may”, the information Management
Journal.
- Barney, J, 1991,”Firm Resources and Sustainable Competitive”, Journal of Management, Vol. 17. No. 1 pp 99-
120.
- Barney, J. (2002), “Strategic management: from informed conversation to academic discipline”, Academy of
Management Executive, Vol. 16, pp. 53-7.
- Black Janice a. And Boal Kimberly B, 1994, Strategic Resources: Traits, Configurations and Paths to
Sustainable Competitive Advantage, Strategic Management Journal, Vol. 15, 131-148.
- Chaharbaghi Kazem, Lynch Richard, 1999, ”Sustainable Competitive Advantage: Towards a Dynamic
Resource-Based Strategy”, Management Decision, pp, 45–50
- Furrer, Sudharshan, Thomas, Alexandre, 2008, ”Resource configurations, generic Strategies, and Firm
Performance Exploring the Parallels Between Resource-Based and Competitive Strategy Theories in a New
Industry”, Journal of Strategy and Management Vol. 1 No. 1, pp. 15-40
- Grand R, M, 1991, ”The Resources Based Theory, of Competitive Adventage: Implication for Strategy
Formulation”, California Management Review.
- Gro¨ßler Andreas, 2007, ”A dynamic view on strategic resources and capabilities applied to an example from
the manufacturing strategy literature”, Journal of Manufacturing Technology Management Vol. 18 No. 3, pp.
250-266
- Javad Abadi, Carlos Cardon, 2010, ”Developing a Framework for Identification and Analysis of the Strategic
Resources and Capabilities in Supply Chains”, The 8th International Conference on Logistics and SCM
Research, BEM Bordeaux Management School.
- Kraatz Matthew S. and Zajac Edward J, 2001, ”How Organizational Resources Affect Strategic Change and
Performance in Turbulent Environments”: Theory and Evidence, Organization Science, Vol. 12, No. 5 pp. 632-
657
- Porter Michael E.,1991, ”Towards a Dynamic Theory of Strategy”, Strategic Management Journal, Vol. 12,
Special Issue: Fundamental Research Issues in Strategy and Economics (Winter, 1991), pp. 95-117
- Valentin E.K, 2001, “ Swot Analysis From Resources Based View”, Journal of Marketing Theory and Practice,
pp 54-69.
241
© 2015 Asian Economic and Social Society. All rights reserved
ISSN (P): 2309-8295, ISSN (E): 2225-4226 Volume 5, Issue 7,
2015, pp. 125-131
Journal of Asian Business Strategy
http://aessweb.com/journal-detail.php?id=5006
DOI: 10.18488/journal.1006/2015.5.7/1006.7.125.131
INNOVATION CAPABILITY: THE ROLE OF ISLAMIC WORK ETHICS
Muhammad Farrukh
SEGi University, Malaysia
Sonia Butt
Bahauddin Zakariya University, Pakistan
Shaheen Mansori
INTI International University, Malaysia
Article History: Abstract

The aim of this research is to investigate the association between
Received: 13 May 2015
Islamic work Ethics (IWE) and innovation capability. The association
Revised received: 16 June
2015 between Islamic work ethics (IWE) and innovation capability was
Accepted: 9 July 2015 assessed with the help of valid questionnaire. A total number of 150
Online available: 11 questionnaires were distributed out of which 120 were received back.
August 2015 The empirical results show a positive relationship between IWE and
innovation capability. This research is expected to underpin the insight
Keywords: into the determinants of innovation capability.
Islamic work ethics,
innovation capability
1. INTRODUCTION1
Every culture has a set of unique and influential conditions that affect people’s life; especially the
culture with religious background has more impact on people’s ethical behavior and thoughts (Porter,
2010). According to Quddus et al. (2009) people’s religious beliefs affect their understanding of
ethics, thus, people portray their beliefs and thoughts in practicing and understanding the ethics in
routine life. From last few decades research on ethics has become one of the tinted are of research for
the researchers and practitioners. A ton of research is carried out western countries pertaining to work
ethics and outcomes.
Most of studies related to work ethics are originated from USA and Europe. A few number of
researches are carried out in non-western work environment, and studies concerning to IWE and its
inference in work place are very rare (Rokhman, 2010; Kumar & Rose, 2010, Ahmad, 2011).
Most of the prior studies investigated the role of Islamic work ethics (IWE) on human resource
management practices such as, commitment, change management, employee job satisfaction and
intention to quit (Yousef, 2001; Rahman et al., 2006; Mohamed et al., 2010; Rokhman, 2010; Haroon
Corresponding author's
Name: Muhammad Farrukh
Email address: mfarrukhiqbal@hotmail.com
125
Journal of Asian Business Strategy, 5(7)2015: 125-131
et al., 2012), firm performance (Abbasi et al., 2012) and innovation (Kumar & Rose, 2010; Awan and
Akram, 2012; Abbasi et al., 2012).
The gap in literature regarding IWE implication in work place, becomes the base of this study,
Moreover, there are very few studies on the association between IWE and innovation capability.
Therefore, the aim of this research is to analyze the relationship between IWE and innovation
capability of employees. Taking into consideration of innovation capability for the successful
performance of organization, this research is expected to underpin the insight into the innovation
capability determinants which can help practitioners to enhance company performance.
1.1. Islamic work ethics

Arguments presented by Porter (2010) suggest that the meaning of work in any person’s life depends
upon the specified history and terms and conditions of his culture. It’s not the culture alone which
shapes the ethical mindset and behaviour; it is also accompanied by the religious background of the
people. It has been noticed by Quddus et al. (2009) that the moral standards and ethical perceptions of
people come from the views of their religion’s background and values. Hence, what people
understand from the term “ethics” and how they implement it in their routine and business life depicts
their religious thoughts, morals and beliefs. Islamic work ethics is not a new concept; it has been
originated for more than 1400 years. Its written evidence in Quran and been demonstrated by the
sayings and actions of the Prophet Muhammad (Ali & Al-Owaihan, 2008: Rice, 1999).
Ahmad (2011) has made his contribution in finding that Islamic work ethics (IWE hereafter) is a
multi-dimensional phenomenon including various dimensions related to economy, society and
morality. According to Jalil et al. (2010), Ethics is a condition of being respectful and the enactment
of good actions, also named as haya in Islamic terminology. Islamic ethics should be visibly depicted
from each and every part of his or her life not only business but personal also. It has been explained
by Ali and Al-Owaihan (2008) that the involvement and participation of the employees who believe is
determined and affected by IWE at the workplace. Business ethics defined by (Bassiouni, 1993, cited
in Wilson, 2006) from Islamic perspective is a relationship of honesty and trust between employers
and employees that indicates their spiritual equality before GOD regardless of their materially
inequality and the certitude that they are all brothers and sisters. Referring to Ali (2005), the IWE
value comprises of four basic principles. The principles are competition, efforts to achieve,
transparency, and sense of responsibility towards the moral conduct. All of these principles together
formulate a social contract and the focal point is achieving goals of the organization (Ali, 2005). The
Islamic work ethics has been gone through both qualitative and quantitative studies from much
different point of views. Previous researches investigated Islamic work ethics from human resource
management perspective such as organizational commitment (Yousef, 2001; Othman et al., 2004;
Rahman et al., 2006) and job satisfaction (Rokhman, 2010). Only a few researchers focused on the
relationship of Islamic work ethics and innovation capability.
According to Alhyasat (2012) there are two types of Islamic work ethics dimensions, one type of
dimension has significant association with organizational citizenship behaviour while other don’t have
any such association. Former type of dimension of Islamic work ethics includes affection and
tolerance with employees and clients, compliance to the people in power and honour job ethics, and
the later includes professionalism and skills, guiding and counselling other Muslims, right use of
authority, equality and honesty, uprightness and employee synergy. Yousef (2001) has searched out
that there is a positive relationship between Islamic work ethics, engagement and employee sensitivity
toward change.
Contrary to this, employee engagement acts as a liaison between Islamic work ethics and employee
sensitivity toward change. It has been proved from the results of Haroon et al., (2012) study that there
is a positive impact of Islamic work ethics on the nurses working in private hospitals regarding job
satisfaction. Similarly another study has reported that commitment can be achieved through the
Islamic management ethics (Salem & Agil, 2012). Kumar and Rose (2010) have found a positive
correlation between Islamic work ethics and innovation capability. Similarly, Abbasi et al. (2012)
126
reported that learning, innovation and organizational performance can be positively influenced by
Islamic work ethics. It has been proved through a research conducted on public sector organizations
by Awan and Akram (2012) that Islamic work ethics has a dominant favourable relation with potential
of innovation and in addition this relationship is supported through knowledge sharing. Very few
studies have been conducted regarding the concept and implementation of Islamic work ethics on sole
and organizational process and its consequences, the study under consideration is supposed to make
important addition to the studies regarding the topic.
1.2. Innovation and innovation capability

Van (1986) defines innovation as not only developing but also implementing of the novel
ideas/practices. According to Oslo (2005) innovation is nothing other than implementing new and
significant improvements in product (good or service), marketing strategies, the business operations,
and managing relations outside the organization. Innovation is one the fundamental factors to create
distinctive competitive advantage for organizations (Calantone et al., 2002; Knight & Cavusgil, 2004;
Günday et al., 2009). Drucker (1954) advocated that innovative capability can help an organization to
get competitive advantage.
The above mentioned definitions of innovations lead us to four basic elements of innovation which
are new concepts, people, processes and institutional context. Innovation is a strategic process in
which new ideas, processes, products, or services are generated, accepted, and implemented
(Calantone et al., 2002). Going towards innovation capability, Burgelman et al. (2004) has defined
that innovation capability is the characteristics of any organization which support and facilitate the
innovation strategies of the organization. While according to Neely and Hii (1998) innovative
capacity is the potential or degree of any firm, region or a nation of generating innovative outputs
which may be product, process, marketing idea or workplace ethics.
The more the potential of producing innovative outputs the more will be the innovative capacity.
Lawson and Samson (2001) have given a slightly modified definition of innovation capability which
is the ability of any firm how continuously and effectively transforms its generation of ideas into
implementation. It is the conversion of concepts and thoughts into new outputs, processes and
systems for the betterment of the firm and its stakeholders. Innovation capability is a value creating
phenomenon in different areas of the organization such as offering new products either tangible or
intangible, being more responsive and willingness to changing trends and policies, exploiting new
ideas, being highly learning organization, and increasing the level of competition in a highly
competitive and dynamic business atmosphere (Neely & Hii, 1998; Terziovski, 2007; Shan & Zhang,
2009). Various factors both internal and external are related to innovation capability and they affect it
(Bullinger et al. 2007; Yeşil & Kaya, 2012). Neely and Hii (1998) has reported three sets of factors
that are directly related to, and effect the innovative capability of the organization which are the
characteristics of the organization, management and environment. The purpose of this study is to find
out that whether there is any association between managerial Islamic work ethics values and
innovation capability or not.
1.3. Islamic work ethic and innovation capability

The relationship between work ethics and organization’s performance has been studied by many
researchers. Prior studies show that work ethics a positive impacton both individual and
organizational outcomes. Individual outcomes include job satisfaction, loyalty and organizational
brotherhood attitude (Koonmee et al., 2010; Sabir et al., 2012) while organizational outcome is the
performance of the firm and its position at product life cycle (Berrone et al., 2007; Donker et al.,
2008; Sabir et al. 2012).
Jalil et al. (2010) reported that organizational functioning and well-being can be achieved in a much
better way through ethical practices and implementation. Further argument leads that ethic-based
organizations have high level of job satisfaction and positive personnel well-beings. Employees are
more committed and better understand their responsibilities towards the organization. Luthans (2002)
has shown many references of the several studies whose results show that a significant positive
127
relationship exists between business ethics and outcomes of the organization. Organizational
innovativeness is the foremost requirement for any organizations to survive in today’s highly
predatory and dynamic environment (Neely & Hii, 1998; Calantone et al., 2002; Terziovski, 2007).
The capability of innovation can be polished through Islamic work ethics values of the organizations
(Abbasi et al., 2012; Awan & Akram, 2012; Kumar & Rose, 2010). Abbasi et al. (2012) has reported
that the integration of Islamic work ethics values into the organizational culture and values help
organization to get better return from their human resources. The Islamic work ethics values of the
managers act as a role model for the employees and encourage exerting extra effort, novelty,
trustworthiness, loyalty to duty and organizations, and good relations among the employees. Jalil et
al. (2010) has proposed that certain qualities such as integrity, honesty, solidarity, commitment and
responsiveness can be implicated and strengthened by sharing Islamic ethical practices throughout the
globe Moreover, some recent studies show a relationship between Islamic work ethics and
organizational innovation (Abbasi et al., 2012). It has been reported in public sector organizations by
Awan and Akram (2012) that capability of innovation is significantly and positively affected by
Islamic work ethics. Therefore, to get the more clear picture of the relationship between Islamic work
ethics and innovation capabilities the following hypothesis has been developed;
H: There is a significant relationship between Islamic work ethics values and innovation capabilities
2. METHODOLOGY
English version of the questionnaire was designed based on the previous scales used for the variables
understudy. A total number of 150 questionnaires were disseminated among the employees of
telecommunication companies in Lahore, Pakistan. 120 questionnaires were returned back by the
respondent and the response rate was 80%.all the items of scale were rated on five point likert scale.
Regression analysis was done by the help of SPSS 19.The target of the regression analysis is to
predict the dependent variable through the values of independent variable, in this study the Innovation
Capability is being predicted with the help of Islamic work ethics value
Table 1: Reliability and normality test

Variable N Number of Items Cronbach’s Alpha M SD SE
Islamic work ethics 120 17 0.80 4.33 0.29 0.04
Innovation capability 120 5 0.72 3.47 0.56 0.05
3. RESULTS AND DISCUSSIONS
The above table (table 1) shows the mean SD and SE in mean of the variables understudy. The mean
score of IWE is 4.33 SD is 0.29 and SE is 0.04. Mean for innovative capability is 3.47 SD and SE is
0.56 & 0.05 respectively.
Above table shows the results of the pearson’s correlation between the variables of the study.It is clear
from the above table that IWE is positivly correlated with the innovation capability (R=0.65, p <
0.001).
Table 2: Regression analysis of Islamic work ethics and innovation capability

Adjusted T value Significance
Variables Beta R square P value F
Islamic work ethics and
Innovation capability 0.65*** 0.4 8.30 0.000 68.68***
Note: *p < 0.05, **p< 0.01, ***p < 0.001
The results of regression analysis in table 3 show that Islamic work ethics has a positive impact on
innovative capability (As beta value is 0.65, p< 0.001). So, these results confirmed that “There is
positive relationship between Islamic work ethics and innovation capability.
128
4. DISCUSSION AND IMPLICATION
The purpose of this research was to investigate the relationship between Islamic work ethics and
innovative capability. The findings of the study showed a positive and significant relationship
between these two variables of the study. The significance of the study is twofold, first it added
knowledge in the domain of Islamic work ethics and innovative capability, and second the findings
will help the practitioners to appreciate the delegation of IWE in designing, formulating and
implementing the change programe in organizations. Ethics in the workplace is important and no one
can afford to avoid it whether work ethic is Islamic or other religions based.
4.1. Limitations and future recommendations

This study is subject to some limitations. This study was confined to only one city and a small sample
size. It is recommended to use a large sample to enhance the generalizability of the study.
It is also recommended to test a mediating or moderating relationship between IWE and innovation
capability. Furthermore, a cross national study using relatively a large sample is highly recommended
to compare and assess the generalizability of this research.
Funding: This study received no specific financial support.

Competing Interests: The authors declare that they have no conflict of interests.
Contributors/Acknowledgement: All authors participated equally in designing and estimation of current
research.
Views and opinions expressed in this study are the views and opinions of the authors, Journal of Asian Business
Strategy shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising
out of the use of the content.
References
Abbasi, A. S., Ghulam, M. M., & Muzammil, H. (2012). Islamic work ethics: How they affect
organizational learning, innovation and performance. Actual Problems of Economics, 138(12),
471-280.
Ahmad, M. S. (2011). Work ethics: An Islamic prospective. International Journal of Human Science,
8(1), 851-859.
Alhyasat, K. M. K. (2012). The role of Islamic work ethics in developing organizational citizenship
behaviour at the Jordanian press foundations. Journal of Islamic Marketing, 3(2), 139-154.
Ali, A. J., & Al-Owaihan, A. (2008). Islamic work ethic: A critical review. Cross Cultural
Management: An International Journal, 15(1), 5-19.
Ali, J. A. (2005). Islamic perspectives on management and organization. UK: Edward Elgar
Publishing.
Awan, K. Z., & Akram, M. (2012). The relationship between Islamic work ethics and innovation
capability and knowledge sharing plays moderation role. International Journal of Economics
and Management Sciences, 1(8), 34-48.
Bassiouni, M. C. (1993). Business ethics in Islam. In: Paul M. Minus, editors. In The Ethics of
Business in a Global Economy. Dordrecht: Kluwer Academic, 117-122.
Berrone, P., Surroca, J., & Tribo, J. A. (2007). Corporate ethical identity as determinant of firm
performance: A test of the mediating role of stakeholder satisfaction. Journal of Business Ethics,
76(1), 35-53.
Bullinger, H. J., Bannert, M., & Brunswicker, S. (2007). Managing innovation capability in SMEs.
Tech Monitor, Special Feature: Innovation & KM by SMEs, May-June, 17-27.
Burgelman, R. A., Christensen, C. M., & Wheelwright, S. C. (2004). Strategic management of
technology and innovation. Boston: McGraw-Hill.
Calantone, R. J., Cavusgil, S. T., & Zhao, Y. (2002). Tacit knowledge transfer and firm innovation
capability. Journal of Business Industrial Marketing, 18(1), 6-21.
Donker, H., Poff, D. & Zahir, S. (2008). Corporate values, Codes of ethics and firm performance: A
Look at the Canadian context. Journal of Business Ethics, 82, 527-537.
129
Drucker, P. F. (1954). The practice of management. New York: Harper and Row Publishers.
Günday, G., Ulusoy, G., Kılıç, K., & Alpkan, L. (2009). Effects of innovation types on firm
performance. International Journal of Production Economics, 133(2), 662-676.
Haroon, M., Zaman, H. M. F., & Rehman, W. (2012). The relationship between Islamic work ethics
and job satisfaction in healthcare sector of Pakistan. International Journal of Contemporary
Business Studies, 3(5), 6-12.
Jalil, A., Azam, F. & Rahman, M. K. (2010). Implementation mechanism of ethics in business
organizations. International Business Research, 3(4), 45-151.
Knight, G. A., & Cavusgil, S. T. (2004). Innovation, organizational capabilities, and the born-global
firm. Journal of International Business Studies, 35, 124-141.
Koonmee, K., Singhapakdi, A., Virakul, B., & Lee, D. J. (2010). Ethics institutionalization, quality of
work life, and employee job-related outcomes: A survey of human resource managers in
Thailand. Journal of Business Research, 63(1), 20-26.
Kumar, N. R., & Rose, R. C. (2010). Examining the link between Islamic work ethic and innovation
capability. Journal of Management Development, 29(1), 79-93.
Lawson, B., & Samson, D. (2001). Developing innovation capability in Organisations: A dynamic
capabilities approach. International Journal of Innovation Management, 5(3), 377-400.
Luthans, F. (2002). Organizational behaviour. Boston: McGraw-Hill.
Mohamed, N., Karim, N. S. A., & Hussein, R. (2010). Linking Islamic work ethic to computer use
ethics, job satisfaction and Organizational commitment in Malaysia. Journal of Business
Systems, Governance and Ethics, 5(1), 13-23.
Neely, A., & Hii, J. (1998). Innovation and business performance: A literature review. The judge
institute of management studies university of Cambridge, Retrieved from
http://ecsocman.hse.ru/data/696/521/1221/litreview_innov1.pdf.
Oslo, M. (2005). The measurement of scientific and technological activities: Guidelines for collecting
and interpreting innovation data, 3rd ed. Organization for economic co-operation and
development statistical office of the European communities.
Othman, A. S. S., Rahman, N. M. A., Malek, Z. A., & Osman, A. R. (2004). Organizational
Commitment and work ethics: An empirical assessment in a Malaysian context. Retrieved from
http://faculty.kfupm.edu.sa/coe/sadiq/proceedings/SCAC2004/30.ASC041.EN.Othman.
Organizational%20Commitment%20and%20Work%20E%20_1_.pdf.
Porter, G. (2010). Work ethic and ethical work: Distortions in the American dream. Journal of
Business Ethics, 96(4), 535-550.
Quddus, M., Bailey, III. H., & White, L. R. (2009). Business ethics: Perspectives from Judaic,
Christian, and Islamic scriptures. Journal of Management Spirituality Religion, 6(4), 323-334.
Rahman, N. M., Muhamad, N., & Othman, A. S. (2006). The relationship between Islamic work ethics
and Organizational commitment: A case analysis. Malaysian Management Review, 41(1), 25-35.
Rice, G. (1999). Islamic work ethic and the implication for the business. Journal of Business Ethic,
18, 345-358.
Rokhman, W. (2010). The effect of Islamic work ethics on work outcomes. Electronic Journal of
Business Ethics and Organization Studies, 15(1), 21-27.
Sabir, M. S., Iqbal, J. J., Rehman, K. U., Shah, K. A., & Yameen, M. (2012). Impact of corporate
ethical values on ethical leadership and employee performance. International Journal of
Business and Social Science, 3(2), 163-171.
Salem, Z. O., & Agil, S. O. S. (2012). The effects of Islamic management ethics on organizational
commitment of employees in Libyan public banks. Australian Journal of Basic and Applied
Sciences, 6(7), 260-270.
Shan, W., & Zhang, Q. (2009). Extension theory and its application in evaluation of independent
innovation capability. Kybernetes, 38(3/4), 457-467.
Terziovski, M. (2007). Building innovation capability in organizations: An international cross-case
perspective. Imperial College Press, London.
Van-de, V. A. H. (1986). Central problems in the management of innovation. Management Science,
32, 509-607.
Wilson, R. (2006). Islam and business. Thunderbird International Business Review, 48(1), 109-123.
130
Yeşil, S., & Kaya, A. (2012). Exploring the link between innovation capability and financial performance. International
Journal of information Technology and Business Management, 6(1), 11-25.
Yousef, D. (2001). Islamic work ethic: A moderator between organizational commitment and job satisfaction in a cross-
cultural context. Personnel Review, 30(2), 52-69.
131
Comparing theories to explain e-commerce adoption
Article in Journal of Business Research · March 2011

DOI: 10.1016/j.jbusres.2009.11.015 · Source: RePEc
CITATIONS READS
114 7,378
3 authors:
Elizabeth E. Grandón Suzanne Altobello

University of Bío-Bío University of North Carolina at Pembroke
48 PUBLICATIONS 1,436 CITATIONS 24 PUBLICATIONS 856 CITATIONS
Peter P. Mykytyn
Southern Illinois University Carbondale
SEE PROFILE
All content following this page was uploaded by Elizabeth E. Grandón on 05 November 2017.

Journal of Business Research 64 (2011) 292–298
Journal of Business Research
Comparing theories to explain e-commerce adoption

a, b b
Elizabeth E. Grandón , Suzanne A. Nasco , Peter P. Mykytyn Jr.
mm College of Business Administration, Universidad del Bío-Bío, Av. Collao 1202, Concepción, Chile
nn College of Business Administration, Southern Illinois University, Mailcode 4629, Carbondale, IL 62901-4629, United States
article info abstract
Article history: E-commerce is a strategy for rapid growth, especially by small and medium sized businesses (SMEs). However, the adoption
Received 1 March 2009 rate of e-commerce by SMEs in Latin America is still undersized. The authors compare the theory of planned behavior (TPB)
Received in revised form 1 September 2009 and the theory of reasoned action (TRA) using structural equation modeling to determine which is better at predicting e-
Accepted 1 November 2009 Available online 24
commerce adoption intentions among 210 SME managers/owners in Chile. Contrary to previous research with American
December 2009
respondents, the study does not find significant differences between the two theories. Thus, academics should select the more
parsimonious model (TRA) to study e-commerce adoption issues in developing countries.
Keywords:
E-commerce
Theory comparison © 2009 Elsevier Inc. All rights reserved.
Structural equation models
Chile
1. Introduction power than its counterpart, the TRA, due to the addition of the perceived
behavioral control construct (Madden et al., 1992; Taylor and Todd, 1995a;
The theory of planned behavior (TPB) (Ajzen, 1991) is an extension of Venkatesh et al., 2003). However, a lack of research results in uncertainty
the theory of reasoned action (TRA) (Fishbein and Ajzen, 1975; Ajzen and when considering whether these findings apply to other types of behavioral
Fishbein, 1980). Both theories hypothesize that an individual's intention to intentions and contexts. Thus, the purpose of this study is to corroborate the
perform the behavior in question is a determinant of that behavior. Intentions predictive validity of these two theories in a different research setting — the
are “indications of how hard people are willing to try, of how much of an intention to adopt e-commerce by managers/owners of SMEs in Chile.
effort they are planning to exert, in order to perform the behavior” (Ajzen,
1991, p. 181). Thus, the individual's attitude toward the behavior and the
subjective norm determine intention. Attitude toward the behavior refers to 2. Theoretical background
the degree to which a person has a favorable or unfavorable evaluation of the
behavior in question. Subjective norm refers to the perceived social pressure Many researchers publishing in MIS journals (e.g., Lee et al., 2006;
to perform or not to perform the behavior. Pavlou and Fygenson, 2006; Fu et al., 2006; Pee et al., 2008; Khalifa and
Shen, 2008; Nor and Pearson, 2008) use the TPB, and by default the TRA, to
In 1991, Ajzen (1991) added a new construct to the TPB: perceived explain behavioral intention to adopt information technology. However, only
behavioral control (PBC). He states that the original model (TRA) was unable a small portion of these studies focus on samples extracted from SMEs, such
to deal with behaviors over which people have incomplete volitional control. as Harrison et al. (1997); Riemenschneider and McKinney (2001–2002);
PBC reflects an individual's perceptions that personal and situational Riemenschneider et al. (2003) and Nasco et al. (2008). Harrison et al. (1997)
impediments to the performance of the behavior may exist. Ajzen (1991) use the TPB to predict small business executives' decisions to adopt
argues that the more favorable the attitude and subjective norm with respect to information technology to achieve a competitive advantage. Normal-ly,
a behavior, and the greater the perceived behavioral control, the stronger researchers follow a three-step process: completion of an elicitation study for
should be an individual's intention to perform the behavior under the preliminary TPB questionnaire, identification of computer-based systems
consideration. firms in the future will adopt, and customization of the final questionnaire sent
Numerous researchers (c.f., Sheppard et al., 1988; Madden et al., 1992; out to executives in SMEs. Results often find strong support for the TPB
Randall, 1994; Orbell et al., 1997) use the TPB and TRA to predict behavioral theory based on attitude, subjective norms, and perceived behavioral control
intentions and/or behavior in several academic disciplines. Many studies find regard-ing IT adoption.
that the TPB has a better explanatory
Also testing the TPB, Riemenschneider and McKinney (2001– 2002)

Corresponding author.
E-mail addresses: egrandon@ubiobio.cl (E.E. Grandón), snasco@cba.siu.edu analyze the beliefs of small business executives regarding the adoption of e-
(S.A. Nasco), mykytyn@cba.siu.edu (P.P. Mykytyn). commerce. They find that all of the component items of
0148-2963/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jbusres.2009.11.015
E.E. Grandón et al. / Journal of Business Research 64 (2011) 292–298 293
the normative and control beliefs differentiated between adopters and non- actual behavior. Ajzen (1991) summarizes 16 studies and concludes that the
adopters of e-commerce. In the behavioral beliefs (attitude) group, however, addition of PBC to the model leads to considerable improvements in the
only some items (e.g., e-commerce enhances the distribution of information, prediction of intention.
improves information accessibility, communication, and the speed with which Few studies in the MIS field directly compare the TPB to the TRA to
things get done) are found to differentiate adopters from non-adopters of e- empirically test model fit as these theories apply to the adoption of IT. Chang
commerce. (1998), Taylor and Todd (1995a), and Venkatesh et al. (2003) are some
Following a similar line of inquiry, Riemenschneider et al. (2003) exceptions. Chang (1998) examines whether 181 university students in Hong
examine the factors that influence web site adoption by SMEs. To gain an Kong would make unauthorized software copies, whereas Taylor and Todd
understanding of how small businesses handle IT adoption issues, they (1995a) examine whether business school students in a midsize university in
propose a combined model using the TPB and the technology acceptance the USA would use the computer resource center. Both studies find similar
model (TAM) of Davis (1989). They test individual models, partially results: the TPB, with the PBC construct, has a better model fit than the TRA.
integrated models, and fully integrated models through structural equation
modeling techniques and find that the fully combined model provides a better Venkatesh et al. (2003) also finds support for the PBC construct. The
fit than either the TPB or the TAM alone. However, the perceived behavioral authors compare the TPB and the TRA, among other models, to determine
control construct of the TPB and the perceived usefulness construct of the their powers in predicting behavioral intentions among individuals in the
TAM are not significant predictors of web adoption in the full model. workplace to adopt information systems, such as an online meeting manager,
database applications, portfolio analyzers, and proprietary accounting
Recently, Nasco et al. (2008) use the TPB to model intentions to adopt e- systems. Four organizations participated over a six-month period, with three
commerce among managers/owners of SMEs in Chile. This novel study points of measurement (post-training, one month after implementation, and
conducted in a developing country finds support for the theory. Using three month after implementation). The TPB explains 37%, 25%, and 21%
hierarchical regression, the authors show that the subjective norm and attitude respectively, at each time point, while the TRA explains slightly lower
constructs positively and significantly predict the intention to adopt e- variance amounts (30%, 26%, and 19% respectively). Thus, the inclusion of
commerce. Contrary to hypotheses, perceived behavioral control does not the PBC construct appears to add to the prediction of behavioral intentions in
significantly influence man-agers/owners´ intention to adopt the technology in the MIS environment.
question. As that study only uses regression and does not focus on a true However, none of these studies compares the TPB and the TRA in
theory comparison, stricter tests are necessary. Table 1 summarizes these international contexts Thus, researchers should determine whether the
studies. research in American settings regarding the ability to predict behavioral
intentions related to e-commerce adoption applies to different cultures of the
world. Moreover, no study examines the predictive ability of both theories in
a developing country. Thus, building on earlier work (Nasco et al., 2008), this
3. TPB vs. TRA project compares the TPB and TRA in predicting e-commerce adoption by
owners and managers of small businesses in a developing country, Chile.
Madden et al. (1992) test the predictive validity of the TRA and the TPB
for ten different behaviors (exercising regularly, getting a good night's sleep,
talking to a close friend, doing laundry, avoiding caffeine, going shopping
with a friend, renting a videocassette, taking vitamin supplements, listening to 3.1. Cultural differences
an album, and washing one's car), utilizing responses from 94 business
students. Results indicate that inclusion of PBC enhances the prediction of Hofstede's (1997) research on cultural dimensions provides a theoretical
behavioral intention and underpinning that may help to explain differences in e-
Table 1
E-commerce-related MIS adoption research in SMEs using the TPB.
Source Critical Adoption measure No. of Participants Industry

a
influencing SMEs
factors
s
Harrison et al. (1997) Attitude Adopt new IT 162 SMEs Business executives Service, non-profit, and manufacturing
s
Subjective norm (n b 200)
Perceived
behavioral
s
control
b
Riemenschneider and Attitude E-commerce 184 SMEs President/CIO, vice president, managers, IS Defense, agriculture, oil and gas, and
s
McKinney (2001–2002) Subjective norm adoption (n b 500) and marketing directors, and others manufacturing
Perceived
behavioral
s
control
s
Riemenschneider et al. Attitude Web site adoption 156 SMEs CEOs, COOs, vice presidents, IS managers, and Service/sales, government, retail, banking,
s
(2003) Subjective norm (web presence) (n b 500) general managers medical, and manufacturing
Perceived
behavioral
ns
control
Perceived
ns
usefulness
s
Nasco et al. (2008) Attitude E-commerce 212 SMEs Managers and owners Wholesale, retail, construction,
s
Subjective norm adoption (n b 200) manufacturing, transportation, and others
Perceived
behavioral
ns
control
s ns
: significant variable (p b 0.1); : non-significant variable (p ≥0.1).
(EE) n represents the maximum number of employees considered in the criteria to define a SME.
(FF) Partial significance (not all the items in the construct were significant).
294 E.E. Grandón et al. / Journal of Business Research 64 (2011) 292–298
commerce adoption between developed and developing countries. Surveying

employees from the IBM Multinational Corporation in 72 countries, including
the USA and Chile, Hofstede identifies four dimensions to distinguish among
different cultures: power distance, individualism, masculinity, and uncertainty
avoidance.
The USA scores substantially higher than Chile on individualism (91 vs.
23) and masculinity (63 vs. 28). Higher individualism, defined as “the interest
of the individual prevails over the interest of the group” (Hofstede, 1997, p.
50), indicates that a culture is more interested in individual goals and pursuits,
rather than communal needs of the larger social group. Hofstede defines
masculinity as pertaining to “societies in which social gender roles are clearly
distinct” (p. 82); higher values of masculinity suggest that traditional gender
roles and dominant societal values such as assertiveness, acquisition of
Fig. 1. Model 1 represents the TRA (full lines). Model 2 represents the standard TPB (full lines
money, and focus on material success are present in such a society.
plus dotted lines).
Chile, on the other hand, scores higher than the USA on power distance control (PBC) determine behavioral intention (BI). Thus, according to the
(63 vs. 40) and uncertainty avoidance (86 vs. 46). Hofstede defines power TPB, BI is a weighted function of A, SN, and PBC, i.e., BI = w 1A + w2SN +
distance as “the extent to which the less powerful members of institutions and w3PBC.
organizations within a country expect and accept that power is distributed This study tests two competing models representing the ante-cedents of
unequally” (Hofstede, 1997, p. 28). In a working environment, larger values intention. Fig. 1 shows Models 1 and 2: Model 1 (full lines) represents the
of power distance mean consider-able dependence of employees on bosses, TRA where the only determinants of intention are attitude and subjective
where employees are unlikely to approach and contradict their bosses directly. norm. Model 2 (full and dotted lines) incorporates PBC as the third factor
Finally, larger values on the uncertainty avoidance dimension, defined as “the influencing intention. Model 2 nests Model 1 when the path from PBC to
extent to which the members of a culture feel threatened by uncertain or intention is equal to 0.
unknown situations” (p. 113), reflect cultures whose members who behave in
ways that avoid or prevent feelings of uncertainty. 4. Research methodology
Differences in Hofstede's cultural dimensions between Chile and the USA 4.1. Subjects
may influence, to a certain extent, e-commerce adoption intentions. In this
regard, uncertainty avoidance and individualism may have the most direct Top managers/owners in this study are from a population of SMEs in
bearing on such perceptions. According to Hofstede's classification, the US Chile. The researchers identified a random sample of 1100 small businesses
culture may be more prone to risk taking and willingness to assume changes, chosen from various business directories in the capital city of Santiago, the
while Chile, on the other hand, exhibits a culture that is less prone to risk Bío-Bío, and the Ninth region of Chile. The Chilean Corporation to Promote
taking and, in general terms, may seek to avoid change. The incorporation of Production defines a small business as one that employs between 10 and 200
a new technology, such as e-commerce, brings structural changes and employees (CORFO, 1994). The sample firms receive a printed survey, which
redesign of organizations (Laudon and Laudon, 2004), which Chilean includes a cover letter, the questionnaire, and a pre-paid return envelope.
individuals may not be as comfortable with as their American counterparts. In Follow-up telephone calls to non-respondents are made two, four, and six
addition, the American culture is very individualistic, which may dictate the weeks after the surveys were mailed. Thirty-five firms did not receive the
willingness of US managers/owners of SMEs toward adopting an electronic survey instruments due to incorrect addresses. Over a 12-week period, the
business model, such as e-commerce. More collectivist cultures, such as researchers receive 228 surveys, representing a response rate of 20.27%.
Chile, may perceive e-commerce as an impersonal way to conduct business,
antithetical to the way a collectivistic culture perceives business relationships.
The TPB and TRA reflect with individual behaviors, but the theories are
applicable to SMEs by considering that the respondents are top
Many researchers apply the TPB and the TRA to American firms, with the managers/owners who are the primary decision makers in their respective
theories demonstrating good explanatory power (Taylor and Todd, 1995a; firms. Similar to Harrison et al.'s (1997) work, in this study, the researchers
Harrison et al., 1997; Riemenschneider et al., 2003). Thus, determining the carefully screened respondents to ensure that each person is a primary
extent to which these theories apply to a collectivist culture with high decision-maker in his/her firm. The researchers eliminate any response from a
uncertainty avoidance index and low individualism index is still an issue. firm from the final analysis if the firm does not meet the definition of SMEs, if
the respondent is not a top manager, or if the firm had e-commerce already in
place. As a result, the final analysis contains 212 firms.
3.2. Research model and determinants of the theories
The TPB (Ajzen, 1991) states that behavior (B) is a direct positive 4.2. Measures
function of behavioral intention (BI) and perceived behavioral control (PBC).
PBC influences behavior indirectly through intentions, as well as directly Measuring latent constructs of attitude, subjective norm, percep-tions of
when the person does not have complete control over that behavior and when behavioral control, and intention in the context of the TPB and TRA involves
the individual's perceptions of control are accurate (Madden et al., 1992). Fig. direct measures that represent manifest indicators. The items in the present
1 graphically shows the direct effect of PBC on behavior and its indirect instrument include appropriately modified items from the study by
effect through intentions. Behavior is a weighted function of intention and
Riemenschneider et al. (2003) to reflect the specific target behavior, the
PBC (B = w1BI + w2PBC). Mykytyn and Harrison (1993) indicate that intention to adopt e-commerce by managers/ owners of SME in Chile. To
“assuming nothing has arisen in the environment to cause a change in plans, a
measure the TPB, fiveitems measure attitude, four items
measure of intention should be the best predictor of behavior” (p. 17). One's
attitude (A), subjective norm (SN), and perceived behavioral measure subjective norm, three items measure perceived
behavioral control, and three items measure intention.
Measurement of all the items is on a 7-point Likert scale,
ranging from
*
strongly agree to (7) strongly disagree. Contact the first author for the 2
1998). Thus, the model fit is perfect (χ (0) = 0.00, p b 1.00). More important
original Spanish questionnaire or the English version of the items. than fit, the t-values associated with each indicator are statistically significant.
The researchers measured the subjective norm latent construct with four
5. Results observed indicators. The CFA for this construct produces satisfactory results
2
with χ (2) = 2.79, (p b 0.25) and Root Mean Square Error of Approximation
5.1. Measurement model (RMSEA) = 0.04. Other model fit indices also indicate quite strong values for
a well-fitting measurement model (GFI = 0.99, AGFI = 0.97; RMSEA = 0.04).
The researchers conducted confirmatory factor analyses (four separate The results of the CFA for the attitude construct are initially not
CFAs involving one for each construct) and Cronbach's alpha reliability acceptable. Two indicators (the second and fifth) from the measure-ment of
analyses prior to testing the competing TPB and TRA models to examine the this construct have high error variances. Coincidentally, both of these items
internal consistency of the items representing each construct. For the are reverse-worded, such that high numbers meant lower attitudes and the
hypothesized models with three indicators (intention and perceived behavioral meaning of reverse-worded items may be difficult for international
control), the models are saturated (the number of parameters is equal to the respondents to understand. Thus, the model for attitude now contains only 3
number of variances and covariances among the observed variables) (Byrne, indicators (A1, A3, and A4). As three indicators lead to a saturated solution,
2
the re-specified model has a perfect fit (χ (0) = 0.00, p b 1.00) and the t-
values associated with each item are statistically significant.
Table 2 Table 2 shows completely standardized loadings, reliability and variance

Results of CFA for intention, attitude, subjective norm, and PBC.
extracted for each indicator. Bold numbers indicate construct reliability and
Construct and indicators Completely Construct Variance variance extracted at the construct level.
standardized reliability (α) extracted
loadings and indicator 5.2. Structural models
reliability
a a
Intention 0.96 0.88 The research team used structural equation modeling (SEM) to test the
I1. My firm strongly intends to 0.90 0.81 0.19
validity of the competing models in this study by examining the relationships
incorporate e-commerce
between the theoretical constructs. Statistical model comparisons provide
within the next year
I2. We have certain plans to 0.98 0.96 0.04 contrasting model fit indices and allow for statistical tests between models,
2
incorporate e-commerce such as likelihood ratios (χ difference test) and Wald tests (Bollen, 1998).
in our organization within
the next year
I3. Our firm has a strong 0.94 0.88 0.12
commitment to incorporating 5.2.1. The full model (TPB)
e-commerce within the The first structural model examined is the full TPB model, with three
next year exogenous constructs and one endogenous construct. The model includes 13
a a
Attitude 0.88 0.71
observed indicators (10 for the exogenous constructs and 3 for the
A1. Incorporating e-commerce in 0.87 0.76 0.24
my firm within the next year endogenous construct). Fig. 2 presents the tested model with LISREL
would be good notations, showing path coefficients between each construct and its respective
A3. Incorporating e-commerce in 0.91 0.83 0.17 indicators (λ) and the error terms associated with each indicator (δ and ε).
my firm within the next year Gamma coefficients (γ) represent the paths from the exogenous constructs to
would be positive
the endogenous construct, which correspond to regression coefficients among
A4. Incorporating e-commerce in 0.73 0.53 0.47
my firm within the next year the constructs. Contact the first author for the covariance matrix used as SEM
would be effective input.
a a
Subjective norm 0.90 0.70
SN1. Most people who are 0.94 0.88 0.12 Within the initial SEM program, the Phi (PH) matrix is a symmetric
important to my firm think
matrix with all parameters freely estimated and the Theta Delta (TD) and
my firm should incorporate
e-commerce within the Theta Epsilon (TE) matrices are diagonal with all parameters
next year
SN2. Most people who influence 0.86 0.73 0.27
the behavior of my firm think
my firm should incorporate
e-commerce within the
next year
SN3. People whose opinions our 0.87 0.75 0.25
firm value would prefer our
firm to incorporate e-commerce
within the next year
SN4. Most firms that are important 0.67 0.45 0.55
to my firm have
adopted e-commerce
a a
Perceived behavioral control 0.70 0.57
PBC1. Incorporating e-commerce 0.87 0.76 0.24
in my firm within the
next year would be easy
in my firm within the
next year would be under
my firm's control
within the next year would
be simple to arrange Fig. 2. Initial theoretical structural model (SEM notation).
*
Construct reliability and variance extracted at the construct level respectively.
freely estimated. Exogenous variables correlate, which is standard practice in

SEM (Loehlin, 1998; Byrne, 1998). The Gamma (GA) matrix is a full matrix
with all parameters fixed to zero. Subsequently, gamma components (1,1),
(1,2), and (1,3) are free to be estimated. Lambda-X (LX) and Lambda-Y (LY)
matrices are set as full matrices with all parameters fixed to zero. The
researchers used the first indicator of each latent variable to set the unit of
measurement.
Output statistical goodness-of-fit measures show that the initial model is
2
not well-fitting: the chi-square value is significant, χ (59) = 198.81 p b 0.001,
and other indices are poor (RMSEA = 0.11, GFI = 0.87, AGFI = 0.79, and
2
χ /df = 3.37). Inspecting the error vari-ance matrix shows that the second
2
indicator PBC2 has high error variance and also has the lowest R value of
0.32. Therefore, the researchers dropped this item from the analysis and re-
estimate the model with 12 indicators.
2
After re-estimation, the results show a slight decrease in χ (196 with 48
degrees of freedom). The RMSEA, however, results in a higher value (0.12)
when compared with the RMSEA of the previous model with 13 indicators.
Fig. 4. The TRA model with path indicators (Model 1).
Other goodness-of-fit indices result in poorer fit (GFI = 0.86, and AGFI =
2
0.78, χ /df = 4.09). A closer look at indicator A4 shows that the variable has a
2
high error variance of 1.07, which contributes to the misfit of the model. following goodness-of-fit indices: χ (39) = 106.97, RMSEA = 0.1, GFI =
Therefore, a follow-up analysis did not include indicator A4. 0.91, and AGFI = 0.86. Fig. 4 shows the results for the TRA.
Table 3 shows the completely standardized (CS) structural coefficients and
Finally, the researchers respecify the model with 11 indicators. This model significance levels for both models. In both, attitude and subjective norm are
shows a better fit to the data; the chi-square value drops considerably to significant predictors of behavioral intention. However, the structural path
106.91 with 38 degrees of freedom (p b 0.001). Although this value is from PBC to intention is not significant (even though the fit of the model still
2
significant, the χ /df ratio is very good at 2.81. All of the other goodness-of-fit meets traditional thresholds) when the fuller model includes PBC.
statistics also support the revised model (RMSEA = 0.094, GFI = 0.91, AGFI
= 0.85, RMR = 0.03, CFI = 0.99, NNFI = 0.98, NFI = 0.98). LISREL output Table 4 shows a comparison of goodness-of-fit indices between the TPB
indicates that attitude, subjective norm, and perceived behavioral control in and the TRA. In general, the goodness-of-fit indices for the TRA model show
the revised structural model for the TPB explain 81% of the variance in a slightly better fit than those of the TPB model. No statistically significant
2
intention. Fig. 3 depicts the final revised model along with the standardized difference between the fit of the two models, as evidenced by the χ difference
paths. 2
test, χ (1) = 0.06 p N 0.05, exists. Thus, the TRA model should be retained as
this model explains the same amount of variance in intention with fewer
parameters to be estimated.
5.2.2. The reduced model (TRA)
The TRA does not include the PBC construct; thus this model is a reduced
model. According to Bentler (1992), nested models emerge when constraints
exist on a more general model, creating a reduced model. Thus, the TRA 6. Discussion and conclusions
model is a nested model within the TPB when the direct path between the
latent exogenous construct of PBC to the endogenous construct of intention is Previous studies (Madden et al., 1992; Taylor and Todd, 1995a; Chang,
set to zero. To compare the models, the researchers use both absolute and 1998) find that the TPB, which incorporates the PBC construct, provides a
2
parsimonious fit indices (PGFI, Normed chi-square, PNFI, and χ difference significant better fit than that of the TRA which does not incorporate the same
test). construct. In this study, even though both models fit the covariance matrix
2
To create the TRA model, constraints set the γ13 path to zero. Fitting this relatively well, results from a χ difference test reveal no significant
model to the overall covariance matrix results in the difference between the models. The results show the validity of the TRA over
the TPB as applied to the domain of adoption of e-commerce among
managers/owners of SMEs in Chile. Attitude and subjective norm are
significant predictors of
Table 3
Structural coefficients and significance levels for the fitted models.
Relationship TRA coefficients TPB coefficients

A →Intention 0.20 0.21
SN →Intention 0.73 0.71
PBC →Intention 0.01
= p b 0.01 (1.96).
Table 4
Fit indices for TPB and TRA models.
2 2
Model Χ RMSEA GFI AGFI PGFI χ /df PNFI
TPB 106.91 0.094 0.91 0.85 0.53 2.81 0.68
TRA 106.97 0.092 0.91 0.86 0.54 2.74 0.69
Fig. 3. Revised structural model for the TPB (Model 2).
intention, whereas PBC is not. Therefore, the model that is more parts of the world. One of the major contributions of this research is that the
parsimonious (in this case, the TRA) should be retained. In fact, both models study is the first research that has rigorously applied statistically sophisticated
explain the same amount of variance on intention (81%). Thus, US techniques to test the TPB and its variations to predict e-commerce adoption
researchers need to think carefully about the addition of constructs that may among managers/owners of SMEs in Chile. Results show that, using the TPB,
not necessarily reflect the international business world. the intention to predict the adoption of e-commerce among managers/owners
These results contradict previous findings regarding the compar-ison of SMEs in Chile is approximately 80%. Similarly, the TRA is as good as the
between TPB and TRA models (Taylor and Todd, 1995a; Chang, 1998), who TPB and could be even preferred, since the TRA explains the same amount of
found support for the TPB. In making model comparisons, differences in the variance on behavioral intention with less estimated parameters (and
contexts between the ways researchers conduct studies are important consequently, fewer questions to ask managers regarding adoption). Thus, the
considerations (Taylor and Todd, 1995a). Subjects in the Chang (1998) and present results illustrate that, regarding the adoption of e-commerce by
Taylor and Todd's (1995a) studies are college students who may have managers/owners of SMEs, the TRA could be an excellent and parsimonious
different perceptions about the adoption of IT than managers/owners of model to explain behavioral intention. Researchers may want to replicate this
SMEs. Perceptions of e-commerce may operate differently in school settings study in similar settings, for example, in other developing countries in South
than in workplace settings and the predicted target behavior here is different America that share some cultural characteristics with the Chilean one, to
from the other studies. corroborate these results. Researchers may also consider testing additional
theories to explain e-commerce adoption, (e.g., TAM) or add theories together
The present findings also do not show a positive linear relationship (e.g., TAM + TRA) to confirm which combination of theories explain better
between PBC and the intention to adopt e-commerce. This finding, however, the adoption of e-commerce among manages/owners of SMEs. Finally, other
is not too surprising, given the mixed results from previous studies regarding studies should investigate whether the antecedents to adoption intentions
the relationship between PBC and intention. Although Chang (1998) finds that change over time.
PBC is the strongest predictor of intention, Venkatesh et al. (2003) finds that
PBC is a significant predictor of intention only in some of the relationships,
and Riemenschneider et al. (2003) report that PBC is not a significant Future research regarding measurement issues could focus on determining
predictor of intention in any of their hypothesized models. The mixed findings the factor structure of the attitude component of the TPB and TRA. Results
could be due to the way the PBC is defined, which may cause inconclusive from reliability and validity analyses of the attitude construct of this study
results in the above mentioned studies, including this one. Armitage and suggest that two independent constructs could better capture the structure of
Conner (2001) and Terry and O'Learly (1995) have brought up this criticism. attitude. For example, as the two items dropped from the attitude construct
As defined, perceived behavioral control represents perceptions of ease or were negatively-worded, future research could consider positive and negative
difficulty of performing the behavior in question (Ajzen, 1991). These attitude toward the adoption of the behavior in question. Future research may
perceptions relate to opportunities and available resources to perform the examine the direct effect of positive and negative attitude on the intention to
behavior in question. adopt IT, particularly e-commerce.
Another possible explanation for the non-significant relationship between

PBC and the intention to adopt e-commerce among managers/owners of 6.2. Post hoc modifications
SMEs in Chile has roots in cultural differences among the countries in which
previous studies have taken place. Nasco et al. (2008) specifically bring up Post hoc analyses on the results of the structural component of the
Hofstede in their discussion which provides relevance to discuss the issue competing models proposed in this study create future research questions.
again here. The uncertainty avoidance dimension of Hofstede (1997) may be Byrne (1998) emphasizes that model respecification and re-estimation are
particularly important in the business context represented in this study. important for exploratory analysis rather than as a confirmatory one.
Americans exhibit a low uncertainty avoidance index, which means that they Modification indices (MI) may be important for determining potential model
are more prone to risk taking and to assume changes. Chileans, on the other respecification. Several possibilities exist. For example, a follow-up study
hand, exhibit a culture that is less prone to risk taking and may avoid changes. involving crossover effects (Taylor and Todd, 1995b) among normative and
The incorporation of e-commerce brings structural changes and redesign of attitudinal compo-nents of the TRA intention model may be important.
organizations; determining the required resources (financial, technological, Similar to previous work in social psychology and consumer research that
human, etc.) associ-ated with the implementation of e-commerce may be a allows for crossover effects in the TRA (e.g., Liska, 1984; Oliver and
difficult task. While American managers of SMEs may be comfortable having Bearden, 1985), Taylor and Todd (1995b) report a study to predict consumer
incomplete information to make the decision to adopt e-commerce, Chilean adoption intention using the TPB. They find that allowing for crossover
managers may feel overwhelmed with the uncertainty of not knowing the effects in the TPB results in improvements: crossover effects from normative
necessary resources to implement e-commerce, and therefore, may not be able beliefs to attitude are significant and improve model fit.
to determine the importance of having these resources available. This fact
may explain why PBC was not found to be a significant predictor of the
intention to adopt e-commerce among Chilean managers of SMEs in In the present study, an examination of modifications for the TPB reveal a
structural tests presented here and in hierarchical results presented in Nasco et high value of 21.48, suggesting a link between one indicator of subjective
al. (2008) and, to a certain extent, why the TRA model is more appropriate norm (SN2) to the attitude construct, which clearly represents a crossover
than the TPB when explaining e-commerce adoption in this same research effect. Therefore, future research could focus on this type of crossover effect
context. to develop an even stronger theory to predict behavioral intention in other
areas of knowledge. Similarly, an examination of modification indices for the
TRA model reveals the same pattern of crossover effects, thus, including a
link between some of the indicators of subjective norm to attitude may be
important for consistency with previous research. For instance, based on
6.1. Contributions to academic research and suggestions for future research previous studies (e.g., Vallerand et al., 1992), Chang (1998) gave support to
justify a causal link path linking subjective norm to attitude in the TPB.
Consequently, future research could focus on determining the impact of this
By using a sound theory such as the TPB, and by default the TRA, this type of crossover effects on the prediction of intention to adopt IT in a variety
study intends to fill in the research gap concerning the lack of strong of domains. In sum, researchers should continue to take
theoretical foundations in IT research involving SMEs in other
established Westernized organizational behavior and managerial theories and Laudon KC, Laudon JP. Management information systems. New Jersey: Prentice Hall; 2004.
examine their applicability to Latin American businesses. Additional goals are Lee SM, Kim I, Rhee S, Trimi S. The role of exogenous factors in technology acceptance: the
to take findings established here (i.e., the importance of the subjective norm case of object-oriented methodology. Information & Management 2006;43:469–80.
construct) and examine whether organizational communications based on the Liska A. A critical examination of the causal structure of the Fishbein/Ajzen attitude behavior
model. Social Psychology Quarterly 1984;47(1):61–74.
model result in higher levels of e-commerce adoption. For example, the Loehlin JC. Latent variable models. An introduction to factor, path, and structural analysis. NY:
government could use local businesses that have adopted e-commerce as an Lawrence Erlbaum Associates; 1998.
example to businesses that have not yet adopted, focusing on the normative Madden TJ, Ellen PS, Ajzen I. A comparison of the theory of planned behavior and the theory of
reasoned action. Personality and Social Psychology Bulletin 1992;18(1):3–9.
behavior of other business owners. Finally, studying the other organizational
Mykytyn PP, Harrison DA. The application of the theory of reasoned action to senior
facilitators and inhibitors of e-commerce adoption in Latin American contexts management and strategic information systems. Information Resources Manage-ment
will help the economy of the region continue to grow. Journal 1993;6(2):15–26.
Nasco SA, Grandón EE, Mykytyn PP. Predicting electronic commerce adoption in Chilean
SMEs. Journal of Business Research 2008;61(6):697–705.
Nor K, Pearson JM. An exploratory study into the adoption of Internet banking in a developing
country: Malaysia. Journal of Internet Commerce 2008;7(1):29–73.
Oliver RL, Bearden WO. Crossover effects in the theory of reason action. Journal of Consumer
References Research 1985;12:324–40.
Orbell S, Hodgkins S, Sheeran P. Implementation intentions and the theory of planned behavior.
Ajzen I. The theory of planned behavior. Organizational Behavior and Human Decision Personality and Social Psychology Bulletin 1997;23(9):945–54.
Processes 1991;50:79-211. Pavlou PA, Fygenson M. Understanding and predicting electronic commerce adoption:
Ajzen I, Fishbein M. Understanding attitudes and predicting social behavior. Englewood Cliffs, an extension of the theory of planned behavior. MIS Quarterly 2006;30(1):115–43.
NJ: Prentice-Hall; 1980. Pee LG, Woon IM, Kankanhalli A. Explaining non-work-related computing in the workplace: a
Armitage CJ, Conner R. Efficacy of the theory of planned behaviour: a meta-analytic review. comparison of alternative models. Information & Management 2008;45:120–30.
British Journal of Social Psychology 2001;40(4):471–99. Randall DM. Why students take elective business ethics courses: applying the theory of planned
Bentler PM. On the fit of models to covariances and methodology to the bulletin. behavior. Journal of Business Ethics 1994;13:369–78.
Psychological Bulletin 1992;112:400–4. Riemenschneider CK, McKinney VR. Assessing belief differences in small business adopters
Bollen KA. Structural equations with latent variables. NY: Academic Press; 1998. and non-adopters of web-based e-commerce. Journal of Computer Information Systems
Byrne BM. Structural equation modeling with LISREL, PRELIS, and SIMPLIS: basic concepts, 2001–2002:101–7.
applications, and programming. NJ: Lawrence Erlbaum Associates Publishers; 1998. Riemenschneider CK, Harrison DA, Mykytyn PP. Understanding IT adoption decisions in small
Chang MK. Predicting unethical behavior: a comparison of the theory of reasoned action and business: integrating current theories. Information and Management 2003;40:269–85.
the theory of planned behavior. Journal of Business Ethics 1998;17:1825–34.
CORFO PYME: Un desafío a la modernización productiva. Corporación de fomento a la Sheppard BH, Hartwick J, Warshaw PR. The theory of reasoned action: a meta-analysis of past
producción; 1994. research with recommendations for modifications and future research. Journal of Consumer
Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information Research 1988;15:325–43.
technology. MIS Quarterly 1989:319–40. Taylor S, Todd P. Understanding information technology usage: a test of competing models.
Fishbein M, Ajzen I. Belief, attitude, intention, and behavior: an introduction to theory Information Systems Research 1995a;6(2):144–77.
and research. Reading, MA: Addison-Wesley; 1975. Taylor S, Todd P. Decomposition and crossover effects in the theory of planned behavior: a
Fu JR, Farn CK, Chao WP. Acceptance of electronic tax filing: a study of taxpayer intentions. study of consumer adoption intentions. International Journal of Research in Marketing
Information & Management 2006;43:109–26. 1995b;12:137–55.
Harrison DA, Mykytyn PP, Riemenschneider CK. Executive decisions about adoption of Terry DJ, O'Learly JE. The theory of planned behaviour: the effects of perceived behavioural
information technology in small businesses: theory and empirical test. Information Systems control and self-efficacy. British Journal of Social Psychology 1995;34:199–220.
Research 1997;8(2):171–95. Vallerand RJ, Deshaces P, Cuerrier J, Pelletier LG, Mongeau C. Ajzen and Fishbein's theory of
Hofstede G. Cultures and organizations. Software of the mine. Intercultural cooperation and its reasoned action as applied to moral behavior: a confirmatory factor analysis. Journal of
importance for survival. New Jersey: McGraw-Hill; 1997. Personality and Social Psychology 1992;62:98-109.
Khalifa M, Shen KN. Explaining the adoption of transactional B2C mobile commerce. Venkatesh V, Morris V, Davis G, Davis F. User acceptance of information technology:
Journal of Enterprise Information Management 2008;21(2):110–24. toward a unified view. MIS Quarterly 2003;27(3):425–78.
See discussions, stats, and author profiles for this publication at:
https://www.researchgate.net/publication/227375525
The Role of Resource Flexibility in

Leveraging Strategic Resources
Article in Journal of Management Studies · July 2011
Impact Factor: 4.26 · DOI: 10.1111/j.1467-6486.2009.00912.x · Source: RePEc
CITATIONS READS
43 174
oo authors, including:
James G. Combs
University of Central Florida
SEE PROFILE
R. Duane Ireland
Texas A&M University
SEE PROFILE
Available from: R. Duane Ireland

Retrieved on: 11 May 2016
Journal of Management Studies 48:5 July 2011
doi: 10.1111/j.1467-6486.2009.00912.x
The Role of Resource Flexibility in Leveraging

Strategic Resources
James G. Combs, David J. Ketchen, Jr, R. Duane Ireland

and Justin W. Webb
Florida State University; Auburn University; Texas A&M University; Oklahoma State University
abstract Recent advances in resource-based theory suggest that the ways managers use
strategic actions to leverage resources has important influences on firms’ resulting competitive
advantages. However, theory dealing with the nature of leveraging remains underdeveloped.
We develop the notion that strategic actions that successfully leverage one resource might not
leverage another resource. In particular, our theorizing distinguishes between flexible
resources that can be leveraged in new markets and rigid resources that must be leveraged in
their current market. Using data from the restaurant industry, we find evidence that top
management team experience (a flexible resource) can be profitably leveraged by franchising,
but that attempting to leverage specific knowledge (a rigid resource) through either franchising
or multi-chaining weakens performance. Our theory and results help explain how different
types of resources can be leveraged using different strategic actions.
INTRODUCTION
Resource-based theory (RBT) has facilitated scholars’ efforts to understand why some
firms outperform others, which is a central goal of strategic management scholarship
(Meyer, 1991). RBT posits that competitive advantage and performance are shaped by
firms’ unique resource endowments (Barney, 1991, 2001). Specifically, a resource is
capable of generating sustainable performance advantages when it satisfies four criteria.
First, resources must be valuable such that they help firms increase efficiency or enhance
buyers’ willingness to pay premium prices. Second, resources need to be sufficiently rare
so that they are not readily available to competitors. The third and fourth criteria are that
resources must not be easy for competitors to imitate or substitute. Resources that satisfy
these criteria are called ‘strategic resources’ (Black and Boal, 1994; Chi, 1994), and
possessing such resources helps firms outperform rivals (Barney, 1991). Empirical evi-
dence appears to support RBT’s predictive ability (Crook et al., 2008).
Address for reprints: David J. Ketchen, Jr, Department of Management, College of Business, Auburn
Univer-sity, Auburn, AL 36849-5241, USA (ketchda@auburn.edu).
© 2010 The Authors

Journal of Management Studies © 2010 Blackwell Publishing Ltd and Society for the
Advancement of Management Studies. Published by Blackwell Publishing, 9600 Garsington
Road, Oxford, OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
Leveraging Strategic Resources 1099
However, critics argue that RBT suffers from a significant problem in that the path
through which resources are managed to generate above normal profits remains elusive
(Priem and Butler, 2001). Investigating how the performance effects of strategic resources
change as environmental contingencies change is one approach to understanding this path
(Miller and Shamsie, 1996). A second approach centres on discovering how the resource
management process shapes the performance implications of strategic resources (Eddleston
et al., 2008). With regard to the latter approach, Sirmon et al. (2007) describe firms’
resource management process as consisting of three core strategic actions: (1) structuring
the resource portfolio; (2) bundling resources and capabilities; and (3) leveraging
resources and capabilities to create customer value.
Our focus is on how managers use strategic actions to leverage the strategic resources
described by RBT. A few studies show that the actions taken to leverage strategic resources
can affect the extent to which resources affect performance (i.e. Eddleston et al., 2008;
Edelman et al., 2005; Frynas et al., 2006). However, scholars have yet to build and test
theory that might explain systematically how different strategic actions successfully
leverage some resources, but not others. We draw upon Leonard-Barton’s (1992)
distinction between core competencies and core rigidities, and a similar distinction in
international business research between location and non-location bound advantages
(Rugman and Verbeke, 1992) to build and test theory around the notion that market-
expansion strategies (i.e. moving into new markets) successfully leverage flexible but not
rigid strategic resources, which are best leveraged through market-focus strategies (i.e.
improving operations within existing markets).
Because a strategic resource’s value can be shaped by a firm’s environment (Amit and
Schoemaker, 1993; Miller and Shamsie, 1996), we investigate the effects of market
expansion strategies (i.e. strategies through which firms expand beyond current geo-
graphic and/or product markets) on the strategic resource–firm performance relation-ship
within a single industry. We concentrate on a service-based industry, partly because service
firms are of increasing importance to economies throughout the world (Wolf, 2005).
Among service industries, we find the restaurant industry particularly appealing given that
it is a vital economic and socio-cultural player in today’s business environ-ment. As
evidence for this assertion, we note that in the United States, the National Restaurant
Association (2009) projected 2009 industry sales to be a record $566 billion. Nearly 50 per
cent of every food dollar is spent away from home (Spielberg, 2006). To support such sales
levels, the restaurant industry employs approximately 9 per cent of the workforce in
developed nations – second only to the government sector (Koteff, 2007).
The fact that important relevant strategic resources and expansion strategies have
been identified in previous research is a second reason for studying the restaurant
industry (Combs et al., 2004). Specifically, we submit that firm-specific top
management team (TMT) experience is not only of great importance in this industry
but is also flexible in that it can be leveraged in multiple geographic and product
markets. Specific knowl-edge embedded in organizational routines is also a valuable
resource in this setting, but it is more rigid in that it is difficult to transfer among
operations in different geographic and product markets.
Franchising and multi-chaining are the two specific expansion strategies we investi-
gate. Franchising involves a firm’s managers contracting with local entrepreneurs to
© 2010 The Authors
Journal of Management Studies © 2010 Blackwell Publishing Ltd and
Society for the Advancement of Management Studies
1100 J. G. Combs et al.
build a chain of outlets that deliver offerings in different geographic locations. In
multi-chaining, managers expand a firm by building additional chains that each have
their own brand name and operating routines. While franchising centres on geographic
expansion, multi-chaining facilitates expansion into new product segments (e.g.
American casual, Italian, seafood). Our theorizing suggests that leveraging flexible
firm-specific TMT experience via franchising or multi-chaining should benefit
restaurant firms’ perfor-mance, but that the positive benefits of specific knowledge will
decline when managers attempt these expansion strategies. Specific knowledge, we
submit, is relatively rigid and not easily leveraged in new markets. Our empirical
analysis offers initial support for our theorizing.
Our central contribution is extending resource-based theory by explaining how dif-
ferent types of resources interact with different strategic moves to affect resource
value. While strategic resources, by definition, must satisfy the criteria of being
valuable, rare, inimitable, and non-substitutable, our theorizing and results suggest that
differences exist among strategic resources in terms of their flexibility. The flexibility
of a strategic resource, we contend, is a function of how underlying characteristics of
the resource interact with characteristics of the strategy used to leverage the resource.
Specifically, we offer theory suggesting that some strategically valuable resources are
more flexible than others with respect to the extent to which they can be leveraged in
new geographic and product markets. Consequently, the effectiveness of expansion
strategies depends on the characteristics of the resources being leveraged.
Our results in support of our theorizing also contribute to the understanding of
franchising and multi-chaining. Researchers are just beginning to investigate how fran-
chising affects firm performance (Combs et al., 2004), and our results show important new
contingencies that appear to impact the performance consequences of this strategy. Multi-
chaining is an important strategy in many industries that has escaped empirical attention.
Although multi-chaining could be viewed as a form of highly related diversi-fication that
may improve performance, our results suggest that its performance impli-cations depend
on the resources being leveraged among the firm’s multiple chains.
LEVERAGING STRATEGIC RESOURCES

Because RBT has enhanced researchers’ ability to explain differences in firm perfor-mance,
it has gained considerable currency and is now a central framework for strategic
management research (Barney et al., 2001). Although RBT research has advanced our
understanding of performance determinants, critics of RBT suggest that it suffers from a
notable problem in that the path by which strategic resources are transformed into
performance advantages is largely hidden. Specifically, Priem and Butler (2001) highlight
two important contingencies whose effects on resource value remain unclear: environ-ment
and firm strategy. In other words, the value of resources may change because of the
trajectory of a firm’s external environment and/or its managers’ strategic actions.
Miller and Shamsie (1996) took an important first step towards illuminating how the
environment interacts with resources to influence value creation. They showed that as the
industry shifted from stable to uncertain, what constituted a strategic resource among
movie studios shifted from long-term relationships with stars and theatre ownership to
© 2010 The Authors

Journal of Management Studies © 2010 Blackwell
Publishing Ltd and Society for the Advancement of
Management Studies
managerial knowledge. Further, Miller and Shamsie (1996) discovered that resources
that had once been valuable became detrimental to the studios’ efforts to cope with
heightened uncertainty. Other scholars have reported similar findings for the contingent
value of resources in different environments (e.g. Brush and Artz, 1999; Coff, 2002;
Eddleston et al., 2008). Thus, evidence is building to suggest that the value of a firm’s
stock of resources is contingent on the environment, as suggested by Priem and Butler
(2001).
Our focus is on the second contingency that Priem and Butler (2001) highlight –
strategy. Firms’ environments may give rise to resource value, but managers need to
recognize this value and engage in a pattern of strategic actions that takes advantage of
a resource’s potential (Priem and Butler, 2001). As Michael Porter noted, ‘the value of
a particular resource is only manifested if you have a particular strategy which realizes
that value’ (Argyres and McGahan, 2002, p. 50).
Building on the recognition that strategic actions are a potentially important contin-
gency, researchers have begun to investigate how different strategic actions shape the
efficacy of strategic resources. A variety of theoretical notions provide support for the
general idea that certain strategic resources can be applied to new areas. For example,
some firms pursue bricolage – a process through which entrepreneurs create value by
overcoming limitations via novel resource combinations (Baker and Nelson, 2005).
Pioneering bundling activities derive from exploratory learning and represent a
pathway through which managers create wholly new capabilities by integrating
existing and/or new resources that were previously unrelated (Ahuja and Lampert,
2001). Also, firms pursuing strategic entrepreneurship often leverage resources that
have provided com-petitive advantages in existing domains in an effort to exploit
opportunities in new domains (Ketchen et al., 2007).
Sirmon et al. (2007) provided a theoretical framework for classifying the different
ways that managers affect resource configurations. They identified three core resource
man-agement practices: (1) structuring the resource portfolio; (2) bundling resources
and capa-bilities; and (3) leveraging resources and capabilities to create customer
value. Structuring the resource portfolio refers to identifying the resources that are
needed to fulfil the firm’s strategy and then making the necessary asset acquisitions
and divestitures. Bundling involves integrating existing and newly acquired resources
to form bundles of strategic resources. Finally, leveraging involves taking actions that
mobilize, coordinate, and deploy strategic resources (Sirmon et al., 2007).
Given our interest in existing strategic resources rather than the processes through
which firms develop new resources, we focus on how firms create value through lever-
aging strategic actions. In doing so, we attempt to take important steps beyond three
studies that have previously investigated resource leveraging effects. Frynas et al.
(2006) analysed three case studies in which firms entered international markets using
first mover strategic actions to leverage superior political resources. Each firm was
able to gain and sustain competitive advantage by garnering foreign government
support before compet-ing firms entered. Interestingly, some competitors using late
mover strategies were able to wrest away valuable market share by leveraging superior
technical skills. Overall, Frynas et al.’s (2006) analysis suggests that managers can
leverage strategic resources in new markets.
© 2010 The Authors

In contrast to Frynas et al.’s (2006) emphasis on international expansion, Edelman et
al. (2005) investigated the extent to which managers’ strategic focus on innovation or
customer service mediated the effects of organizational and human resources on perfor-
mance among 192 small firms. Although neither organizational resources nor human
resource skills were directly related to performance, Edelman et al. (2005) found a
positive relationship between such resources and firm performance when the level of
strategic focus on innovation or customer service were included as mediating variables.
These results imply that perhaps strategic focus is the key to leveraging strategic
resources. Not all leveraging strategies appear successful, however. In a third study
involving resource leveraging, Eddleston et al. (2008) investigated the effect of small,
family-firm managers’ attempts to leverage their innovative capacity through strategic
planning and found that such attempts actually diminished performance.
These studies build knowledge by demonstrating that different strategic actions
affect the efficacy of different strategic resources. They show us that the value
provided by a resource can change depending on the strategic actions used to leverage
the resource. The next step, it appears, is to go beyond knowing broadly that resource
value changes with leveraging actions and to develop theory that can explain when a
particular strategic action might successfully leverage a particular strategic resource.
Accordingly, our theo-rizing below takes a first step towards explaining how particular
resource characteristics impact the value of a resource when managers engage in
strategic actions that move the firm into new geographic and product markets.
STRATEGIC RESOURCE CHARACTERISTICS AND

EXPANSION STRATEGIES
Of the four resource characteristics that define a strategically valuable resource – i.e.
value, rarity, inimitability, and non-substitutability (Barney, 1991) – value is
determined externally by market forces (Priem and Butler, 2001). Thus, it is this
attribute that potentially changes when managers attempt to leverage existing resources
into new markets via expansion strategies. We draw on two prior resource
classifications to distinguish between flexible strategic resources that retain their value
across multiple geographic and product markets and rigid strategic resources that are
specific to their current market and thus lose value when deployed elsewhere. Leonard-
Barton (1992) studied 20 new technology projects in five large firms and found that
whereas success depended heavily on managers’ ability to draw on the firm’s core
capabilities, the more that a project’s technological needs deviated from what the firm
had accomplished previously, the more that core capabilities became ‘core rigidities’.
The reason is that firms’ core capabilities are comprised of technological knowledge
that is reinforced by organizational routines, values, norms, and reward systems;
however, the new knowl-edge that is required for a project’s success is often
incompatible with these reinforcing elements of an existing core competency. New
knowledge often requires a different set of routines, values, norms, and reward systems
to be effectively managed, and the existing capabilities that Leonard-Barton observed
were rigid with respect to the new technolo-gies required by many projects’ mandates.
© 2010 The Authors

Management Studies
A second, related distinction is between location and non-location bound strategic
resources in an international context (Rugman and Verbeke, 1992). Non-location
bound strategic resources can be exploited globally with little adaptation to local or
regional differences in market structure or consumer preference, allowing firms to
build substan-tial economies of scale. Location-bound strategic resources, in contrast,
benefit the multinational corporation only in a particular country or region. Rugman
and Verbeke (1992) showed that this distinction is important for explaining how
multinational firms organize for international expansion.
Generalizing these concepts broadly within the logic of the RBT, we submit that
some strategic resources are flexible in that their value is retained when leveraged
through expansion strategies that deploy them across geographic and product markets.
Such resources are elastic in that they maintain strategic value as they are repeatedly
redeployed in different contexts. Examples include generalized managerial talent, or
knowledge about evaluating and integrating acquisitions, both of which can be suc-
cessfully transferred across industries (e.g. Harris and Helfat, 1997; Zollo and Singh,
2004). In contrast, rigid strategic resources create competitive advantages in only one
or a narrow set of geographic or product markets. Organizational routines imbued with
specific knowledge often act as rigid strategic resources, particularly when they are
developed in one part of an organization and extended to other parts (Szulanski, 1996).
Maritan and Brush (2003), for example, show how the unique path dependent histories
of production facilities challenged one firm’s ability to transfer manufacturing routines
among its facilities. Other strategic resources, such as organi-zational culture, might be
flexible or rigid, depending on how they manifest within a particular firm. For
example, a culture that values openness and change might be flexible whereas one that
emphasizes frugality and discipline might be rigid (Denison and Mishra, 1995).
The contrast between flexible and rigid strategic resources is related to but distinct from
similar constructs. In the RBT literature, a dynamic capability is an organizational
capability that can be adapted to changing environmental conditions (Teece et al., 1997). A
dynamic capability is flexible by definition, but a flexible strategic resource as defined here
is not necessarily dynamic – a flexible resource can be deployed in multiple settings
without adjusting its basic nature (cf. Rugman and Verbeke, 1992).
The concept of a rigid strategic resource is related to but distinct from the concept of a
specific asset in transaction cost theory (Williamson, 1991). Specific assets are physical or
human resources that have their value tied to a particular transaction. Although rigid
resources and specific assets are both defined by their superior ability to create value in
their current use, specific assets are not necessarily difficult to imitate (cf. Williamson,
1991). Thus, rigid strategic resources can be thought of as a subset of specific assets that
are not only tied to their current use, but also elude imitation.
We theorize that the success of market expansion strategies depends on the
flexibility of the resources that underlie the firm’s competitive advantage; market
expansion strategies are an effective way to leverage flexible resources, while rigid
resources are best leveraged by market-focused strategies. Testing these notions
requires us to identify the characteristics of strategic resources that affect flexibility.
Fortunately, RBT has advanced significantly in terms of identifying resource charac-
© 2010 The Authors

teristics that affect competitors’ ability to imitate. Using the context of US restaurant
chains as an example, we submit that flexible resources have characteristics that facili-
tate internal deployment without making competitor imitation more likely. Rigid
resources, in contrast, are difficult to imitate, but the same features that make them
difficult to imitate also make them difficult for a firm to leverage them via market
expansion strategies.
Restaurant Firms’ Expansion Strategies

We focus on two expansion strategies that are central ways that restaurant chains leverage
resources in new markets – franchising and multi-chaining. Franchising facilitates
geographic market expansion. It is defined by a contractual agreement in which one firm
(the franchisee) pays an up-front fee and ongoing royalties to the focal firm (the fran-
chisor) for the right to sell the firm’s products and services and to use the franchisor’s
trademark and/or business format. When considering growth, restaurant chains must decide
whether to employ managers in company-owned outlets or enlist franchisees to own and
manage outlets. Firms can focus exclusively on growth via franchising (e.g. Subway), keep
all of their units company-owned (e.g. The Cheesecake Factory), or use a combination of
the two forms (e.g. Panera Bread Company). Franchising typically allows faster growth
into new geographic markets than company ownership because franchisees incur a large
proportion of the costs of developing new outlets (Shane, 1996). Further, as owners,
franchisees will monitor their outlets closely (Krueger, 1991); as such, outlets can be
located far from the chain’s headquarters without notably increasing monitoring costs
(Brickley and Dark, 1987).
Multi-chaining refers to one firm’s ownership of multiple restaurant chains. Thus, it
is a strategy that allows restaurant firms to expand into different product markets. For
example, Darden Restaurants owns chains focused on the seafood (Red Lobster),
Italian (Olive Garden), and steakhouse (LongHorn Steakhouse) segments of the
restaurant industry, among others (Bahama Breeze, The Capital Grille, and Seasons
52). Similarly, Brinker International operates chains within the fast casual (Chili’s),
Southwestern (On the Border Mexican Grill and Cantina), and Italian (Maggiano’s
Little Italy) segments. Although multiple chains might share important administrative
functions such as pur-chasing, they are distinct at the outlet level. Customers rarely
know that the chains have common ownership. Our interest is in the overall
distribution of outlets among multiple chains in that a firm with outlets distributed
evenly over three chains is more actively managing multiple chains than a firm with
almost all of its outlets in a single chain and the remainder spread among a series of
one-, two-, or three-outlet experiments that may or may not grow into larger chains.
Six of the largest restaurant chains by sales are part of companies that own multiple
chains, and the eight largest multi-chain restaurants account for over 40 per cent of
total sales among all 400 chains contained in Restaurant and Institutions’ 2007
listing. Thus, multi-chaining is an important vehicle for market expansion within the
restaurant industry. Although this type of growth is within one industry, multi-chaining
can be viewed as a type of highly related diversification, given the many semi-distinct
niches (e.g. Italian, Southwestern, steakhouses, seafood).
© 2010 The Authors

Management Studies
Restaurant Firms’ Strategic Resources
We examine two resources that have been identified as strategic resources in prior research,
particularly for restaurant firms – firm-specific top management team (TMT) experience
and specific knowledge embedded in operating routines. Although there are other resources
that potentially have strategic value, managerial experience (Combs and Ketchen, 1999;
Shane, 1996) and specific knowledge (Kaufmann and Eroglu, 1999; Knott, 2003) have
been highlighted as key success factors in this competitive context. We also focus on these
two strategic resources because we believe that they furnish strong examples of the contrast
between flexible and rigid resources. Other resources that are important among restaurant
chains, such as a chain’s brand reputation (e.g. Michael, 2000), are less clear in that they
seem to have attributes that make them flexible in some contexts and rigid in others
(Nayyar, 1990). Next, we explain why these two resources are considered valuable, rare,
inimitable, and non-substitutable in this context.
Firm-specific TMT experience can be an important strategic resource (Castanias and
Helfat, 1991). A TMT with many years working together in a firm is likely to have
developed important firm-specific wisdom and skills. In the restaurant industry, unifor-
mity and predictability are important goals in that consumers expect consistent, high-
quality service (Bradach, 1997). Therefore, intimate knowledge of how a particular
restaurant concept ‘works’ is a key for executives to help firms achieve superior perfor-
mance (Combs and Ketchen, 1999). Much of this knowledge is embodied within TMT
executives and gained over time as executives learn the characteristics of beneficial
strategic actions and how to implement these actions efficiently and effectively
(Castanias and Helfat, 1991).
Long-term executives also have experienced greater opportunities to develop intra-and
inter-firm relationships that are vital to making informed decisions (Shen, 2003). For
example, a seasoned executive may form a strong relationship over time with a key
supplier, who may share valuable intelligence about competitors’ actions. Also, relation-
ships developed over time with franchisees may engender the trust needed to persuade
franchisees to adopt operational changes and participate in promotional activities (Bradach,
1997). Because TMT experience and their important stakeholder relationships build over
time, it is not a resource that less-tenured teams can easily replicate (Castanias and Helfat,
1991; Dierickx and Cool, 1989). Therefore, at least in the restaurant indus-try, long-tenured
TMT executives may be considered a strategic resource.
Specific knowledge, which is the other resource we examine, is defined as ‘knowledge
that is costly to transfer to agents’ ( Jensen and Meckling, 1995, p. 4; see also Ambrosini
and Bowman, 2001). Restaurants must develop chain-specific operational procedures that
are effectively communicated to and implemented by outlet managers. Such stan-
dardization is central to competitive advantage because it can reduce costs, such as those
associated with quality control and service consistency (Kaufmann and Eroglu, 1999).
Standardization is an important element in how restaurant firms leverage their resources
because customers expect little variance as they move from one of a chain’s outlets to
another (Kaufmann and Eroglu, 1999). Although standardized operational routines may
appear to be imitable, wholesale imitation is unlikely because much of the knowledge
behind these routines is tacit, complex, and deeply embedded in the firm (Combs and
© 2010 The Authors

Ketchen, 1999). Indeed, one of the firms in our sample, Luby’s Cafeterias, claimed that
it took seven years for managers to develop the knowledge needed to operate an outlet
successfully (Luby’s Cafeterias, 1993).
Even when successful routines are based on explicit forms of knowledge, two factors
make imitation difficult. First, they are based on what Barney (1995, p. 54) calls ‘numer-
ous small decisions’. While each operational routine, such as the way guests are greeted,
can be easily imitated, the routines are part of a package of patterns and practices that is
difficult for competitors to copy in its entirety. Second, competitors are often focused on
building their own specific knowledge and thus do not seek out and copy what others are
doing (Knott, 2003). In sum, standardized routines that are imbued with firm-specific
knowledge are unlikely to be successfully imitated and as such, are likely to serve as a
source of sustainable competitive advantage when leveraged appropriately.
Expansion Strategies and the Flexibility of Strategic Resources

Our contention that firm-specific TMT experience is a flexible strategic resource, at
least in the context we studied, is based on the concept of dominant logic (Prahalad and
Bettis, 1986). Dominant logic is ‘the way in which managers conceptualize the
business and make critical resource allocation decisions’ (Prahalad and Bettis, 1986, p.
490). Domi-nant logic can guide managers’ environmental scanning efforts and furnish
a roadmap for successful action (Lampel and Shamsie, 2000). Further, diversified firms
can enhance business-unit performance through the application of a dominant resource
management logic among divisions (D’Aveni et al., 2004). Importantly, the
development of managers’ dominant logic appears idiosyncratic to a firm and its
managers and evolves in a path dependent way that makes it difficult for competitors
to imitate (Cote et al., 1999). Thus, at least under circumstances where a dominant
managerial logic can apply in different situations, it can generate resource-based
performance advantages (Castanias and Helfat, 1991).
The concept of dominant logic implies limits to where firm-specific managerial
expe-rience can be effectively deployed because managers might attempt to generalize
logic that is effective in one context into another where it is not appropriate (Prahalad
and Bettis, 1986). However, adapting a single business to new geographic areas
(Sambharya, 1996) or across highly related businesses such as different restaurant
chains (Prahalad and Bettis, 1986) involves very similar strategic needs wherein the
dominant logic developed by executives with lengthy tenures should apply. As an
example, Darden Restaurants owned Olive Garden and Red Lobster at the time of our
study. Whereas each chain has a distinct building template, menu, and market segment,
they share a similar service orientation, price point, and marketing approach that
should allow managers to apply their managerial dominant logic in both chains.
Even in the extreme case of international expansion into a foreign market that is not
familiar with a chain’s product, such as Starbuck’s expansion into tea-drinking China (e.g.
Harrison et al., 2005), managers must understand what is core to their success and what
adaptations will not threaten the brand (Kaufmann and Eroglu, 1999). In short, dominant
logic is the schema that develops among top managers for conceptualizing the business and
guiding decision making (D’Aveni et al., 2004). Whereas it can have severe
© 2010 The Authors

Management Studies
limits (Bettis and Prahalad, 1995), dominant logic has been found to be of strategic
benefit for managing operations that share similar strategic and operational logics
(D’Aveni et al., 2004), which we believe is the case for restaurant firms’ expansion into
new geographic markets and multiple chains.
Franchising and firm-specific TMT experience. Franchising is an alternative to internal

growth through company ownership that overcomes managerial limits to growth (Penrose,
1959), allows rapid expansion across geographic markets (Shane, 1996), and increases
economies of scale in purchasing and advertising that, in turn, improves performance
(Combs et al., 2004). The challenge, however, is to maintain quality standards and
standardization while growing through franchising. Franchisees are not under the same
level of control as employees; as a result, some will make unsanctioned adaptations to the
chain’s offerings, fail to join in chain-wide promotions, and under-invest in quality to the
detriment of the chain (Kaufmann and Eroglu, 1999; Michael, 2000).
Because of this challenge, experienced top managers appear to be in the best
position to oversee and gain value from franchisees. Consistent with the concept of
dominant logic, long-tenured executives possess idiosyncratic and detailed wisdom
about how new tasks and routines can be established without disrupting the level of
service quality provided to customers (Combs and Ketchen, 1999). Such executives are
likely to have established relationships with the firm’s largest and most influential
franchisees, which can make it easier for top managers to learn about franchisees’ local
adaptations that might be useful elsewhere in the chain (e.g. both the Egg McMuffin
and the Filet-o-Fish were created by McDonald’s franchisees and then spread to other
outlets), and to persuade franchisees to participate in promotional campaigns and
invest in quality (Bradach, 1997).
In short, the expansion strategy of franchising should enhance the value of firm-
specific TMT knowledge. As the guardians of standardization, the firm’s company-
owned outlets should also benefit from an experienced TMT (Bradach, 1997), but
because employee-managers at company-owned outlets lack incentive to innovate or
make local adaptations, top managers’ ability to identify the next important adaptation
or innovation might go unutilized. Thus, as the chain increases its use of franchising,
the value of the TMT’s experience should be increasingly exploited. Stated formally:
Hypothesis 1: The use of franchising moderates the firm-specific TMT experience–

performance relationship such that as the use of franchising increases, the more
positive the relationship between firm-specific TMT experience and performance.
Multi-chaining and firm-specific TMT experience. Multi-chaining is an important way

for restaurant chains to expand, but unlike franchising, its expansionary focus is across
product categories. Because successfully managing a restaurant chain that serves ham-
burgers, for example, should be strategically similar to managing a chain that serves tacos,
the concept of dominant logic suggests that multi-chaining should be one way to leverage
the deep operational knowledge and long-term relationships of an experienced TMT (cf.
Prahalad and Bettis, 1986). Possessing significant insights about existing chains should
help these executives facilitate physical resource-sharing among chains to reap
© 2010 The Authors

available synergies – synergies that are key to better performance as a multi-chain firm.
Consequently, long-tenured teams should be well suited to the challenges of
developing multiple restaurant concepts.
Although there is very little direct investigation of restaurant firms’ multi-chaining
activities, firm specific managerial experience is important to the success of the similar
strategy of related diversification (Michel and Hambrick, 1992), and a case study of a
multi-chain retailer found that the managerial processes, values, and norms used by top
managers influenced how successfully the chain integrated its activities (Mitronen and
Moller, 2003). While we would expect firm-specific TMT experience to increase the
efficiency of firms with only a single chain, firm-specific TMT experience provides
additional synergies, such as economies of scale and scope, within the context of multi-
chaining that are not possible in single-chain firms. Accordingly, we predict that:
Hypothesis 2: Multi-chaining moderates the firm-specific TMT experience–

performance relationship such that the more multi-chaining is undertaken, the more
positive the relationship between firm-specific TMT experience and performance.
Expansion Strategies and the Rigidity of Strategic Resources

Above we argued that the dominant logic of managers with firm-specific experience is
flexible enough to apply across geographic regions and new products. In contrast, we
contend that much of a restaurant chain’s specific knowledge cannot be as easily applied in
new markets. The RBT concept of causal ambiguity helps explain why specific knowledge
is potentially rigid with respect to expansion strategies (Reed and DeFillippi, 1990). The
concept suggests that the bundles of routines that support competitive advantage can be so
complex that even members of a firm might not understand how any particular routine
contributes to competitive advantage; such ambiguity makes it difficult for competitors to
imitate routines (Barney, 1991; Lippman and Rumelt, 1982). While causal ambiguity thus
helps protect a firm’s specific knowledge from potential imitators, it also undermines
internal expansion efforts. Indeed, causal ambiguity surrounding which routines will and
will not transfer seems to be a key impediment to managers’ attempts to transfer routines
among different facilities in the same company (Maritan and Brush, 2003; Szulanski,
1996). Thus, we argue that because of causal ambiguity, specific knowl-edge embedded in
operational routines is rigid in the sense that wholesale transfer of the firm’s routines is
unlikely to be successful as firms expand into new markets.
Franchising and specific knowledge. Based on the premise that causal ambiguity is
important to the transfer of capabilities, we submit that franchising will have a negative effect
on the relationship between specific knowledge and performance. Transferring knowledge to
franchisees and enforcing practices reflected in that knowledge is more difficult than
transferring knowledge to employee managers working in company-owned outlets (Bradach,
1997). Because franchisees are not employees of the firm, they are not required to attend
training beyond that specified in their initial contract, nor are they required to implement the
firm’s operational or promotional improvements (Bradach, 1997). Given that what is in the best
interest of the chain overall is not always in the best interest of the
© 2010 The Authors

Management Studies
local franchisee and vice versa, franchisees are not always open to adopting new
routines or enforcing existing routines (Brickley and Dark, 1987; Kaufmann and
Eroglu, 1999). Indeed, this is a key reason why customers are warned in some
franchisors’ advertise-ments that ‘participation may vary’.
Training franchisees appears to be an important part of transferring routines to them
(Michael and Combs, 2008), but there is strong evidence that franchisee resistance to
franchisors’ routines creates variance among outlets in pricing ( Lafontaine, 1999), local
service delivery strategy (Yin and Zajac, 2004), and product consistency (Kaufmann and
Eroglu, 1999). Overall, above the minimum knowledge required to operate an outlet,
franchisees absorb only the organizational routines that they choose to absorb (Bradach,
1997). Thus, although specific knowledge will generally be positively related to perfor-
mance, as chains increase their exposure to franchising (as opposed to growth through
company ownership), the transfer and use of such knowledge might be increasingly
inhibited, thereby attenuating this relationship. Therefore:
Hypothesis 3: The use of franchising moderates the specific knowledge–

performance relationship such that as the use of franchising increases, the less
positive the relation-ship between specific knowledge and performance.
Multi-chaining and specific knowledge. Specific knowledge is often tailored to the

needs of a particular restaurant concept. Although some knowledge may be shared as a
firm develops a new restaurant concept, much of its specific knowledge will not easily
transfer between chains. For example, the speed-oriented food preparation system (and
the knowledge contained within it) that Jack-in-the-Box has developed might not be
useful or appropriate in its less-automated, freshness-oriented Qdoba Mexican Grill
restaurants wherein the final preparations are conducted in front of customers and
catering is a large part of the business. Consequently, chain-specific knowledge must
be generated for each new chain, and the concept of causal ambiguity (Reed and
DeFillippi, 1990) suggests that it will be very challenging for managers to know which
routines can be applied in the new setting. Managers might attempt to inappropriately
generalize some specific knowledge to a new restaurant concept, adversely affecting
performance (e.g. Haleblian and Finkel-stein, 1999). Thus, although specific
knowledge positively affects performance, transfer of such knowledge via multi-
chaining may undermine its effectiveness. As a result, the specific knowledge–
performance relationship will be weaker when multi-chaining is used. Accordingly:
Hypothesis 4: Multi-chaining moderates the specific knowledge–performance

relation-ship such that the more multi-chaining undertaken, the less positive the
relationship between specific knowledge and performance.
METHOD
Sample
The sampling frame consisted of all public US restaurant firms contained in the COM-
PUSTAT database in 1994. Firms that did not constitute a chain (i.e. at least four outlets
© 2010 The Authors

– Hawes and Crittenden, 1984) and firms whose restaurant operations were less than
70 per cent of the firms’ total sales (Rumelt, 1974) were eliminated. Ninety-four firms
met these criteria; data for reputation (a control variable) and specific knowledge were
gathered via an expert panel survey on these firms. Measures of the remaining
constructs in the study were gathered from the public documents on the 65 firms that
survived as public entities through 1996. The significant loss of firms over the three-
year period is not surprising given the easy entry and exit in this and other fragmented
industries where franchising is common (Lafontaine and Shaw, 1998). Our sample size
is relatively small, which means that if the real relationships that we seek to identify
are fairly small, our sample might not have the statistical power to detect them (i.e.
Type II error). Thus, non-significant results might not provide definitive evidence that
our theory is wrong; it would be possible that effects exist but are not large enough to
be detected within our sample. Conversely, our ability to find statistically significant
results in support of our hypotheses would suggest that these effects do exist and these
effects are of a moderate to large effect size.
Although researchers often prefer more recent data, Robins (2004) argues that data
remain highly relevant over time unless dramatic shifts take place involving the variables
of interest. In our case, it does not appear that the nature of resources, expansion strategies,
and performance have changed fundamentally since the mid-1990s. To verify this, we
compared the 252 restaurant franchisors listed in Entrepreneur magazine in 1995 with the
191 listed in 2009. Although Entrepreneur’s listing of mostly private firms is
fundamentally different from our sample of large public restaurant chains, it appears to be
representative of the population of established franchisors (Lafontaine, 1995); more-over,
these variables have been heavily used (and explained in more detail) in the franchising
literature (e.g. Lafontaine, 1992; Shane, 1996).
To see whether relationships among these frequently-studied variables have changed, we
calculated the correlations among important descriptive variables reported by Entre-
preneur magazine – the proportion of outlets franchised (a measure that is also used in our
hypotheses tests), total number of outlets, total start-up costs, initial franchise fee, the
royalty rate, and the number of geographic areas in which the franchisor is seeking
franchisees – and compared the 15 bivariate correlations among these variables in 1995
with the same correlations from 2009. Although most of the variables are increasing over
time in absolute terms – e.g. fees, start-up costs, and the number of outlets are all growing
– in terms of the correlations among the variables, only the positive relationship
between franchisors’ initial franchise fee and total start-up costs has changed. The
correlation is stronger today than it was in 1995. The primary purpose of this fee
appears to be to recapture the franchisors’ costs associated with helping a franchisee
launch their business (Lafontaine, 1992), so it might be that franchisors are doing a
better job of re-setting these fees as start-up costs rise. Overall, however, we found
little evidence that relationships among variables are systematically shifting.
Strategic Resources
We used an expert panel survey of hospitality management educators to assess specific
knowledge. Our decision to use an expert panel survey versus a direct survey of restau-
© 2010 The Authors
Management Studies
rant managers was based on our concern that managers may be too steeped in their own
firm to provide an unbiased evaluation of their firm’s specific knowledge vis-à-vis com-
petitors. Chen et al. (1993) showed that expert opinions provided by relevant academics
can offer valid measures of unobservable constructs. Following the generation of items,
pre-tests with nine senior food-service executives in three restaurant chains and with six
hospitality academics, and a factor analysis of pre-test results, four items were selected to
represent specific knowledge. Next, 400 hospitality management educators were
identified through the Council for Hotel, Restaurant, and Institutional Educators. In each of
ten regions, 40 experts were asked to rate an average of ten chains headquartered in their
geographic region. Respondents were asked to rate only those chains about which they felt
confident making an expert judgment. The survey was constructed in 1994 and distributed
following Dillman’s (1978) total design method. The Dillman total design method is based
on research about survey characteristics that enhance response rates, including the content
of the cover letter, size of the paper, use of graphics, presentation of items, incentives to
participate, and the frequency, timing, and format of follow-up.
There were 226 surveys returned. Forty-four respondents did not rate any chains
because they did not feel qualified to make the requested judgments, leaving a usable
response rate of 45.5 per cent. On average, raters were knowledgeable about 5.4
chains. The average number of ratings for a chain was 9.3 and average inter-rater
reliability was 0.81. Respondents were not different from non-respondents in terms of
the type of institution where they were employed ( public or private, F = 0.02; p =
0.91) or the length of hospitality programme offered (2 or 4 years, F = 0.91; p = 0.34).
Thus, we did not find evidence of non-response bias.
Specific knowledge was assessed via four survey items asking the experts about
how difficult it would be to either write down or train someone in a chain’s operations.
Almost all of the firms in our sample have a dominant chain that is significantly larger
than the others, and we measured specific knowledge by asking about the firm’s
dominant chain. In the two cases where the firm had two dominant chains of roughly
the same size, the chains were very similar in style (i.e. both fast food in one case and
both full-service in the other) and had almost identical scores on specific knowledge.
In these two cases, the scores on the two dominant chains were averaged. Although this
approach does not measure the specific knowledge across all chains, it is consistent
with our theory in that we are interested in the impact of specific knowledge from the
dominant chain on expansion efforts into new product segments.
The items were: Comparing this company to all other food service operations, (1)
How long would it take to train competent assistant managers? (anchored by ‘not long’
and ‘very long’ using a 7-point Likert scale); (2) How long would it take to train
competent hourly employees? (not long/very long); (3) How difficult would it be to
communicate job requirements to unit-level managers? (not difficult/very difficult);
and (4) How difficult would it be to include all of the unit manager’s job tasks in an
operations manual? (not difficult/very difficult). Together, the items capture a degree of
complexity in a chain’s routines that could reflect rareness and inimitability. Reliability
was 0.86 and average inter-rater agreement was r *WG( J) = 0.66.
To measure top management experience, we first had to define ‘top managers’.
Following Bantel and Finkelstein’s (1995) recommendation, we defined top managers as
© 2010 The Authors

those with membership on the board of directors. This approach furnishes a clear
distinction between first- and second-tier executives. Board membership indicates
mem-bership in the firm’s dominant coalition (Bantel and Finkelstein, 1995). We
measured top management experience as the average of the number of years that top
managers had been affiliated with their firm as of 1995. Data were derived from 10-Ks,
annual reports, and proxy statements.
Expansion Strategies
Franchising and multi-chaining are the expansion strategies on which we focus. Fran-
chising is a central means for rapid geographic expansion (Shane, 1996) and multi-
chaining takes firms into new product segments. Franchising was measured as the
change in the ratio of units franchised to total units owned by a company from 1994 to
1995. This measure was chosen because it reflects the degree to which firms are trying
to increase or decrease their use of franchising.
Multi-chaining was measured using a specialized Herfindahl index. Herfindahl
indices typically use a firm’s sales in each SIC code to derive a measure of
diversification strategy. Because all of our firms are in the same SIC code, we
calculated the multi-chaining index (MC) using the percentage of outlets in each chain
for a given company ( pi), such that:
MC = 1− ∑ p2i
where pi is the percentage of outlets for chain i. Thus, our measure of multi-chaining
registers both the number of chains within the firm and the distribution of outlets
among the chains. Multi-chaining was measured in 1995. Outlet data were collected
from firms’10-Ks and annual reports.
Performance
Performance was examined using two commonly used measures of profitability:
return on assets (ROA) and earnings per share (EPS). We collected performance
data for 1996, the year following the measurement of independent variables. Both
measures are at the firm-level of analysis and thus aggregate the performances of
multiple chains within the firm. Performance data were obtained from COMPUSTAT.
Control Variables
To strengthen our tests, we controlled for each firm’s age, size, and the reputation of its
largest chain. Age and size have been used frequently in prior franchising research to
depict resources (Combs and Ketchen, 2003). Using 10-Ks and annual reports, age
was measured as the number of years since the firm initiated restaurant operations.
Size, which was not normally distributed, was measured as the natural log of total
sales. These data were derived from COMPUSTAT.
© 2010 The Authors

Management Studies
Some firms might simply possess better brand- and service-building capabilities as they
engage in expansion strategies. Thus, we also controlled for reputation quality of the firm’s
largest chain. Reputation was measured by taking the sum of the average ratings on the
following four items: Comparing this company to all other food service operations, (1)
How well respected is this company? (anchored by ‘not respected’ and ‘very respected’
using a 7-point Likert scale); (2) How good of a value is this company perceived to provide
for the price? (not good/very good); (3) How strong is this company’s reputation for
consistent quality and service? (not strong/very strong); and (4) How strong is this
company’s brand name recognition in its service area? (not strong/very strong). Reli-ability
was 0.87 and average inter-rater agreement was r *WG( J) = 0.65.
RESULTS
Table I displays descriptive statistics and correlations among our variables. Table II
shows the results of our hypotheses’ tests. The hypotheses were tested using
hierarchical moderated regression. One-tailed tests are statistically correct when
hypotheses are theory-driven and directional ( Jones, 1952). Thus, following other
scholars testing such hypotheses (e.g. Lazzarini et al., 2008; Love and Cebon, 2008;
Sanders and Hambrick, 2008), one-tailed tests were used for hypotheses tests while all
other (non-directional) significance levels are based on two-tailed tests. Because of the
measures’ greatly varying scales, the independent variables were standardized prior to
creating interaction terms (Aiken and West, 1991). Two sets of models were run, one
predicting ROA and one predicting EPS. Both fully specified models were significant
(F(11,53) = 2.74; p  0.01 for ROA, and F(11,53) = 3.40; p  0.01 for EPS).
Given our small sample, we wished to know our actual statistical power. The effect size
etas attributable to the change in R2 from the addition of the interaction terms for ROA and
EPS were ‘large’ (i.e. Cohen, 1977) at 0.39 and 0.41, respectively; both yield a 77 per cent
probability, or power, of detection. This is close to the standard desired power of 80
Table I. Descriptive statisticsa
Variable Mean S.D. 1 2 3 4 5 6 7 8
1. Return on assets 1.84 11.37

2. Earnings per share 0.20 0.89 0.76**
3. Age 22.63 16.27 0.20 0.17
4. Sizeb 441.37 751.57 0.15 0.14 0.24†
5. Brand name reputation 19.63 3.20 0.18 0.21 -0.09 0.23
6. TMT experience 9.80 6.92 0.20 0.15 0.26* 0.18 -0.03
7. Specific knowledge 13.80 3.84 -0.03 -0.16 -0.26* -0.28* 0.17 0.13
8. Franchising 0.02 0.10 -0.11 -0.18 -0.09 0.00 -0.11 -0.27* -0.20
9. Multi-chaining 0.12 0.18 -0.08 -0.19 -0.01 0.24† 0.12 0.05 0.17 0.14
Notes:
(GG) n = 65.
(HH) (000,000).
†
p  0.10, * p  0.05, ** p  0.01.
© 2010 The Authors

Table II. Regression results
Variable Hypothesis ROA results EPS results
Controls Main Interactions Controls Main Interactions

effects effects
Controls
Age 0.19 0.13 0.11 0.16 0.06 0.03
Size 0.07 0.07 0.07 0.06 0.04 0.10
Reputation 0.17 0.19 0.21 0.25† 0.18
Resources
TMT experience 0.16 0.31* 0.14 0.27*
Specific knowledge -0.02 -0.08 -0.18 -0.24†
Strategic actions
Franchising -0.02 -0.07 -0.11† -0.25†
†
Multi-chaining -0.11 -0.12 -0.19 -0.23
Interactions
TMT experience ¥ franchising 1 0.54** 0.85***
TMT experience ¥ multi-chaining 2 -0.21 -0.07
Specific knowledge ¥ franchising 3 -0.30† -0.54**
Specific knowledge ¥ multi-chaining 4 -0.40*** -0.27**
2
R 0.08 0.11 0.36 0.08 0.17 0.41
DR2 0.03 0.25*** 0.09 0.24***
F 1.72 1.03 2.74** 1.74 1.72 3.40**
*
p  0.10, * p  0.05, ** p  0.01, *** p  0.001.
per cent. About 66 per cent of studies in strategic management had the power to detect
such effects in Mazen et al.’s (1987) evaluation of statistical power. Thus, whereas our
sample is small, our actual statistical power is enough to detect the large effects that are
evident in these data.
Hypotheses 1 and 2 predicted that as a flexible strategic resource, firm-specific TMT
experience would continue to benefit the firm as it expands across multiple geographic
and product domains. Franchising is a key strategy for rapid geographic expansion, so
Hypothesis 1 stated that franchising moderates the firm-specific TMT experience–
performance link such that the relationship will be more positive as franchising
increases. The hypothesized relationship is supported for both ROA (b = 0.54; p 
0.01) and EPS (b = 0.85; p  0.001). Figure 1 plots the interactions for Hypothesis 1; it
shows that the value of TMT experience, in terms of ROA and EPS, increases with
higher levels of franchising. Hypothesis 2 predicted that multi-chaining moderates the
relationship between firm-specific TMT experience and performance such that the
relationship will be more positive as multi-chaining increases. However, the interaction
was not signifi-cant in the predicted direction for either ROA (b = -0.21; NS) or EPS (b
= -0.07; NS). Thus, no support for Hypothesis 2 was found.
Hypotheses 3 and 4 investigated whether expansion strategies would weaken the impact
of a rigid strategic resource – i.e. specific knowledge – on performance. Specifi-cally,
Hypothesis 3 anticipated that franchising negatively moderates the specific
© 2010 The Authors

1.00
0.80 Low Fran
0.60 High Fran
0.40
0.20
ROA
0.00
-0.20
-0.40
-0.60
-0.80
-1.00
Low High
Firm-Specific TMT Experience
1.00
Low Fran
0.50
High Fran
EPS 0.00
-0.50
-1.00
-1.50
Low High
Firm-Specific TMT Experience
Figure 1. Interactions of firm-specific TMT experience and franchising on ROA and EPS
knowledge–performance relationship. The interaction was weakly significant for ROA (b =

-0.30; p  0.10) and significant in the EPS model (b = -0.54; p  0.01), which we interpret
as overall support for Hypothesis 3. Consistent with expectations, the plotted interactions
in Figure 2 show that the value of specific knowledge, in terms of both ROA and EPS,
decreases with higher levels of franchising. Finally, Hypothesis 4 predicted that multi-
chaining negatively moderates the specific knowledge–performance relationship.
© 2010 The Authors

0.40
0.30 Low Fran
0.20 High Fran
0.10
0.00
ROA
-0.10
-0.20
-0.30
-0.40
-0.50
Low High
Specific Knowledge
0.60
0.40
Low Fran
0.20
High Fran
0.00
EPS
-0.20
-0.40
-0.60
-0.80
-1.00
Low High
Specific Knowledge
Figure 2. Interactions of specific knowledge and franchising on ROA and EPS
The interaction terms were negative and significant for both models (b = -0.40; p 
0.001 for ROA and b = -0.27; p  0.001 for EPS), offering support for Hypothesis 4.
The plotted interactions provided in Figure 3 show that EPS and ROA both decrease
when firms leverage high levels of specific knowledge with multi-chaining strategies.
Given the low inter-rater agreement of r *WG( J) = 0.66 for specific knowledge, as a
robustness test we re-ran the analysis removing the five firms with the lowest r *WG( J)
© 2010 The Authors

Management Studies
0.60
Low Multichaining
0.40
High Multichaining
0.20
0.00
ROA-0.20
-0.40
-0.60
-0.80
Low High
Specific Knowledge
0
.
6
0
Low Multich
EPS 0
.
4
0
0
.
2
0
0
.
0
0
-
0
-
0
-
0
-0.60
Low
High
Specific Knowledge
Figure 3. Interactions of specific knowledge and multi-chaining on ROA and EPS
scores. All significance effects remained significant and the explained variance for
both models improved slightly despite the additional reduction in sample size. Thus,
we do not believe that the inter-rater agreement scores materially affected results.
DISCUSSION
Extant evidence supports RBT’s basic tenets (Crook et al., 2008), but rich opportunities
remain to develop understanding of paths, patterns, and practices whereby strategic
resources enhance performance (Sirmon et al., 2007). Building on past evidence suggest-
ing that the way a resource is leveraged affects its value-creating potential (i.e. Eddleston
© 2010 The Authors

et al., 2008; Edelman et al., 2005; Frynas et al., 2006), we developed and tested theory
that takes a step towards explaining the types of resources that can and cannot be
leveraged via expansion strategies that deploy a resource across multiple geographic
and product markets. Specifically, we depicted flexible strategic resources as those that
can be deployed in multiple markets with little adaptation, whereas rigid resources
require significant adaptation when deployed in new markets.
Implications for Resource-Based Theory

We sought to shed additional light on RBT by building knowledge about one of the
theory’s key concepts: strategic resources. Our theorizing suggested that the
effectiveness of expansion strategies differ based on whether strategic resources are
flexible or rigid. First, we argued that TMTs with substantial firm-specific experience
would develop a dominant logic about how the business is effectively managed that
they could flexibly apply in new markets. As shown in Figure 1, results indicate that
this is true with respect to reaching new geographic markets via franchising. For both
performance measures, firm-specific TMT experience improves performance among
heavy franchisors, but not among firms that use more company-ownership. This
evidence supports our contention that experienced managers develop a dominant logic
regarding how to leverage their relationships with existing franchisees and their skills
working with new franchisees to expand the chain via franchising.
We did not similarly find evidence that multi-chaining leverages top managers’ firm-
specific experience, which we had expected based on the notion that dominant logic
applies across strategically similar businesses such as restaurant chains. Non-findings,
especially with a small sample, should not be treated as evidence that a relationship
does not exist. Nevertheless, the fact that we did not observe an interaction between
multi-chaining and top management experience raises questions about the limits of
dominant logic in defining the scope of a business.
Prahalad and Bettis (1986) suggest two limits to the flexibility of dominant logic. First,
the very act of moving from one business to two businesses requires a new logic, which
could have hampered the flexibility of firm-specific managerial skill for some of the firms
in our sample. That is, managers might understand the new chain in terms of its similarity
with the firm’s primary chain, but they still must learn to manage two (or more) separate
chains. Second, despite strategic similarities among businesses, each new busi-ness adds to
overall strategic variety, which limits managers’ ability to respond quickly and
appropriately to competitors’ moves. Overall, our results are suggestive that domi-nant
logic among TMT members is a flexible resource that can aid expansion, but defining the
boundaries of its flexibility requires additional investigation.
Our focus was on one of the three resource management processes described by
Sirmon et al. (2007) – i.e. the leveraging of existing resources. Our results, however,
have implications for how managers might effectively perform the other two resource
man-agement processes – i.e. structuring and bundling – as well as for the leveraging
processes that we study. The process of structuring deals with acquiring, constructing,
and divest-ing resources. Bundling deals with improving existing resources and
combining them in ways that create new capabilities.
© 2010 The Authors

Management Studies
In terms of structuring, results suggest that managers might actively develop flexible
resources in their firm’s portfolio. This seems desirable, given the potential to profitably
apply such resources in multiple settings. Building a dominant logic embedded in
managerial experience appears to be one path. Another possibility might be found in the
firm’s core competencies (Prahalad and Hamel, 1990). A core competency is a skill or set
of skills that is central to a firm’s competitive advantage; often, core competencies can be
leveraged in multiple markets (Lei et al., 1996). Classic examples include Honda’s exper-
tise in small engines and Casio’s expertise in miniaturization, which brought these firms
into a set of highly related businesses (Marino, 1996). It might be, for example, that a
restaurant firm can develop a core competency in conceptualizing, designing, and branding
restaurant chains, which should be highly flexible vis-à-vis expansion through multi-
chaining. Another possibility is that managers can build dynamic capabilities that can be
leveraged in new markets (Teece et al., 1997). There is evidence, for example, that some
repeat acquirers are able to capture what they learn in each acquisition and use it to
improve the performance of future acquisitions (Zollo and Singh, 2004). Thus, the concepts
of dominant logic, core competencies, and dynamic capabilities offer potential starting
points for structuring flexible resources into the firm’s resource portfolio.
The notion that competitive advantage can rest on resources that are rigid and not
easily adapted has implications for how managers choose to leverage resources and the
bundling processes that are applied to them as part of the resource management pro-
cesses (Sirmon et al., 2007). We argued that specific knowledge embedded in organiza-
tional routines is relatively rigid because managers confront causal ambiguity
surrounding which routines will transfer (Reed and DeFillippi, 1990). Studies of intra-
organizational attempts to transfer routines suggest that recipients often do not under-
stand routines in the same way as senders (Szulanski, 1996) and structural differences
between sending and receiving units can make it difficult to understand how to adapt
transferred knowledge (Maritan and Brush, 2003). As depicted in Figures 2 and 3, our
results are consistent with this logic in that when a firm’s primary chain is imbued with
specific knowledge, expanding via franchising or multi-chaining appears to dampen
performance. In contrast, firms that stayed focused on a single chain or through
company ownership appear to better leverage their specific knowledge for enhanced
performance.
In terms of leveraging, our theory and evidence suggests that the best strategies for
leveraging rigid resources are those that ‘stick to their knitting’ (Peters and Waterman,
1982). Consistent with this logic, in a study of the prevalence of sustained above
average industry performance, several of the firms that Wiggins and Ruefli (2002)
identified as sustained above average performers were restaurant chains that used the
slow-growth strategy of company-ownership. Generalizing to the broader context of
diversification, whereas the weight of the evidence suggests that firms perform best
with moderate levels of diversification (Palich et al., 2000), single business firms
perform almost as well (Rumelt, 1974) and many top performing firms are highly
focused strategically (Wiggins and Ruefli, 2002). Perhaps a distinguishing feature
between those firms that should stick close to their current business and those that
should pursue related diversification is the rigidity of the resources that support
competitive advantage in the firms’ primary business.
© 2010 The Authors

In terms of bundling, rigid resources suggest what Sirmon et al. (2007) called ‘stabi-
lizing’ processes wherein managers make incremental improvements to existing resources.
Whereas a bundling approach of incremental improvement can be effectively combined
with strategic focus in slow-moving industries (Nadkarni and Narayanan, 2007), fast-paced
industries require strategic flexibility (Baum and Wally, 2003; Brown and Eisenhardt,
1997). Even in a stable industry, an environmental jolt can turn a competitive advantage
into a liability (Audia et al., 2000). Thus, long-term success for firms relying on rigid
strategic resources might need to focus on ‘enriching’ and ‘pio-neering’ bundling activities
(Sirmon et al., 2007). Enriching involves extending a current resource by combining it with
complementary resources. For example, perhaps manag-ers can build a strong
organizational identity that decreases the rigidity of specific knowledge routines by
strengthening franchisees’ identity with the firm and its routines (cf. Cornelissen et al.,
2007). Pioneering involves combining resources in ways so as to create totally new
resources to address emerging market needs. Ultimately, effective bundling with rigid
resources might involve some form of ambidexterity wherein the firm exploits its flexible
resources in current markets while simultaneously pioneering new flexible resources that
can be leveraged across markets, thereby helping the firm cope with environmental change
(Gupta et al., 2006).
Implications for Franchising and Multi-Chaining Research

Our findings also provide insights for research on franchising and multi-chaining.
Fran-chising is an important path to geographic growth, but researchers are only
beginning to examine its performance implications. Extant inquiry tends to either view
franchising’s mere existence as evidence that it is efficient or view franchising as a
necessary evil (Combs et al., 2004). Our results suggest that the efficacy of franchising
depends on the characteristics of firms’ resource endowments. As more work assesses
the performance consequences of franchising, the interaction between franchising and
the franchisor’s resources should be considered. More generally, efforts to discover
whether franchising is either ‘good’ or ‘bad’ appear misdirected. Instead, franchising’s
implications for per-formance appear to be highly contingent.
Multi-chaining is an important phenomenon (Mitronen and Moller, 2003), but it has
largely escaped research attention. Because multi-chaining can be viewed as a unique case
of highly related diversification, studying it offers the potential to shed additional light on
the paths and patterns through which related diversification can enhance firm performance.
Theory suggests that even though specific knowledge is less generalizable to new chains,
as a form of highly related diversification, multi-chaining should leverage ‘back office’
skills such as superior site selection or creative marketing (Palich et al., 2000). However,
we found no relationship between multi-chaining and performance. It is possible that
managers sometimes establish additional chains when they see their existing chain
approaching market saturation, and these secondary chains fail to live up to the profit
potential of the existing chain. If so, the addition of multiple chains in these cases dilutes
profits. Looking to the future, additional research focused on better understand-ing the
reasons for and consequences of multi-chaining could not only increase under-standing of
this phenomenon but perhaps also help advance theory about diversification.
© 2010 The Authors

Management Studies
Limitations
The results of our study should be viewed in light of the limitations of our research
design; these limitations also serve to highlight future research opportunities. Whereas
the resource, strategy, and performance measures we studied are important, especially
in our focal industry, none of the three variable sets is exhaustive and our focus on
public firms in one industry limits generalizability. Although our results seem likely to
apply directly to public firms in other settings where franchising and multi-chaining
are common (e.g. hotels), studies involving multiple settings and among small and
medium-sized private firms will be required in order to establish that resource
flexibility matters for performance.
A number of limitations surface due to the data used to test our hypotheses. First, our
data only permit investigation of a one-year performance lag. This limited timeframe
highlights opportunities for future inquiry to examine the longer-term ramifications of
resource-expansion strategy decisions and the time firms need to realize the full value of
leveraged resources. Another data-related limitation of the present study is the age of the
data. We note, however, that franchising and multi-chaining remain widely used strat-egies
in the restaurant industry. Similarly, the resources upon which we focused (i.e. TMT
experience and specific knowledge) are important resources in this context that do not
appear to have changed in form over time, unlike technological resources that have
significantly changed the way restaurants are managed. Indeed, our comparison of
restaurant chains in 1995 and 2009 using Entrepreneur magazine’s data suggests that
whereas the absolute value of some factors are rising (especially costs), the relationships
among frequently studied variables remain relatively stable. Finally, as noted previously,
our small sample size can cause smaller effects to go undetected. The fact that we found
statistically significant results suggests that the effects of interest in this study were
moderate to large. However, the degree to which resources are flexible may vary, so future
research may benefit from the use of larger samples to determine smaller, yet practically
important levels of flexibility.
CONCLUSION
Understanding the relationship between resources and performance is a key task of
research grounded in RBT. This study takes an important step forward in RBT by
distinguishing between flexible and rigid strategic resources. In addition, our results
indicate that the firm’s pattern of market expansion strategies must align with the
characteristics of its strategic resources for the latter to obtain their full performance-
enhancing value. Our hope is that these findings can be integrated with future RBT
inquiry that further uncovers how different bundles of resources, strategies, and other
contingencies interact as researchers continue to strive towards a more complete under-
standing of the determinants of firm performance.
ACKNOWLEDGMENTS
An earlier version of this paper was presented during the conference ‘Trajectories, paths, patterns and
practices of strategizing and organizing’ sponsored by the Journal of Management Studies. We thank
Vera Street for her help gathering some of the data used in this project.
© 2010 The Authors

REFERENCES
Ahuja, G. and Lampert, C. M. (2001). ‘Entrepreneurship in the large corporation: a longitudinal study of
how established firms create breakthrough inventions’. Strategic Management Journal, 22, 521–43.
Aiken, L. S. and West, S. G. (1991). Multiple Regression: Testing and Interpreting Interactions . Newbury Park, CA:
Sage.
Ambrosini, V. and Bowman, C. (2001). ‘Tacit knowledge: some suggestions for operationalization’. Journal
of Management Studies, 38, 811–29.
Amit, R. and Schoemaker, P. (1993). ‘Strategic assets and organizational rent’. Strategic Management Journal,
14, 33–46.
Argyres, N. and McGahan, A. M. (2002). ‘An interview with Michael Porter’. Academy of Management Executive,
16, 43–52.
Audia, P. G., Locke, E. A. and Smith, K. G. (2000). ‘The paradox of success: an archival and a laboratory
study of strategic persistence following radical environmental change’. Academy of Management Journal, 43,
837–53.
Baker, T. D. and Nelson, R. E. (2005). ‘Creating something from nothing: resource construction from
entrepreneurial bricolage’. Administrative Science Quarterly, 50, 329–66.
Bantel, K. A. and Finkelstein, S. (1995). ‘The determinants of top management teams’. Advances in
Group Processes, 12, 139–65.
Barney, J. B. (1991). ‘Firm resources and sustained competitive advantage’. Journal of Management, 17, 99–120.
Barney, J. B. (1995). ‘Looking inside for competitive advantage’. Academy of Management Executive , 9, 49–61.
Barney, J. B. (2001). ‘Resource-based theories of competitive advantage: a ten-year retrospective on the
resource-based view’. Journal of Management, 27, 643–50.
Barney, J. B., Wright, M. and Ketchen, D. J., Jr (2001). ‘The resource-based view of the firm: ten years after
1991’. Journal of Management, 27, 625–41.
Baum, R. J. and Wally, S. (2003). ‘Strategic decision speed and firm performance’. Strategic Management
Journal, 24, 1107–18.
Bettis, R. and Prahalad, C. K. (1995). ‘The dominant logic: retrospective and extension’. Strategic Management
Journal, 16, 5–14.
Black, J. A. and Boal, K. B. (1994). ‘Strategic resources: traits, configurations and paths to sustainable
competitive advantage’. Strategic Management Journal, 15, 131–48.
Bradach, J. L. (1997). ‘Using the plural form in the management of restaurant chains’. Administrative Science
Quarterly, 42, 276–303.
Brickley, J. A. and Dark, F. H. (1987). ‘The choice of organizational form: the case of franchising’. Journal
of Financial Economics, 18, 401–20.
Brown, S. L. and Eisenhardt, K. M. (1997). ‘The art of continuous change: linking complexity theory and
time-paced evolution in relentlessly shifting organizations’. Administrative Science Quarterly , 42, 1–34.
Brush, T. H. and Artz, K. W. (1999). ‘Toward a contingent resource-based theory: the impact of information
asymmetry on the value of capabilities in veterinary medicine’. Strategic Management Journal, 20, 223–
50. Castanias, R. P. and Helfat, C. E. (1991). ‘Managerial resources and rents’. Journal of Management, 17,
155–72.
Chen, M., Farh, J. and MacMillan, I. C. (1993). ‘An exploration of the expertness of outside informants’.
Academy of Management Journal, 36, 1614–32.
Chi, T. (1994). ‘Trading in strategic resources: necessary conditions, transaction cost problems, and choice
of exchange structure’. Strategic Management Journal, 15, 271–90.
Coff, R. W. (2002). ‘Human capital, shared expertise, and the likelihood of impasse in corporate acquisi-
tions’. Journal of Management, 28, 107–28.
Cohen, J. (1977). Statistical Power Analysis for the Behavioral Sciences . New York: Lawrence Erlbaum Associates.
Combs, J. G. and Ketchen, D. J. (1999). ‘Explaining interfirm cooperation and performance: toward a
reconciliation of predictions from the resource-based view and organizational economics’. Strategic
Management Journal, 20, 867–88.
Combs, J. G. and Ketchen, D. J. (2003). ‘Why do firms use franchising as an entrepreneurial strategy? A
meta-analysis’. Journal of Management, 29, 443–65.
Combs, J. G., Michael, S. C. and Castrogionvanni, G. J. (2004). ‘Franchising: a review and avenues to
greater theoretical diversity’. Journal of Management, 30, 907–32.
Cornelissen, J. P., Haslam, S. A. and Balmer, J. M. T. (2007). ‘Social identity, organizational identity and
corporate identity: towards an integrated understanding of processes, patternings and products’. British
Journal of Management, 18, S1–S16.
© 2010 The Authors
Management Studies
Cote, L., Langley, A. and Pasquero, J. (1999). ‘Acquisition strategy and dominant logic in an engineering
firm’. Journal of Management Studies, 36, 919–52.
Crook, T. R., Ketchen, D. J., Combs, J. G. and Todd, S. Y. (2008). ‘Strategic resources and performance:
a meta-analysis’. Strategic Management Journal, 29, 1141–54.
D’Aveni, R. A., Ravenscraft, D. J. and Anderson, P. (2004). ‘From corporate strategy to business-level
advantage: relatedness as a resource congruence’. Managerial and Decision Economics, 25, 365–81.
Denison, D. R. and Mishra, A. K. (1995). ‘Toward a theory of organizational culture and effectiveness’.
Organization Science, 6, 204–24.
Dierickx, I. and Cool, K. (1989). ‘Asset stock accumulation and sustainability of competitive advantage’.
Management Science, 35, 1504–14.
Dillman, D. A. (1978). Mail and Telephone Surveys. New York: Wiley.
Eddleston, K. A., Kellermanns, F. W. and Sarathy, R. (2008). ‘Resource configuration in family firms:
linking resources, strategic planning and technological opportunities to performance’. Journal of Man-
agement Studies, 45, 26–50.
Edelman, L. F., Brush, C. G. and Manolova, T. (2005). ‘Co-alignment in the resource-performance
relationship: strategy as a mediator’. Journal of Business Venturing, 20, 359–83.
Frynas, J. G., Mellahi, K. and Pigman, G. A. (2006). ‘First mover advantages in international business and
firm-specific political resources’. Strategic Management Journal, 27, 321–45.
Gupta, A. K., Smith, K. G. and Shalley, C. E. (2006). ‘The interplay between exploration and exploitation’.
Academy of Management Journal, 49, 693–706.
Haleblian, J. and Finkelstein, S. (1999). ‘The influence of organizational acquisition experience on acquisition
performance: a behavioral learning perspective’. Administrative Science Quarterly, 44, 29–56.
Harris, D. and Helfat, C. (1997). ‘Specificity of CEO human capital and compensation’. Strategic
Harrison, J. S., Chang, E., Gauthier, C., Joerchel, T., Nevarez, J. and Wang, M. (2005). ‘Exporting a North American
concept to Asia: Starbucks in China’. Cornell Hotel and Restaurant Administration Quarterly, 46,
275–83.
Hawes, J. M. and Crittenden, W. F. (1984). ‘A taxonomy of competitive retailing strategies’. Strategic
Jensen, M. C. and Meckling, W. (1995). ‘Specific and general knowledge, and organizational structure’.
Journal of Applied Corporate Finance, 8, 4–18.
Jones, L. V. (1952). ‘Test of hypotheses: one-tailed vs. two-sided alternatives’. Psychological Bulletin, 49, 43–6.
Kaufmann, P. J. and Eroglu, S. (1999). ‘Standardization and adaptation in business format franchising’.
Journal of Business Venturing, 14, 69–85.
Ketchen, D. J., Ireland, R. D. and Snow, C. C. (2007). ‘Strategic entrepreneurship, collaborative
innovation, and wealth creation’. Strategic Entrepreneurship Journal, 1, 371–85.
Knott, A. M. (2003). ‘The organizational routines factor market paradox’. Strategic Management Journal, 24,
929–44.
Koteff, E. (2007). ‘NRA’s outgoing leader Anderson leaves a legacy of political clout’. Nation’s
Restaurant News, 41, 6.
Krueger, A. B. (1991). ‘Ownership, agency, and wages: an examination of franchising in the fast food
industry’. Quarterly Journal of Economics, 106, 75–101.
Lafontaine, F. (1992). ‘Agency theory and franchising: some empirical results’. RAND Journal of Economics, 23,
263–83.
Lafontaine, F. (1995). ‘A critical appraisal of data sources on franchising’. In Dant, R. P. and Kaufmann, P.
J. (Eds), Franchising: Contemporary Issues and Research. New York: Haworth.
Lafontaine, F. (1999). ‘Franchising versus corporate ownership: the effect on price dispersion’. Journal of
Business Venturing, 14, 17–34.
Lafontaine, F. and Shaw, K. L. (1998). ‘Franchising growth and franchisor entry and exit in the U.S. market:
myth and reality’. Journal of Business Venturing, 13, 95–112.
Lampel, J. and Shamsie, J. (2000). ‘Probing the unobtrusive link: dominant logic and the design of joint
ventures at General Electric’. Strategic Management Journal, 21, 593–602.
Lazzarini, S. G., Claro, D. P. and Mesquita, L. F. (2008). ‘Buyer-supplier and supplier-supplier alliances:
do they reinforce or undermine one another?’. Journal of Management Studies, 45, 561–84.
Lei, D., Hitt, M. A. and Bettis, R. (1996). ‘Dynamic core competences through meta-learning and strategic
context’. Journal of Management, 22, 549–70.
Leonard-Barton, D. (1992). ‘Core capabilities and core rigidities: a paradox in managing new product
development’. Strategic Management Journal, 13, 111–25.
© 2010 The Authors

Lippman, S. A. and Rumelt, R. P. (1982). ‘Uncertain imitability: an analysis of interfirm differences in
efficiency under competition’. Bell Journal of Economics, 13, 418–53.
Love, E. G. and Cebon, P. (2008). ‘Meanings on multiple levels: the influence of field-level and
organizational-level meaning systems on diffusion’. Journal of Management Studies, 45, 239–67.
Luby’s Cafeterias, I. (1993). 1993 Annual Report.
Marino, K. E. (1996). ‘Developing consensus on firm competencies and capabilities’. Academy of Management
Executive, 10, 40–51.
Maritan, C. A. and Brush, T. H. (2003). ‘Heterogeneity and transferring practices: implementing flow
manufacturing in multiple plants’. Strategic Management Journal, 24, 945–59.
Mazen, A. M. M., Hemmasi, M. and Lewis, M. F. (1987). ‘Assessment of statistical power in contemporary
strategy research’. Strategic Management Journal, 8, 403–10.
Meyer, A. D. (1991). ‘What is strategy’s distinctive competence?’. Journal of Management, 17, 821–33.
Michael, S. C. (2000). ‘The effect of organizational form on quality: the case of franchising’. Journal of
Economic Behavior and Organization, 43, 295–318.
Michael, S. C. and Combs, J. G. (2008). ‘Entrepreneurial failure: the case of franchisees’. Journal of Small
Business Management, 46, 73–90.
Michel, J. G. and Hambrick, D. C. (1992). ‘Diversification posture and top management team character-
istics’. Academy of Management Journal, 35, 9–37.
Miller, D. and Shamsie, J. (1996). ‘The resource-based view of the firm in two environments: the Hollywood
film studios from 1936 to 1965’. Academy of Management Journal, 39, 519–43.
Mitronen, L. and Moller, K. (2003). ‘Management of hybrid organisations: a case study in retailing’.
Industrial Marketing Management, 32, 419–29.
Nadkarni, S. and Narayanan, V. K. (2007). ‘Strategic schemas, strategic flexibility, and firm performance:
the moderating role of industry clockspeed’. Strategic Management Journal, 28, 243–70.
National Restaurant Association (2009). 2009 Restaurant Industry Pocket Factbook. Available at:
http:// www.restaurant.org/pdfs/research/2009Factbook.pdf (accessed 28 January 2010).
Nayyar, P. R. (1990). ‘Information asymmetries: a source of competitive advantage for diversified service
firms’. Strategic Management Journal, 11, 513–19.
Palich, L. E., Cardinal, L. B. and Miller, C. C. (2000). ‘Curvilinearity in the diversification-performance
linkage: an examination of over three decades of research’. Strategic Management Journal, 21, 155–74.
Penrose, E. T. (1959). The Theory of the Growth of the Firm. New York: John Wiley.
Peters, T. H. and Waterman, R. H. (1982). In Search of Excellence. New York: Harper & Row.
Prahalad, C. K. and Bettis, R. A. (1986). ‘The dominant logic: a new linkage between diversity and
performance’. Strategic Management Journal, 7, 485–501.
Prahalad, C. K. and Hamel, G. (1990). ‘The core competence of the corporation’. Harvard Business Review,
68, 79–91.
Priem, R. L. and Butler, J. E. (2001). ‘Is the resource-based “view” a useful perspective for strategic
management research?’. Academy of Management Review, 26, 22–40.
Reed, R. and DeFillippi, R. J. (1990). ‘Causal ambiguity, barriers to imitation, and sustainable competitive
advantage’. Academy of Management Review, 15, 88–102.
Robins, J. A. (2004). ‘When does the age of data matter? Notes on the selection of data for strategy research’.
In Ketchen, D. J. and Bergh, D. D. (Eds), Research Methodology in Strategy and Management, Vol. 1.
Greenwich, CT: Elsevier, 251–71.
Rugman, A. M. and Verbeke, A. (1992). ‘A note on the transactional solution and the transaction cost theory
of multinational strategic management’. Journal of International Business Studies, 23, 461–771. Rumelt, R. P.
(1974). Strategy, Structure, and Economic Performance. Cambridge, MA: Harvard University. Sambharya, R. B.
(1996). ‘Foreign experience of top management teams and international diversification
strategies of U.S. multinational corporations’. Strategic Management Journal, 17, 739–47.
Sanders, W. G. and Hambrick, D. C. (2008). ‘Swinging for the fences: the effects of CEO stock options on
company risk taking and performance’. Academy of Management Journal, 50, 1055–78.
Shane, S. A. (1996). ‘Hybrid organizational arrangements and their implications for firm growth and
survival: a study of new franchisors’. Academy of Management Journal, 39, 216–34.
Shen, W. (2003). ‘The dynamics of the CEO-board relationship: an evolutionary perspective’. Academy of
Management Review, 28, 466–76.
Sirmon, D. G., Hitt, M. A. and Ireland, R. D. (2007). ‘Managing firm resources in dynamic environments
to create value: looking inside the black box’. Academy of Management Review, 32, 273–92.
Spielberg, S. (2006). ‘Industry eyes 15th year of real growth in ’06: NRA predicts 5.1% jump in U.S.
foodservice sales, with 2.7% in menu price hikes’. Nation’s Restaurant News, 22, 37–8.
© 2010 The Authors
Management Studies
Szulanski, G. (1996). ‘Exploring internal stickiness: impediments to the transfer of best practice within the
firm’. Strategic Management Journal, 17, 27–43.
Teece, D. J., Pisano, G. and Shuen, A. (1997). ‘Dynamic capabilities and strategic management’.
Strategic Management Journal, 18, 509–33.
Wiggins, R. R. and Ruefli, T. (2002). ‘Sustained competitive advantage: temporal dynamics and the
incidence and persistence of superior economic performance’. Organization Science, 13, 82–105.
Williamson, O. E. (1991). ‘Comparative economic organization: the analysis of discrete structural alterna-
tives’. Administrative Science Quarterly, 36, 269–96.
Wolf, A. (2005). The Service Economy in OECD Countries. Paris: OECD Directorate for Science,
Technology and Industry.
Yin, X. and Zajac, E. J. (2004). ‘The strategy/governance structure fit relationship: theory and evidence in
franchising arrangements’. Strategic Management Journal, 25, 365–83.
Zollo, M. and Singh, H. (2004). ‘Deliberate learning in corporate acquisitions: post-acquisition strategies and
integration capability in U.S. bank mergers’. Strategic Management Journal, 25, 1233–56.
© 2010 The Authors
Journal of Business Research 58 (2005) 726 – 735
The entrepreneur’s business model: toward a unified perspective

a,* b c
Michael Morris , Minet Schindehutte , Jeffrey Allen
a b c
Witting Chair in Entrepreneurship, Syracuse University, Syracuse, NY 13244, USA Miami University, Oxford, OH 45056, USA University of Central
Florida, Orlando, FL 32816, USA
Received 29 September 2002; accepted 6 November 2003
Abstract
Highly emphasized in entrepreneurial practice, business models have received limited attention from researchers. No consensus exists
regarding the definition, nature, structure, and evolution of business models. Still, the business model holds promise as a unifying unit of
analysis that can facilitate theory development in entrepreneurship. This article synthesizes the literature and draws conclusions regarding
a number of these core issues. Theoretical underpinnings of a firm’s business model are explored. A six-component framework is proposed
for characterizing a business model, regardless of venture type. These components are applied at three different levels. The framework is
illustrated using a successful mainstream company. Suggestions are made regarding the manner in which business models might be
expected to emerge and evolve over time.
D 2003 Elsevier Inc. All rights reserved.
Keywords: Activity sets; Architecture; Business model; Strategy; Model dynamics
1. Introduction 2. Literature review
Ventures fail despite the presence of market opportu- 2.1. What is a ‘business model’?
nities, novel business ideas, adequate resources, and
talented entrepreneurs. A possible cause is the underlying No generally accepted definition of the term ‘‘business
model driving the business. Surprisingly, little attention has model’’ has emerged. Diversity in the available definitions
been given to business models by researchers, with much poses substantive challenges for delimiting the nature and
of the published work focusing on Internet-based models. components of a model and determining what constitutes a
The available research tends to be descriptive in nature, good model. It also leads to confusion in terminology, as
examining approaches to model construction, noting business model, strategy, business concept, revenue model,
standard model types, citing examples of failed and and economic model are often used interchangeably. More-
successful models, and discussing the need for new models over, the business model has been referred to as architecture,
as conditions change. Yet, no consensus exists regarding design, pattern, plan, method, assumption, and statement.
the definition or nature of a model, and there has been no It is possible to bring order to the various perspectives.
attempt to prioritize critical research ques-tions or establish A content analysis of key words in 30 definitions led the
research streams relating to models. The purpose of this authors to identify three general categories of definitions
study is to review existing perspec-tives and propose an based on their principal emphasis. These categories can be
integrative framework for charac-terizing the labeled economic, operational, and strategic, with each
entrepreneur’s business model. comprised of a unique set of decision variables. They
represent a hierarchy in that the perspective becomes more
comprehensive as one progressively moves from the
economic to the operational to the strategic levels.
At the most rudimentary level, the business model is
pp Corresponding author. Tel.: +1-315-443-
defined solely in terms of the firm’s economic model. The
3164. E-mail address: mhmorris@syr.edu (M.
Morris).
0148-2963/$ – see front matter D 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.jbusres.2003.11.001
M. Morris et al. / Journal of Business Research 58 (2005) 726–735 727
concern is with the logic of profit generation. Relevant business plan, but the plan deals with a number of start-up
decision variables include revenue sources, pricing method- and operational issues that transcend the model. It is not a
ologies, cost structures, margins, and expected volumes. strategy but includes a number of strategy elements. Simi-
Hence, Stewart and Zhao (2000) approach the model as ‘‘a larly, it is not an activity set, although activity sets support
statement of how a firm will make money and sustain its profit each element of a model.
stream over time.’’ At the operational level, the model
represents an architectural configuration. The focus is on 2.2. What do we know about business models?
internal processes and design of infrastructure that enables the
firm to create value. Decision variables include produc-tion or Interest in business models is relatively recent, with much
service delivery methods, administrative processes, resource of the research appearing in the past decade, a time period
flows, knowledge management, and logistical streams. Mayo associated with the ‘‘new economy.’’ The popularity of the
and Brown (1999) refer to ‘‘the design of key interdependent term is evidenced in a keyword search using the Google
systems that create and sustain a compet-itive business.’’ search engine and the ABI-Inform database. Results from
Definitions at the strategic level emphasize overall direction in these two sources indicated 4,326,812 and 2387 entries,
the firm’s market positioning, interactions across respectively, for ‘‘business model.’’
organizational boundaries, and growth opportunities. Of The largest volume of research has come from electronic
concern is competitive advantage and sustainability. Decision commerce (Mahadevan, 2000). Early work focused on
elements include stakeholder identification, value creation, capturing revenue streams for web-based firms. Subsequent
differentiation, vision, values, and networks and alliances. efforts identified model types based on product offerings,
Slywotsky (1996) refers to ‘‘the totality of how a company value-creating processes, and firm architecture, among other
selects its customers, defines and differentiates its offerings, variables. For a detailed inventory of these models, see
defines the tasks it will perform itself and those it will http://digitalenterprise.org/models/models.html. As it be-came
outsource, configures its resources, goes to market, creates evident that the number of potential models was limitless,
utility for customers and captures profits.’’ researchers began focusing on model taxonomies.
Among the available definitions, strategic elements are In spite of this foundation, progress in the field has been
most prominent. Further, an analysis of models frequently hindered by lack of consensus over the key components of a
cited as being well conceptualized (e.g., Dell, Nucor, Wal- model. Table 1 presents a synopsis of available perspectives
Mart, IKEA, Walgreen) suggests that the elements making regarding model components. The perspectives are notable
these models unique transcend the architecture of the firm both for their similarities and differences. The number of
or how it makes money. More than the sum of its parts, the components mentioned varies from four to eight. A total of 24
model captures the essence of how the business system will different items are mentioned as possible components, with 15
be focused. Accordingly, the following inte-grative receiving multiple mentions. The most frequently cited are the
definition is proposed: ‘‘A business model is a concise firm’s value offering (11), economic model (10), customer
representation of how an interrelated set of decision interface/relationship (8), partner network/ roles (7), internal
variables in the areas of venture strategy, archi-tecture, and infrastructure/connected activities (6), and target markets (5).
economics are addressed to create sustainable competitive Some items overlap, such as customer relationships and the
advantage in defined markets.’’ firm’s partner network or the firm’s revenue sources, products,
To illustrate the distinction between a business model and value offering.
and related concepts, consider Dell Computer, a firm that This lack of consensus has hindered progress on a
has grown to over US$32 billion in annual sales in just 20 number of related issues. Few insights are available
years. The company’s products include a mix of PCs, regarding the conditions that make a particular model
notebooks, workstations, servers, and software products. appropriate, ways in which models interact with organiza-
Their business concept involves selling customized com- tional variables, existence of generic model types, and
puter solutions directly to customers at competitive pri-ces. dynamics of model evolution, among other questions.
However, the Dell business model integrates strategic Attempts at model decomposition acknowledge the exis-
considerations, operational processes, and decisions relat- tence of interdependencies among components but shed
ed to economics. It is designed around elimination of little light on the nature of the relationships. Limited
intermediaries, systems built to order, highly responsive progress has also been made in establishing methodologies
customer service, moderate margins, rapid inventory turn- for evaluating model quality.
over, speedy integration of new technologies, and a highly
efficient procurement, manufacturing, and distribu-tion 2.3. Theoretical underpinnings of business models
process. Adherence to these elements guides opera-tional
decision making and the firm’s ongoing strategic direction. Issues of theory represent another area receiving scant
attention. A notable exception can be found in Amit and
The business model is related to a number of other Zott (2001), who approach the business model construct as
managerial concepts. It captures key components of a a unifying unit of analysis that captures value creation
728 M. Morris et al. / Journal of Business Research 58 (2005) 726–735
Table 1
Perspectives on business model components
Source Specific components Number E-commerce/ Empirical Nature of data
general support (Y/N)
Horowitz (1996) Price, product, distribution, organizational 5 G N
characteristics, and technology
Viscio and Pasternak (1996) Global core, governance, business units, services, 5 G N
and linkages
Timmers (1998) Product/service/information flow architecture, 5 E Y Detailed case
business actors and roles, actor benefits, revenue studies
sources, and marketing strategy
Markides (1999) Product innovation, customer relationship, infrastructure 4 G N
management, and financial aspects
Donath (1999) Customer understanding, marketing tactics, corporate 5 E N
governance, and intranet/extranet capabilities
Gordijn et al. (2001) Actors, market segments, value offering, value activity, 8 E N
stakeholder network, value interfaces, value ports,
and value exchanges
Linder and Cantrell (2001) Pricing model, revenue model, channel model, 8 G Y 70 interviews
commerce process model, Internet-enabled commerce with CEOs
relationship, organizational form, and value proposition
Chesbrough and Value proposition, target markets, internal value chain 6 G Y 35 case studies
Rosenbaum (2000) structure, cost structure and profit model, value
network, and competitive strategy
Gartner (2003) Market offering, competencies, core technology 4 E N Consulting
investments, and bottom line clients
Hamel (2001) Core strategy, strategic resources, value network, 4 G N Consulting clients
and customer interface
Petrovic et al. (2001) Value model, resource model, production model, 7 E N
customer relations model, revenue model, capital model,
and market model
Dubosson-Torbay Products, customer relationship, infrastructure 4 E Y Detailed
et al. (2001) and network of partners, and financial aspects case studies
Afuah and Tucci (2001) Customer value, scope, price, revenue, connected 8 E N
activities, implementation, capabilities, and sustainability
Weill and Vitale (2001) Strategic objectives, value proposition, 8 E Y Survey research
revenue sources, success factors, channels, core
competencies, customer segments, and IT infrastructure
Applegate (2001) Concept, capabilities, and value 3 G N
Amit and Zott (2001) Transaction content, transaction structure, 4 E Y 59 case studies
and transaction governance
Alt and Zimmerman (2001) Mission, structure, processes, revenues,legalities, 6 E N Literature
and technology synthesis
Rayport and Value cluster, market space offering, resource system, 4 E Y 100 cases
Jaworski (2001) and financial model
Betz (2002) Resources, sales, profits, and capital 4 G N
arising from multiple sources. They argue for a cross- (Barney, 1999) and relates to transaction cost economics
theoretical perspective, concluding that no single theory (Williamson, 1981).
can fully explain the value creation potential of a venture. Most perspectives on models include the firm’s offer-
The business model construct builds upon central ideas ings and activities undertaken to produce them. Here,
in business strategy and its associated theoretical traditions. management must consider the firm’s value proposition,
Most directly, it builds upon the value chain concept choose the activities it will undertake within the firm, and
(Porter, 1985) and the extended notions of value systems determine how the firm fits into the value creation net-
and strategic positioning (Porter, 1996). Because the busi- work. Based on Schumpeter’s (1936) theory of economic
ness model encompasses competitive advantage, it also development, value is created from unique combinations
draws on resource-based theory (Barney et al., 2001). In of resources that produce innovations, while transaction
terms of the firm’s fit within the larger value creation cost economics identifies transaction efficiency and
network, the model relates to strategic network theory boundary decisions as a value source. Positioning within
(Jarillo, 1995) and cooperative strategies (Dyer and Singh, the larger value network can be a critical factor in value
1998). Further, the model involves choices (e.g., vertical creation. As part of its positioning, the firm must establish
integration, competitive strategy) about firm boundaries appropriate relationships with suppliers, partners, and
customers.
Models implicitly or explicitly address the internal com- dation level, a need to make generic decisions regarding
petencies that underlie a firm’s competitive advantage. This what the business is and is not and ensure such decisions
is consistent with resource-based theory, where the firm is are internally consistent. Because the foundation level
viewed as a bundle of resources and capabilities (Barney et addresses basic decisions that all entrepreneurs must make,
al., 2001). Competitive advantage can emerge from it permits general comparisons across ventures and the
superior execution of particular activities within the firm’s identification of universal models. At the propri-etary level,
internal value chain, superior coordination among those the model’s purpose is to enable development of unique
activities, or superior management of the interface between combinations among decision variables that result in
the firm and others in the value network. Also, where the marketplace advantage. At this level, the frame-work
model has proprietary innovative elements, resource becomes a customizable tool that encourages the
advantage theory holds relevance (Hunt, 2000). entrepreneur to focus on how value can be created in each
The economics of the venture is featured prominently in of the six decision areas. The usefulness of any model is
business model research. An effective model encom-passes limited, however, unless it provides specific guidance and
unique combinations that result in superior value creation, discipline to business operations, necessi-tating a third
in turn producing superior returns to the firm, consistent level in the model. The rules level delin-eates guiding
with Schumpeterian theory (Schumpeter, 1936). At the principles governing execution of decisions made at levels
same time, the growth and profit aspirations of one and two.
entrepreneurs vary considerably. Aspirations reflect the
firm’s relationship to the entrepreneur’s career and life and 3.1. Foundation level: defining basic components
influence enterprise objectives. Business models will differ
for ventures with more moderate versus more ambitious At its essence, a well-formulated model must address
aspirations. Various theoretical traditions have implications six key questions (see Table 2). These questions have been
for entrepreneurial intentions regarding the nature and derived based on commonalities among the various per-
scope of the venture. Self-efficacy theory is a case in point, spectives found in the literature, including those summa-
with its emphasis on role of an entrepre-neur’s cognitive rized in Table 1. Moreover, each has a foundation in the
capabilities and skills assessment in determining outcomes. theoretical work discussed earlier. The most consistently
Alternatively, the theory of effec-tuation suggests that emphasized components concern the value proposition, the
entrepreneurs make conjectures about the future, determine customer, internal processes and competencies, and how
what can be done, and goals emerge over time (Wiltbank the firm makes money. To these four, a competitive
and Sarasvathy, 2002). strategy element has been added, reflecting the need to
An additional theoretical perspective approaches the translate core competencies and the value proposition into
business model as interrelated components of a system that a sustainable marketplace position. Finally, a useable
constitutes the firm’s architectural backbone. With systems framework should apply to all types of ventures, reflecting
theory, the business is viewed as an open system with the design consid-erations necessary to accommodate
varying levels of combinatorial complexity among subsys- differing levels of growth, time horizons, resource
tems and bounded by the environment and open strategies, and exit vehicles. Thus, the sixth decision area
information exchange (Petrovic et al., 2001). captures growth and time objectives of the entrepreneur.
Let us examine each in more detail.
3. Model development: an integrative framework 3.1.1. How will the firm create value?
This first question concerns the value offering of the
Building on these conceptual and theoretical roots, it is firm. Decisions here address the nature of the product/
possible to develop a standard framework for service mix, the firm’s role in production or service
characterizing a business model. To be useful, such a delivery, and how the offering is made available to
framework must be reasonably simple, logical, measurable, customers. There is no business without a defined value
comprehensive, and operationally meaningful. In seeking proposition, and the creation of value provides a justifi-
generalizability, the extant perspectives tend to cation for the business entity. Its inclusion in the model is
oversimplify a firm’s model. The challenge is to produce a supported by the work of Afuah and Tucci (2001),
framework that is applicable to firms in general but which Chesbrough and Rosenbaum (2002), and Rayport and
serves the needs of the individual entrepreneur. Jaworski (2001), among others.
Accordingly, a framework is proposed that consists of
three increasingly specific levels of decision making, 3.1.2. For whom will the firm create value?
termed the ‘foundation’, ‘proprietary,’ and ‘rules’ levels. This question focuses on the nature and scope of the
Further, at each level, six basic decision areas are market in which the firm competes. To whom will the firm
considered. The need for three levels reflects the different sell and where in the value chain will it operate? Customer
managerial purposes of a model. There is, at the foun- types, their geographic dispersion, and their interaction
.
.
.
.
.
.
.
speculative model
Table 2
Six questions that underlie a business model
Component 1 (factors related to the offering): How do we create value? requirements have significant impacts on how an organiza-
(select from each set) tion is configured, its resource requirements, and what it
offering: primarily products/primarily services/heavy mix
sells. Failure to adequately define the market is a key factor
offering: standardized/some customization/high customization
offering: broad line/medium breadth/narrow line associated with venture failure. Support for the role of
offering: deep lines/medium depth/shallow lines customer considerations in delineating a firm’s model can
offering: access to product/ product itself/ product bundled with other be found in Gordijn et al., 2001, Markides, 1999, and
firm’s product Timmers, 1998.
offering: internal manufacturing or service delivery/ outsourcing/
licensing/ reselling/ value added reselling
offering: direct distribution/indirect distribution (if indirect: single or
multichannel)
Component 2 (market factors): Who do we create value for? (select from

each set)
. .
type of organization: b-to-b/b-to-c/ both
local/regional/national/international
.
where customer is in value chain: upstream supplier/ downstream
supplier/ government/ institutional/ wholesaler/ retailer/ service
provider/ final consumer
. .
broad or general market/multiple segment/niche market
transactional/relational
Component 3 (internal capability factors): What is our source of
competence? (select one or more)
.
production/operating systems
.
selling/marketing
.
information management/mining/packaging
.
technology/R&D/creative or innovative capability/intellectual
.
financial transactions/arbitrage
.
supply chain management
.
networking/resource leveraging
Component 4 (competitive strategy factors): How do we

competitively position ourselves? (select one or more)
.
image of operational excellence/consistency/dependability/speed
.
product or service quality/selection/features/availability
.
innovation leadership
.
low cost/efficiency
.
intimate customer relationship/experience
Component 5 (economic factors): How we make money? (select from

each set)
.
pricing and revenue sources: fixed/mixed/flexible
.
operating leverage: high/medium/low
.
volumes: high/medium/low
.
margins: high/medium/low
Component 6 (personal/investor factors): What are our time, scope, and

size ambitions? (select one)
.
subsistence model
.
income model
.
growth model
3.1.3. What is the firm’s internal source of advantage? The
term ‘core competency’ is used to capture an internal
capability or skill set that the firm performs relatively better
than others (Hamel, 2001). Hence, Federal Express delivers a
benefit of on-time delivery based on its competency at
logistics management, and the organization is configured
around this competency. Development and enhancement of
this competency solidify the firm’s role in the external value
chain and become the focus for the internal value chain. These
competencies lie at the heart of the business model
(Applegate, 2001; Viscio and Pasternack, 1996) . A firm can
attempt to build advantage around one or more competen-cies,
with general sources of advantage identified by various
observers (e.g., Siggelkow, 2002).
3.1.4. How will the firm position itself in the marketplace?

Core internal competencies provide the basis for exter-nal
positioning. The model must delineate how the entre-
preneur intends to achieve advantage over competitors
(Amit and Zott, 2001). The challenge is to identify salient
points of difference that can be maintained. The entrepre-
neur attempts to define a unique, defensible niche enabling
the firm to mitigate ongoing developments in the environ-
ment. Given the ability of firms to quickly imitate one
another, the entrepreneur seeks a positioning basis that is
more than transitory.
3.1.5. How will the firm make money?

A core element of the firm’s business model is its
economic model (Linder and Cantrell, 2000). The eco-
nomic model provides a consistent logic for earning
profits. It can be approached in terms of four subcompo-
nents: operating leverage or the extent to which the cost
structure is dominated by fixed versus variable costs; the
firm’s emphasis on higher or lower volumes in terms of
both the market opportunity and internal capacity; the
firm’s ability to achieve relatively higher or lower margins;
and the firm’s revenue model, including the flexibility of
revenue sources and prices.
3.1.6. What are the entrepreneur’s time, scope, and size

ambitions?
Entrepreneurs create different types of ventures,
ranging from lifestyle firms to rapid growth companies.
Differences among venture types have important
implications for com-petitive strategy, firm architecture,
resource management, creation of internal competencies,
and economic perfor-mance. As such, an integrated
business model must capture the entrepreneur’s time,
scope, and size ambitions or what might be termed the
firm’s ‘investment model.’ Examples of four such models
are subsistence, income, growth, and speculation. With
the subsistence model, the goal is to survive and meet
basic financial obligations. When employing an income
model, the entrepreneur invests to the point that the
business is able to generate on ongoing and stable income
stream for the principals. A growth model finds significant
initial investment, but also substantial reinvestment in an to a particular venture. Where the foundation level is
attempt to grow the value of the firm to the point that it generic, the proprietary level becomes strategy specific.
eventually generates a major capital gain for investors. In a The foundation level model is fairly easy to replicate by
speculative model, the entrepreneur’s time frame is shorter competitors; the proprietary level is not. Replication is
and the objective is to demonstrate venture potential before especially difficult because of interactions among the
selling out. proprietary-level components. In the earlier Dell Comput-
er example, the ‘Dell Direct Method’ results from pro-
3.2. Proprietary level: creating unique combinations prietary approaches to defining the value proposition and
organizing internal logistical flows.
While the foundation level is adequate to capture the
essence of a model for many firms, sustainable advantage 3.3. Rules level: establishing guiding principles
ultimately depends on the ability of the entrepreneur to
apply unique approaches to one or more of the founda-tion Once implemented, a model’s success can be tied to a basic
components. Having determined that the firm will sell set of operating rules. These guidelines ensure that the model’s
some combination of goods directly to businesses or will foundation and proprietary elements are reflected in ongoing
sell in consumer markets at high margins and low volumes, strategic actions. Eisenhardt and Sull (2001) discuss the
the entrepreneur identifies novel ways to ap-proach such concept of ‘‘strategy as simple rules’’ (see also Nelson and
decisions. This is referred to as the propri-etary level of the Winter, 1982). They discuss ‘‘priority rules’’ that determine
model, as it entails innovation unique how Intel allocates manufacturing capacity and ‘‘boundary
Table 3
Characterizing the business model of Southwest Airlines
Foundation level Proprietary level Rules
Component 1: Sell services only Short haul, low-fare, high-frequency, Maximum one-way fare should
___
Factors related Standardized offering point-to-point service not exceed US$
to offering Narrow breadth Deliver fun Maximum food cost per person
Shallow lines Serve only drinks/snacks should be less than US$___
Sell the service by itself Assign no seats/no first class
Internal service delivery Do not use travel agents/intermediaries
Direct distribution Fully refundable fares, no advance
purchase requirement
Component 2: B2C and B2B (sell to individual travelers Managed evolution from regional airline Specific guidelines for selecting
Market factors and corporate travel departments) to servicing to 59 airports in 30 states cities to be serviced
National Careful selection of cities based on fit 85% penetration of local markets
Retail with underlying operating model
Broad market
Transactional
Component 3: Production/operating systems Highly selective hiring of employees At least 20 departures per day
Internal capability that fit profile; intense focus on from airport
factors frontline employees Maximum flight distance
Do not operate a hub and-spoke route system.
___
should be less than miles
Fly into uncongested airports of small cities, Maximum flight time should be
less congested airports of large cities ___
less than minutes
Innovative ground operations approach Turnaround of flights should be
Independent baggage handling system 20 minutes or fewer
Use of Boeing 737 aircraft
No code sharing with other airlines
Component 4: Image of operational excellence/ Differentiation is achieved by Achieve best on-time record
Competitive strategy consistency/dependability stressing on-time arrival, low fares, in industry
factors passengers having a good time (spirit of fun)
Airline that love built
Component 5: Fixed revenue source Short-haul routes and high frequency of Maintain cost per passenger
___
Economic factors High operating leverage flights combined with consistently low prices mile below US$
High volumes and internal efficiencies result in annual
Low margins profitability regardless of industry trends
Component 6: Growth model Emphasis on growth opportunities Managed rate of growth

Growth/exit factors that are consistent with business model
rules’’ that govern the types of movies Miramax decides to ‘‘fit,’’ where the former is concerned with a coherent
make. Girotto and Rivkin (2000) explain how Yahoo! adheres configuration of key activities within the firm and the latter
to a set of guiding rules in the formation of partnerships, a addresses the appropriateness of the configuration given
critical part of the firm’s business model. At Dell Computer, a external environmental conditions.
rule might involve turning inventory in 4 days or less. Rules Internal fit includes both consistency and reinforce-ment
are important at the level of execution of the business model. within and between the six subcomponents of the model.
Consistent adherence to basic principles can distinguish two An economic model with high operating leverage, low
companies having otherwise similar models. margins, moderate volumes, and fixed revenue sour-ces
may, by itself, be untenable. Further, the economics must
3.4. Applying the framework fit with the other components of the model. A given
economic model might not be workable when selling in a
Southwest Airlines has a robust business model that has regional b-to-b market where significant investment in
sustained company growth for 30 years, including the customer relationships is required or when selling a value
aftermath of the 9/11 terrorist tragedy that devastated the offering involving extensive customiza-tion. If the
industry. Not surprisingly, the Southwest model has been economic model calls for penetration pricing with low
copied in whole or part by others (e.g., JetBlue, RyanAir, margins and high fixed costs, this may imply a value
United Express). While some have achieved achieve note- proposition that centers on medium to low quality, a target
worthy performance, none of these firms has achieved the market that is fairly broad and relatively price elastic,
level of success as Southwest, especially in head-to-head competitive positioning based on cost leadership, and a
competition with the firm. Southwest’s superiority in growth-oriented investment model.
exploiting this model also makes it clear that a well- Ultimately, each component affects and is affected by
conceptualized business model affects and is affected by the other components. While each is vital, the firm’s
such organizational variables as culture and leadership investment model effectively delimits decisions made in all
quality. the other areas. For instance, a speculative business, with
In Table 3, the Southwest model is first captured at the its shorter time horizon, may require a cost structure with
foundation level. Here, the focus is on what the firm is doing, lower operating leverage and a customer focus that is not
as opposed to how. This level is concerned with basics of the predicated on long-term customer relationships.
firm’s approach in terms of a standardized set of questions. Alternatively, if one is building a lifestyle business, the
Next, at the proprietary level, Southwest’s model reflects firm is apt to have a more narrowly defined product and
innovation that has changed the ways in which other airlines market focus, may be more dependent on customer
operate, while reflecting an approach that is difficult to relationships, and is likely to require an economic model
replicate. From Table 3, it can be seen how the model that includes lower volumes. With the lifestyle venture, it
components are exploited for advantage in an innovative yet may not be necessary to invest as much in the model’s
internally consistent manner. The proprietary model centers on proprietary elements.
Southwest’s core competency, its unique operating system. External fit is concerned with consistency between
This operating system (e.g., employee policies, airport and choices in the six areas of the model and conditions in the
route selection, no code sharing, independent baggage external environment. As environmental conditions change,
handling, standardization of aircraft) makes possible a unique the model may require adaptation or wholesale change.
value proposition (short haul, low-fare, direct service that is Rindova and Kotha (2001) describe the ‘‘morph-ing’’ of
on-time and ‘fun’). Finally, it would be easy to deviate from Yahoo’s business model from provider of search functions
this model given competitive and regulatory pressures. to supplier of content to source of interactive services.
However, a number of ‘rules’ help management avoid When confronting highly turbulent conditions, a strong
strategic or tactical moves that are inconsistent with the internal fit can undermine the firm’s adaptability in the face
model. Rules regarding maximum fares or flight turnaround of a poor external fit. Companies must work to disrupt their
times effectively delimit appro-priate courses of management own advantages and those of competitors. Adaptability
action, while reinforcing the strategic intent of the firm in the may require models with loosely fitting elements or
minds of employees. introduction of new elements that change the dynamics
among existing elements.
4. Business model fit, evolution, sustainability 4.2. Emergence and evolution of models
4.1. The issue of fit Although some entrepreneurs have a clearly formulat-
ed model when undertaking a venture, many start with
Sustainability requires that model components demon- partially formed models and incomplete strategies. A
strate consistency, as in the Southwest example. Consisten- process of experimentation may be involved as the model
cy can be described in terms of both internal and external emerges (and a viable model may never emerge). Lessons
are being learned regarding what is required to make ent activities than competitors or about performing similar
money on a sustainable basis. As competencies are activities in different ways. He juxtaposes strategy against
developed, keener insights may result regarding sources of operational effectiveness, a concern with performing similar
innovation or advantage as they relate to those activities better than competitors. The business model has
competencies. The entrepreneur is also likely to become elements of both strategy and operational effectiveness. For
more strategic in his/her view of business operations over instance, a low-cost advantage deriving from a novel ap-
time. proach to distribution might be central to the way in which the
In terms of the proposed framework, a firm’s model firm creates value, reflecting Porter’s (1996) notion of strat-
might be expected to evolve from the foundation level egy. Similarly, the model might call for internal manufactur-
toward a more complete articulation of the proprietary and ing, where production processes are fairly similar to those of
rules levels. Initially, the entrepreneur may have a fair competitors and the firm’s competitiveness in this area is a
picture of the foundation level and limited notions about function of operational effectiveness.
some components at the proprietary level. As the firm Central to Porter’s (1996) recent work is the concept of
develops and learns, it is able to flesh out more compo- ‘‘activity sets.’’ Organizations configure activities in unique
nents at the proprietary level, furthering its advantage, and ways, with advantage deriving from how activities fit with
develops rules that guide operations and ongoing growth. and reinforce one another. Activity systems can be mapped
Model evolution can also be linked to the type of venture so as to capture the evolution of organizations along
being pursued. Models for survival, lifestyle, growth, and discernable developmental paths. Siggelkow (2002)
speculative ventures might be expected to vary in formal- characterizes activity sets in terms of core elements,
ity, sophistication, and uniqueness. For instance, the pro- elaborating elements, and interactions. He notes the
prietor of a lifestyle business may have an implicit model emergence over time of seven core elements in his study of
in mind at start-up, and the model may never develop the Vanguard Group. Implied in this work is a large
beyond basic decisions at the foundation level. He/she universe of potential core elements from which a subset is
might develop a set of rules of thumb that support the basic created and elaborated upon as a firm evolves.
model, such as how much inventory must move at certain The business model organizes these core elements and
times of the year. This entrepreneur may periodi-cally activities around six key decision areas. The model
deviate from the model, introducing elements that are captures all of a firm’s core elements, although more than
inconsistent with existing elements. Alternatively, a more one core element can fall into a given decision area.
formal, comprehensive, and potent model is needed to Further, each of the six decision areas and the interac-tions
provide direction and attract resources to a high growth between areas are supported by a variety of activity sets.
venture. Decisions at the proprietary level become vital for Unfortunately, the mapping referred to by Porter and
sustainable advantage. Siggelkow (2001) occurs after the fact. The business model
Conceptually, it is possible to envision a business model represents a framework for doing this constructing in the
life cycle involving periods of specification, refinement, early stages of a venture and for conducting predictive,
adaptation, revision, and reformulation. An initial period what-if scenario analysis. For early stage entrepreneurs,
during which the model is fairly informal or implicit is many of the potentially most productive activity sets are
followed by a process of trial and error, and a number of core less apparent, as the firm has little experience, highlighting
decisions are made that delimit the directions in which the the importance of entrepreneurial vision.
firm can evolve. At some point, a fairly definitive, formal
model is in place. Subsequently, adjustments are made and The business model encourages the entrepreneur to (a)
ongoing experiments are undertaken. Siggelkow (2002) conceptualize the venture as an interrelated set of strategic
characterized such adjustments in terms of augmentation, choices; (b) seek complementary relationships among ele-
reinforcement, and deletion. A basically sound model will ments through unique combinations; (c) develop activity
typically withstand economic downturns and modest distur- sets around a logical framework; and (d) ensure consisten-
bances but can become dysfunctional if major discontinuities cy between elements of strategy, architecture, economics,
occur. When external changes undermine a model, it typically growth, and exit intentions. Strategic choices that charac-
cannot be recalibrated; a new model must be constructed. terize a venture are made both intentionally and by default.
Hence, Grove (1997) describes ‘‘strategic inflection points’’ in The business model makes the choices explicit. The model
the respecification of Intel’s model over time. is a relatively simple way to delimit and organize key
decisions that must be made at the outset of a venture. At
4.3. Linking the business model to strategic management the foundation level, the model provides a framework for
concepts deciding what not to do (e.g., what services not to offer)
and assists the entrepreneur in assessing consistencies and
The business model is consistent with a number of recognizing trade-offs among decisions. At the proprietary
emerging concepts from the field of strategic management. level, truly unique configurations are produced that can
Strategy, in Porter’s (1996) view, is about performing differ- result in sustainable advantage.
5. Conclusions needed. Methods are also needed for appraising the model’s fit
with changing environmental conditions; just as critical are
The business model can be a central construct in entre- issues surrounding model implementation. One chal-lenge
preneurship research. This article has sought to provide concerns the translation of model components into operational
direction in addressing some of the more vexing questions decisions, where the importance of fit will likely differ by
surrounding models. The model represents a strategic activity area. Another challenge involves exper-imenting with
frame-work for conceptualizing a value-based venture. The new strategic moves in ways that do not compromise the
pro-posed framework allows the user to design, describe, model. Finally, further insights are needed into the dynamics
categorize, critique, and analyze a business model for any of model emergence and evolution.
type of company. It provides a useful backdrop for strate-
gically adapting fundamental elements of a business. By
specifying the elements that constitute a model, the frame-
References
work enhances the ability to assess model attributes. A
model that ignores one or more of the specified
Afuah A, Tucci CL. Internet business models. New York: McGraw-Hill/
components will suffer in terms of its comprehensiveness, Irwin, 2001.
while incon-sistency can manifest itself both in terms of Alt R, Zimmerman HD. Introduction to special section on business models.
the fit among decision areas within a given component as Electron Mark 2001;11(1):3 – 9.
well as the fit between components. Amit R, Zott C. Value creation in e-business. Strateg Manage J 2001;
22(2):493 – 520.
With the proposed framework, each of six components is Applegate LM. Emerging e-business models. Harvard Bus Rev 2001;
evaluated at three levels. At the foundation level, the model is 79(1):79 – 87.
defined in terms of a standardized set of decisions that can be Barney JB. How firm capabilities affect boundary decisions. Sloan
quantified. A benefit of this standardization is the ability to Manage Rev 1999;40(1):19 – 32.
make comparisons across models from a broad universe of Barney J, Wright M, Ketchen D. The resource-based view of the firm: ten
years after. J Manage 2001;27(6):625 – 41.
ventures. New avenues for empirical research become Betz F. Strategic business models. Eng Manag J 2002;14(1):21 – 7.
possible, ranging from the creation of general model taxon- Chesbrough H, Rosenbaum RS. The role of the business model in capturing
omies and investigations of relationships among the foun- value from innovation (working paper). Boston: Harvard Business
dation level variables to modeling relationships between the School, 2000.
model and a host of endogenous and exogenous variables. Donath R. Taming e-business models. ISBM business marketing web con-
sortium 3 (1). State College (PA): Institute for the Study of Business
Also promising is the ability to identify model archetypes. By Markets, 1999; 1 – 24.
applying codes to the various decision choices across the six Dubosson-Torbay M, Osterwalder A, Pigneur Y. E-business model design,
components of the framework at the foundation level, classification and measurements. Thunderbird Int Bus Rev 2001;44(1):
statistical tools can be used to identify dominant patterns 5–23.
among decision choices. Dyer J, Singh H. The relational view: cooperative strategy and sources of
interorganizational competitive advantage. Acad Manage Rev 1998;
At the proprietary level, considerable scope for inno- 23(4):660 – 79.
vation exists within each model component. The model Eisenhardt KM, Sull D. Strategy as simple rules. Harvard Bus Rev 2001;
becomes a form of intellectual property, with some 79(1):107 – 16.
entrepreneurs actually obtaining patents for their models. Gartner. Available at: http://www3.gartner.com, 2003.
Considerable work remains to properly understand busi- Girotto J, Rivkin J. Yahoo! Case 9-700-013. Boston: Harvard Business
School Press, 2000.
ness model innovation. A beginning point is to develop Gordijn J, Akkermans J, Van Vliet J. Designing and evaluating e-business
measures of model innovativeness. A further step involves models. IEEE Intell Syst 2001;16(4):11 – 7.
determining how the relative importance of component Grove A. Only the paranoid survive. New York: HarperCollins Business,
innovation varies depending on industry or market 1997.
characteristics. Hamel G. Leading the revolution. Boston (MA): Harvard Business School
Press, 2001.
The business model can serve as a focusing device for Horowitz AS. The real value of VARS: resellers lead a movement to a new
entrepreneurs and employees, especially when supported service and support. Mark Comput 1996;16(4):31 – 6.
by a set or rules or guidelines that derive from decisions Hunt SD. A general theory of competition. Thousand Oaks (CA): Sage
made at the proprietary level. Rules provide a clearer sense Publications, 2000.
Jarillo JC. Strategic networks. Oxford: Butterworth-Heinemann, 1995.
of the firm’s value proposition and are a source of guidance
Linder JC, Cantrell S. Changing business models. Chicago: Institute for
regarding actions that might compromise the value Strategic Change, Accenture, 2000.
equation. Additional research is needed regarding what Mahadevan B. Business models for Internet-based e-commerce. Calif
constitutes a rule, types of rules, differences between rules Man-age Rev 2000;42(4):55 – 69.
and objec-tives, qualities of good rules, and ways in which Markides C. A dynamic view of strategy. Sloan Manage Rev 1999;40(3):
55 – 63.
rules become dysfunctional.
Mayo MC, Brown GS. Building a competitive business model. Ivey Bus J
Other areas requiring further investigation include the 1999;63(3):18 – 23.
ability of entrepreneurs and others to assess model quality. Nelson R, Winter SG. The Schumpeterian tradeoff. Am Econ Rev 1982;
Systematic approaches for assessing model viability are 72(1):114 – 33.
Petrovic O, Kittl C, Teksten RD. Developing business models for e-busi-ness. Vienna: International Electronic Commerce Conference, 2001 [31.10.2001-
4.11.2001].
Porter ME. Competitive advantage. New York: Free Press, 1985.
Porter ME. What is strategy. Harvard Bus Rev 1996;74(6):61 – 78.
Porter ME, Siggelkow N. Contextuality within activity systems. In: Nagao D, editor. Research in best paper proceedings. Washington (DC): Acad-emy of
Management, 2001. p. 1 – 7.
Rayport JF, Jaworski BJ. E-commerce. New York: McGraw-Hill/Irwin, 2001.
Rindova VP, Kotha S. Continuous morphing: competing through dynamic capabilities. Acad Manage J 2001;44(6):1280 – 3.
Schumpeter J. Theory of economic development. Cambridge (MA): Har-vard University, 1936.
Siggelkow N. Evolution towards fit. Adm Sci Q 2002;47(1):125 – 59. Slywotzky AJ. Value migration. Boston (MA): Harvard Business Review
Press, 1996.
Stewart DW, Zhao Q. Internet marketing, business models, and public policy. J Public Policy Mark 2000;19(Fall):287 – 96.
Timmers P. Business models for electronic markets. Electron Commer Europe 1998;8(April):1 – 6.
Viscio AJ, Pasternack BA. Toward a new business model. Strateg Bus 1996;2(1):125 – 34.
Weill P, Vitale MR. Place to space. Boston: Harvard Business School Press, 2001.
Williamson OE. The economics of organization: the transaction cost ap-proach. Am J Sociol 1981;87(4):548 – 77.
Wiltbank R, Sarasvathy S. Patterns of business model change as measures of causation and effectuation. In: Reynolds P, editor. Research in fron-tiers of
entrepreneurship research. Babson Park (MA): Babson College Press, 2002. p. 328.
The impact of standardized innovation management systems on innovation

capability and business performance: An empirical study
Article in Journal of Engineering and Technology Management · June 2016
DOI: 10.1016/j.jengtecman.2016.06.002
CITATIONS READS
18 958
3 authors:
Moises Mir Marti Casadesus

Universitat de Girona Universitat de Girona
12 PUBLICATIONS 65 CITATIONS 145 PUBLICATIONS 3,526 CITATIONS
Luc Honore Petnji Yaya

Universitat Internacional de Catalunya
23 PUBLICATIONS 487 CITATIONS
SEE PROFILE
Universidad de Girona View project
El deleite de los clientes en servicios: Customer delight in services View project
All content following this page was uploaded by Marti Casadesus on 23 October 2017.

J. Eng. Technol. Manage. 41 (2016) 26–44
Journal of Engineering and Technology

Management
The impact of standardized innovation management systems on

innovation capability and business performance: An empirical study
a, a b
Moises Mir *, Martí Casadesús , Luc Honore Petnji
qq a
Universitat de Girona, Escola Politècnica Superior, C/M Aurèlia Capmany, 61-17071 Girona, Spain
rr Universitat Internacional de Catalunya, Faculty of Economics and Social Sciences, Immaculada 22, 08017 Barcelona, Spain
ARTICLE INFO ABSTRACT
Article history:
Received 23 September 2015 This paper provides a novel analysis concerning the impact of a Standardized Innovation Management
Received in revised form 20 April 2016 System (SIMS) on company innovation capability, innovation performance, and corporate results that is an
Accepted 6 June 2016 Available online 23 unprecedented contribution to innovation management literature. This research is focused on the standard
June 2016 UNE 166002 that is one of the first national certifiable SIMSs that exists globally. It is one of the first SIMSs
in the world that has reached a sufficient level of acceptance to enable an empirical study. In order to analyze
Keywords: its impact on companies, a relational model was developed and analyzed using hypotheses concerning with
Innovation management system the main dimensions.
Standard
UNE 166002
ã 2016 Elsevier B.V. All rights reserved.
Business performance
1. Introduction
It is widely recognized that innovation is a key factor for corporate competitiveness (Porter, 1980; Roberts, 1998) and economic development
(Schumpeter, 1934). Moreover, the innovation process is considered an essential corporate process (Matias and Coelho, 2011) that has to be
managed properly to get profit from it. Nowadays, innovation can be managed through Standardized Innovation Management Systems (SIMSs).
These systems, which have recently appeared on the innovation scene, are sets of standards designed to help companies navigate the complex
process of innovation, systematize their activities and enhance efficiency of its management.
The present empirical research explores the impact of a Standardized Innovation Management System (SIMS) on Innovation Capability (IC)
and on Business Performance (BP) of firms.
Before the recent emergence of SIMSs, no specific standard for managing the corporate innovation process were available. This lack of
resources was often addressed through the use of Total Quality Management (TQM) systems, which are still frequently used today (Kanji, 1996),
or through the application of the ISO 9001 standard for quality management. These systems, paradigms of the new management standards, have
been studied deeply, and their benefits, limits and disadvantages have been discussed in the context of creativity and innovation ( Rossetto, 1995;
Kanji, 1996; Kondo, 1996, 2000; Keogh and Bower, 1997; Mathur-De-Vré, 2000; Bossink, 2002; Prajogo and Sohal, 2003; Prajogo and Hong,
2008; Jayawarna and Holt, 2009). In fact, several empirical studies have found a significant positive relationship between TQM,
* Corresponding author.
E-mail addresses: u1034468@correu.udg.edu, moi025@hotmail.com (M. Mir), marti.casadesus@udg.edu (M. Casadesús), arinopetnji@hotmail.com (L.H. Petnji).
http://dx.doi.org/10.1016/j.jengtecman.2016.06.002
0923-4748/ã 2016 Elsevier B.V. All rights reserved.
M. Mir et al. / Journal of Engineering and Technology Management 41 (2016) 26–44 27
quality and innovation at the organizational level (Prajogo and Sohal, 2003). Other R&D studies have found that TQM principles are useful for
building a range of capabilities far beyond the realm of quality (Prajogo and Hong, 2008). Quality guidelines are often used in R&D departments
as an organizational tool for building innovative capability (Perdomo-Ortiz et al., 2006; Prajogo and Sohal, 2006; Santos-Vijande and Alvarez-
Gonzalez, 2007). Consequently, it was sensible to develop management standards, following quality principles and the knowledge developed, to
manage innovation. However, Zairi (1994) suggested that most companies found it extremely difficult to apply TQM concepts and techniques in
the area of innovation. Moreover, some studies that have examined the relationships between the quality standards, such as ISO 9001, and
innovation have found that standardizing the innovation process results in greater process control, which may disrupt the level of freedom
necessary for creativity and R&D processes (Kondo, 1996, 2000; Mathur-De-Vré, 2000; Jayawarna and Pearson, 2001; Jayawarna and Holt,
2009). Furthermore, popular management discourse of fundamental contradictions between innovation and standardization exists, arguing that
standardization can imply rigidity and a site for resistence for ones, as well as a medium and outcome for change and incremental innovation for
others (Wright et al., 2012).
In recent years, it is clear that the normative paradigm has changed with the emergence of the first iterations of SIMSs in various countries
such as Brazil, Colombia, Denmark, France, Ireland, Mexico, Portugal, Russia, Spain and United Kingdom (see Table 1). The normative
contexts of these frameworks are changing considerably and will continue to change in the future, especially given the recent publication of the
European standard CEN-TS 16555-1 Innovation Management: Innovation Management System, published by the European Committee for
Standardization (CEN, 2013) in July 2013, and with the recent creation of the ISO/TC 279 Innovation Management Technical Committee that is
currently developing the ISO 50501 standard for innovation management at international level forecasted for March 2018.
SIMSs guidelines, regardless of the country or territorial scope of focus, share a common objective: to systematically and ef ficiently manage
company innovation processes to improve innovative capability and business performance (Mir and Casadesús, 2011a,b). This goal has not been
empirically studied before the present study.
One of the first SIMSs frameworks to be developed was the Spanish UNE 166002:2006 Management of R&D&I: Requirements of the
R&D&I Management System (AENOR, 2006) published by the Spanish Association for Normalization (AENOR) that a new version is available
in 2014. The aim of this SIMS is to enhance Research, Development and Innovation (R&D&I) of companies. This SIMS is an appropriate topic
for study because of the novelty of its focus, the fact that the data it has produced has not previously been studied empirically, and because this
standard has been implemented and certified in more than 500 companies in Spain (Mir et al., 2012, 2014). Spain has been chosen as the focus of
this study because it is a forerunner country implementing a certifiable standard on innovation management. However, the fact that Spain faced
an unprecedented financial crisis in the period that is analyzed in this study cannot be ignored.
The first case studies that were conducted on SIMS implementations and analyzes in Spain were focused on sectors such as construction
(Correa et al., 2007; Pellicer et al., 2008, 2010, 2012, 2014; Yepes et al., 2016), manufacturing (Mir and Casadesús, 2008, 2011a,b),
biotechnology (Romero-Cuevas et al., 2010) and nanotechnology (Law, 2010). Other studies have made comparisons with other standards (Law,
2010; Mir and Casadesús, 2011a,b Mir, 2012), studied patterns of diffusion over time (Mir et al., 2012, 2014) and proposed combining standards
across systems (Law, 2010; Mir and Bernardo, 2012). Nonetheless, the existing literature indicates that no empirical study has focused on the
benefits of SIMSs for innovation capability and business performance of companies. Therefore, the objective of this paper is to explore this gap,
based on a series of issues first presented by Mir and Casadesús (2011a,b), who question whether the SIMSs tangibly improve innovative
capabilities, innovation outcomes, and economic results of companies.
This scientific and independent study will be of interest to the scientific community as well as entrepreneurial and social circles, both in Spain
and internationally, given that the model developed on the basis of hypotheses was found to be robust. This strength means that it is possible to
expect that the results can be generalized to other countries with similar national
Table 1
Standardized Innovation Management Systems (SIMSs) in alphabetical order by region.
Geographical Year of Standard

scope publication
Brazil 2011 ABNT NBR 16501:2011 – Guidance for the research, development and Innovation (R&D&I) management system.
Colombia 2008 NTC 5801:2008 – R&D&I Management: Requirements of the R&D&I management system.
Denmark 2010 DS-hæfte 36:2010 – User oriented innovation management.
Europe 2013 CEN-TS 16555-1:2013 – Innovation management: Innovation management system.
France 2013 FD X50-271:2013 – Innovation management – Guide for innovation management implementation.
International 2018 forecasted ISO 50501 – Innovation management system guidelines.
Ireland 2009 NWA 1:2009 – Guide to good practice in innovation and product development processes.
Mexico 2008 NMX-GT-003-IMNC-2008 – Technology management system Requirements.
Portugal 2007 NP4457:2007 – R&D&I Management: Requirements of the R&D&I management system.
Russia 2010 GOST R 54147:2010 – Strategic and innovation management. Terms and definitions.
Spain 2006 UNE 166002:2006 – R&D&I Management: Requirements of the R&D&I management system. (New version
available in 2014)
United Kingdom 2008 BS 7000-1:2008 – Design management systems – Part 1: Guide to managing innovation.
Source: authors’ own construction.

28 M. Mir et al. / Journal of Engineering and Technology Management 41 (2016) 26–44
standards, or to wider territorial scope standards, including the future innovation management standards currently being developed by ISO.
The article is structured as follows. This introductory section is followed by a second section that provides a literature review. The third
section presents the hypotheses and the research model. The fourth section describes the methodology, and the fifth section describes the data
analysis and results of the empirical study. The last section provides the discussion and conclusions with practical implications, limitations and
opportunities for further research.
2. Literature review
Previous studies that have examined the effects of management standards on innovation have focused on quality management standards
(QMS), environmental management standards (EMS) and integrated management systems (IMS). These studies have been widely discussed in
the literature, and their positive and negative judgments on innovation, creativity and R&D have been highlighted. However, while
contradictions arise at the operational level, there is widespread agreement that TQM, and consequently management standards, has favorable
effects on innovation at the strategic level (Kondo, 1996, 2000; Prajogo and Sohal, 2004, 2006; Prajogo and Hong, 2008).
Quality management systems are limited in managing innovation when applied to tangible tasks involved in innovation processes (Prajogo
and Hong, 2008). Furthermore, previous studies have focused narrowly on R&D departments, but innovation extends beyond R&D production
and processes. The current holistic understanding of innovation processes (Lawson and Samson, 2001) acknowledges the existence of four types
of innovation: product, process, organization and marketing innovation (OECD, 2005). A management system designed to manage innovation
must address the entire innovation process and not only the quality of the R&D department. SIMSs guidelines intend to build on this
understanding by managing innovation processes systematically and strategically that not only takes place within R&D departments, but also in
all other departments (Mir and Casadesús, 2011a,b).
Numerous authors have expressed the need to systematize the quality process in R&D settings (Robins et al., 2006; Jayawarna and Holt,
2009) to facilitate knowledge transfer for the possible application of results (Mathur-De-Vré, 1997, 2000). Meeting this need involves
maintaining registries and documents to avoid losing knowledge while continuously improving R&D processes (Pellicer et al., 2008, 2010),
communication in multidisciplinary teams (Valcarcel and Rios, 2003; Robins et al., 2006) and innovation task planning. This practice is also
understood to equip companies with organization, control and management systems (Pellicer et al., 2008) that benefit from clear goals,
designated resources and strategic approaches to reducing inherent uncertainties during the initial stages of innovation. Systemization of
innovation should be pursued through process management (Pellicer et al., 2010). This, in turn, is believed to increase the efficiency (in time and
cost) of projects while anticipating the changing needs of clients and economic environments.
This is critical given that SIMSs frameworks provide standards for managing the innovation process ( Pellicer et al., 2008) that can be
integrated into other management systems because of their similar structure (Pellicer et al., 2008; Law, 2010; Mir and Casadesús, 2011a,b; Mir
and Bernardo, 2012). However, the usefulness of SIMSs frameworks and the benefits that these systems provide in relation to innovative
capability and business performance have not yet been demonstrated (Mir and Casadesús, 2011a,b) until today. Although recent exploratory
studies in the construction sector have been conducted, they were performed in the form of a single company case study (Pellicer et al., 2012,
2014; Yepes et al., 2016). Thus, prompting calls to empirically conduct studies with a larger number of companies (Pellicer et al., 2014; Yepes et
al., 2016).
Besides, it is worth taking into account the literature regarding the Resource-Based View (RBV) and the Dynamic Capabilities (DC)
perspective to examine the competitive relevance of Innovation Capability (IC). On the one hand RBV theories focus on resources. Speci fically,
the accumulation of Valuable, Rare, and Inimitable as well as Nonsubstituible (VRIN) resources as a basis of enterprise competitiveness and
economic rent (Barney,1986). According to Newbert (2007), the value and rare resources are related to competitive advantage and competitive
advantage is related to performance. On the other hand, DC that can be de fined as organization capabilities to integrate, learn and reconfigure
internal and external resources (Teece et al., 1997) were shown to be considered as latent abilities or skills (Helfat et al., 2007; Teece, 2007).
However, Eisenhardt and Martin (2000) recognize them as processes or routines. Moreover, Zollo and Winter, (2002) suggested that the only
way these dynamic capabilities can be a source of competitive advantage is if they are applied “sooner, more astutely, or more fortuitously.”
Clearly, the ability to apply dynamic capabilities “sooner or more astutely” is itself a capability. Given that in a rapidly changing market, some
firms are more likely to be agile, more able to change quickly and more alert to changes in their competitive environment, they will be able to
adapt to changing market conditions more rapidly than competitors and thus can gain competitive advantage. To the extent that the ability to
change quickly, and alertness to changes in the market are costly for others to imitate, these abilities can be a source of sustained competitive
advantage (Eisenhardt and Martin, 2000).
More recently, Lin and Wu (2012) study empirically assessed both the effects of RBV and DC on performance and suggested that the
strategic management should consider DC and RBV in combination instead of separately. In this context, SIMS and IC may be seen as types of
DC to the extent that it may help the firms not only to be aware of the environment changes (market and technology) but also to learn latent
abilities and skills for innovation (Helfat et al., 2007; Teece, 2007). Furthermore, as SIMS is a management system issued to manage the
innovation process more efficiently, it is apparent that it may lead to a faster adaptability. Besides, it may also lead to application of changes that
may increase the difficulties to
imitate the firm VRIN resources and maintain or enhance competitive advantage by implementing innovation routines (Eisenhardt and Martin,
2000) such as the ones defined in a SIMS.
In a similar way, Lawson and Samson (2001) propose that innovation management can be viewed as a form of organizational capability.
Besides, they argued that excellent companies invest and nurture this capability from which they execute effective innovation processes. In
addition, innovation management leads to innovations in new product, services and processes as well as superior business performance results.
Drawing on that, the authors believe that SIMS is also a form of organisational capability as a system or a mechanism or a framework for
innovation management (that is implemented according to the guidelines and requirements of a standard). Thus, as it is a systematic management
system, it is not static. On the contrary, it is dynamic as it is frequently reviewed and enhanced by a continuous improvement methodology
(under Deming PDCA philosophy). Just like ISO 9000 and ISO 14000, it may be considered as a kind of organizational learning system. As
Lawson and Samson, (2001) and others, claim that the ability to learn and the ability to change are likely to be among the most important
capabilities that a firm can possess.
As the UNE 166002 standard aims to systematize and manage the innovation process in an ef ficient manner to enhance innovation capability
and business performance (Mir and Casadesús, 2011a,b), there is a need for empirical examinations of the effects of this standard on companies.
Furthermore, the conclusions of this study will be of great importance for the upcoming standards for innovation management. Apart from case
studies of the Spanish standard (Mir and Casadesús, 2011a,b; Pellicer et al., 2014; Yepes et al., 2016), exploratory analyzes of SIMSs
frameworks in other countries such as Mexico (Pedroza et al., 2013) and Portugal (Peetri et al., 2013) have been conducted. The SIMSs
structures established in these countries are still in their infancy and are not yet mature enough to warrant empirical study of their impacts.
However, the number of certificates being granted is increasing. In Portugal, for example, over 150 companies are now certi fied under their
national standard for innovation management NP4457:2007 (Peetri et al., 2013).
In summary, previous studies on management standards and the effects of those standards on innovation have focused on quality
management standards and systems. While such standards are still being discussed (Kondo, 2000; Prajogo and Sohal, 2006; Prajogo and Hong,
2008), no empirical studies have examined the role of innovation management standards. One reason for this absence is the lack of data
collection to date, and data has been systematically gathered for the first time in the present study. UNE 166002 is an opportunity for research,
because certification with this standard is growing; over 500 certificates have been issued to companies (Mir et al., 2012, 2014). However, the
number of previous studies analyzing the impact of UNE 166002 is limited by the small number of case studies that have been performed ( Mir
and Casadesús, 2011a,b; Pellicer et al., 2014; Yepes et al., 2016). Introduction of SIMSs in Europe (CEN, 2013) is in its infancy stage, and
national standards for innovation management have been published in numerous countries in recent years, including Brazil, Colombia, Denmark,
France, Ireland, Mexico, Portugal, Russia, Spain and the United Kingdom, but no empirical studies of their impact on companies have been
produced.
Because of the relevance and importance of studying SIMSs frameworks and the effects that these systems may have on the innovative
capability of companies in which they are implemented, the current paper aims to answer the following question about the impact of the SIMS
on companies in relation to the empirical data: Do standardized innovation management system (SIMS) improve the innovative capability (IC)
and business performance (BP) of companies?
INNOVATION BUSINESS
CAPABILITY PERFORMANCE
SIMS
Fig. 1. SIMSMA – Standardized Innovation Management System Model Analysis.

Source: adapted from Lawson and Samson (2001)
3. Hypotheses
The objective of this empirical study is to investigate whether a relationship exists between SIMS, innovation capability (IC) and business
performance (BP). No empirical studies that address these issues, either scientific or professional, have been identified in the previous review of
the literature. This lack of previous studies demonstrates the novelty of the subject and the pioneering nature of this paper.
The study was first defined using a conceptual hypotheses model called the Standardized Innovation Management System Model Analysis
(SIMSMA). The SIMSMA is the first approach (Fig. 1), adapted from Lawson and Samson (2001) that makes it possible to examine three
principal dimensions:
xlv) Standardized Innovation Management System (SIMS)

xlvi) Innovation capability (IC)
xlvii) Business performance (BP)
Drawing on this model from Lawson and Samson (2001), combined with the innovation capability measurement framework (Saunila and
Ukko, 2012), the authors propose a new model, SIMSMA2, that includes all the hypotheses that can shed light on the objective of this study,
especially the hypotheses related to the SIMS. The method is also inspired by previous models that were developed to examine the relationship
between TQM and innovation (Prajogo and Sohal, 2006). The proposed model, SIMSMA2, employs six dimensions in a diamond shape, as
shown in Fig. 2. Below, each one of the proposed hypotheses included in the model is described.
H1. The types of innovation that develop influences the innovation capability of the company
Types of innovation (product innovation, process innovation, marketing innovation, organizational innovation) are defined in the Oslo
Manual (OECD, 2005), and two additional types of technological innovation are defined in the standard UNE 166002 (AENOR, 2006). As a
holistic model of innovation capability, innovation capability itself is not a separately identifiable construct. Innovation capability is shaped by
reinforcing practices and processes within a firm. In consequence, these processes are key mechanisms for stimulating, measuring and
reinforcing innovation. Until a few years ago there was no agreement on the variables that in fluence innovation capability, even though there
were many different proposals. One of the most useful models for the present study is composed of the seven major elements proposed by
Lawson and Samson
Financial
Crisis
H5
H4 H6
H10
Types of
Innovation H2
H1
Innovation H3 Business
Capability Performance
H7
Innovation H8
Performance
H9d
H9a H9b H9c
SIMS
UNE 166002
Fig. 2. Research framework SIMSMA2.

Source: authors’ construction
(2001). However, since 2008 an European agreement is available in the form of a published standard: CWA 15899:2008 Standardization of an
innovation capability rating for SMEs, which defines nine categories or groups of major elements (CEN, 2008) based on the literature on
innovation management and best practice models. Innovation Capability is measured in the present study according this standard which is indeed
very similar to the model of Lawson and Samson (2001).
H2. The types of innovation that develop influences the business performance
Considering studies that have been conducted on innovation and firm performance (Geroski et al., 1993; Han et al., 1998; Roberts, 1999;
Calantone et al., 2002; Cho and Pucik, 2005; Artz et al., 2010; Therrien et al., 2011; Gunday et al., 2011), of the four types of innovation defined
in the Oslo Manual (OECD, 2005), only product and process innovation positively and significantly affect firm performance. No evidence has
been found for a significant and positive relationship between non-technological innovation (organizational and marketing innovation) and firm
performance (Atalay et al., 2013). In order to analyze business performance, key performance indicators (KPI) from EFQM (2005) and CIS
(2010) are used, including, for example, annual turnover and staff size.
H3. The innovative capability of a company influences the business performance

Company R&D expenditures are related to subsequent sales revenues (Franko, 1989). Thus, innovative firms may acquire larger market
shares and achieve higher growth rates and profits (Geroski et al., 1993). For example, Acs and Audretsch (1989), using counts, and Archibugi
and Pianta (1996), using patent indicators, confirm that technology and performance are associated. However, how firms leverage innovative
capability and innovativeness to enhance business performance is poorly understood.
In addition, elements of innovation capability can have effects on business performance related to personnel, customers, processes and
finances (Neely, 1998). These elements do not follow one another in any particular order. Every element of innovation capability can lead to
improvements in business performance, either directly or indirectly through another element (Neely, 2005; Saunila and Ukko, 2012).
In order to examine the hypotheses, an “innovation capability” construct with nine major elements, based on CWA 15899 (CEN, 2008) is
proposed. These elements are: innovation culture, strategy, competence and knowledge, technology, product and service, process, structure and
network, market, and project management. These elements can be roughly matched with the seven elements suggested by Lawson and Samson
(2001), which are vision and strategy, harnessing the competence base, organizational intelligence, creativity and idea management,
organizational structures and systems, culture and climate, and management of technology.
H4. The economic crisis influences the innovative capability of a company

The question of how the financial crisis influences the companies to invest in innovation or vice versa still induces discussion. On the one
hand, some suggest that economic cycles are the consequence of innovation. On the other hand, others argue that innovative activities and
innovative organizations are reshaped by economic crises. Regardless of the debate, the 2008 financial crisis has negatively influenced the short-
term willingness of companies to invest in innovation. While on the whole firms’ investment in innovation declined during the economic
downturn, a small but significant minority of firms are “swimming against the stream” and have increased their expenditures on innovation
(Archibugi et al., 2013). In contrast, small firms are possibly more vulnerable to stopping investment in innovative projects during the crisis
because they are more likely to face difficulties in accessing external finance (Paunov, 2012). Drawing on that, this study hypothesis that:
H5. The economic crisis influences the business performance
H6. The economic crisis influences the types of innovation that develop
Considering the high impact of the crisis in Spain during the progress of this study, it is important to analyze the impact of the crisis on the
innovative capability of companies, business performance and types of innovation that develop as presented in the previous three hypotheses.
Companies tend to reduce investment in innovation during times of economic crisis (Mir and Casadesús, 2011a,b). The economic crisis caused,
at least during its initial stages, a destruction of innovation investment (Archibugi et al., 2013). In the light of this, it is necessary to separate the
impact of the implementation of SIMS from the impact of the crisis.
H7. Innovation capability influences innovation performance

Innovation can only occur if a firm has the capability to innovate (Laforet, 2011). The stronger the innovation capability possessed by a firm,
the more effective will be their innovation performance (Lawson and Samson, 2001). In order to evaluate these hypotheses, an innovation
capability construct is proposed with nine major elements, as mentioned above. Furthermore, according to Saunila and Ukko (2012), the
innovation potential can be divided into five categories, namely leadership and decision-making processes, organizational structures and
communication, collaboration and external links, organizational culture and climate, and individual creativity and know-how. Organizations that
exploit these aspects effectively during innovation processes are expected to enjoy successful outcomes from innovation activities ( Saunila and
Ukko, 2012).
H8. Innovation performance impacts on business performance

Organizations that exploit innovation potential effectively during innovation processes are expected to enjoy successful outcomes from
innovation activities, which will in turn affects the organization’s long-term performance (Saunila and Ukko, 2012). With reference to the
relationship between innovation and profitability, previous studies (Geroski et al., 1993; Geroski, 1997) have found a positive, although modest,
direct effect, but a large indirect effect due to the relative insensitiveness of innovating firms to adverse macroeconomic shocks.
H9a. Managing innovation through the UNE 166002 (SIMS) impacts innovative capability
H9b. Managing innovation through the UNE 166002 (SIMS) impacts innovation performance
On the one hand, TQM practices are effective organizational tools for building innovative capability (Perdomo-Ortiz et al., 2006; Prajogo and
Sohal, 2006; Santos-Vijande and Alvarez-Gonzalez, 2007; Prajogo and Hong, 2008). On the other hand, Sitkin et al. (1994) highlighted that
TQM systems must not be viewed only as sets of tools, methods and practices. Rather, and more importantly, TQM guidelines offer universal,
profound and valuable knowledge that may be tailored to suit specific organizational needs in areas such as quality and innovation. TQM
practices effectively enhance the performance of R&D functions (Prajogo and Hong, 2008). Both aspects suggest that the implementation of
SIMS, using UNE 166002 for example, will enhance innovative capability and innovation performance of firms as is proposed in previous
hypotheses.
H9c. Managing innovation through the UNE 166002 (SIMS) influences business performance
H9d. Managing innovation through the UNE 166002 (SIMS) moderates the effect of the impact of economic crisis on company innovative
capability
H10. The crisis factor moderates the effect of innovation capability on innovation performance
The implementation of a SIMS, for example under the UNE 166002 standard, improves innovative capability and business performance as
seen in previous case studies (Correa et al., 2007; Pellicer et al., 2008; Law, 2010; Mir and Casadesús, 2011a,b; Matias and Coelho, 2011;
Pellicer et al., 2012, 2014; Yepes et al., 2016). However, the economic crisis has reduced investment in innovation (Archibugi et al., 2013). Both
concepts are evaluated with the previous hypotheses.
4. Methodology
An empirical analysis based on a survey was specifically designed for the present study. The survey was designed using standards and
guidelines, namely UNE 166002 (AENOR, 2006), CWA 15899 (CEN, 2008), EFQM Framework for innovation (EFQM, 2005), Oslo Manual
(OECD, 2005) and the Community Innovation Survey (CIS, 2010).
The survey, which consisted of 51 questions, was distributed to 1000 Spanish companies. This number of companies was selected because it
was already known that approximately 500 companies possessed a Spanish UNE 166002 SIMS certificate, and the aim was to achieve roughly
equal responses from certified and non-certified companies. A list of 500 certified companies under UNE 166002 requirements was constructed,
after collating information from all the certification bodies in Spain. A similar list of 500 other companies neither implemented nor certified
under UNE 166002 standard, but comparable to the first sample on all other variables, was also constructed. For example, one of the variables
that were controlled in both samples was company size, with a very similar balance in both samples of big companies (31.5%), medium
companies (31.5%) and small companies (37%). The questionnaire was answered mainly by the R&D Director of each company (44%), the
Technical Manager (29%) or the General Manager (15%). In only few companies (12%) the questionnaire was answered by an employed with
low responsibilities. Moreover, the level of studies of the respondents was very high in both samples (96% with a university degree). The Annex
presents the questionnaire that was used for this study.
The survey reached targeted recipients in 347 of the companies and 107 surveys were answered. After rejecting some incomplete
questionnaires, 73 of valid completed questionnaires were retained. This later figure represented a response rate
Table 2
Descriptive summary of the survey.
Population: Spanish companies of any size or sector

Location: Spain
Total sample size: 347
Number of fully completed surveys: 73
Method for obtaining information: Online survey
Response rate: 21%
Sampling Error: 8.3 at 95% confidence interval
Fieldwork period: April 2011 to December 2011
Source: authors’ construction

of 21% with a sampling error of 8.3% at 95% confidence. 57.5% of the total were certified under the UNE 166002 standard and were actively
using its directives, and 42.5% were not. Table 2 includes a summary of survey characteristics.
As in most surveys, there is the possibility of bias due to non-response. Non-responses could mean that companies are not very satis fied with
their innovation skills or capabilities, which would produce more impressive results if all companies had responded. There is no way to be sure
about this question without developing further research in the future to test the results.
The European Agreement CWA 15899:2008 Standardization of an Innovation Capability Rating for SMEs provides an overall Innovation
Capability (IC) score, from 1 to 4, for companies as a mean value of the 37 scores obtained in its own questionnaire. These 37 questions were
included in the present research survey, and the global average for the IC value was found to be 2.97.
Fig. 3 shows a first approximation of the preliminary aggregate results (Table 3) for the IC and SIMS dimensions, dividing the sample
between companies certified under SIMS and those that are not. Of the companies certified with SIMS, 85.71% have a high (>2.5) IC value (A),
and companies without SIMS certification have an even balance of 53.33% (A) high and 46.67%
(II) low (< = 2.5) IC values. It is observed that SIMS certification may be highly positively correlated with IC. Furthermore, companies were
asked about changes in their business performance (BP), through a direct question about
their improvement over the preceding four years. Of the total number of companies, approximately half (Table 4) have experienced improved
results (BP), while the other half have not (51.4% and 48.6%, respectively). Conducting the same analysis as before, as is represented in Fig. 4,
66.7% of the companies with SIMS certification reported improved business performance over the previous four years while only 28.6% of
companies not certified under the standard reported this result. Although it could be assumed that most companies would not experience
improved business performance during a period of economic crisis, it is interesting to note that 66.7% of companies with SIMS certification
experienced improved business performance during this period from 2007 to 2010. This result shows that there may be a possible positive
relationship between SIMS and BP.
5. Data analysis and results
5.1. Assessment of the measures
Because scales assessing innovation capability have never been tested previously, an initial Exploratory Factor Analysis (EFA) of the
measures was undertaken to test the scales’ underlying factors. SPSS 19 software package was used to perform
90%
80%
70%
60%
50%
40%
30%
20%
10% B
0%
A IC
No
Yes
SIMS
Fig. 3. Graph statistic for SIMS and IC.
Table 3
Data statistics for SIMS and IC.
SIMS (UNE 166002) IC – Innovation Capability
A High degree B Low degree Total general

Yes 85.71% 14.29% 100.00%
No 53.33% 46.67% 100.00%
Total 72.22% 27.78% 100.00%
Table 4
Data statistics for SIMS and BP.
SIMS (UNE 166002) BP – Business Performance
Yes No Total
Yes 66.7% 33.3% 100.0%
No 28.6% 71.4% 100.0%
Total 51.4% 48.6% 100.0%
80,0%
70,0%
60,0%
50,0%
40,0%
30,0%
20,0%
10,0% No
0,0%
Yes BP
Yes
No
SIMS
Fig. 4. Graph statistic for SIMS and BP.
arrays of EFA and the Normalized Varimax was used as the principal component factor rotation method (Hair et al.,1998). All items that loaded
poorly or which cross-loaded were rejected.
The overall results of the remaining items indicate that the Kaiser-Meyer-Olkin (KMO) measure was 0.86 and significant at p < 0.000. The
corresponding approximate Chi-Square was equal to 1014.95 with df = 253. Moreover, only items with loading values higher than 0.5 were
retained. The overall results confirmed five factors, namely “Strategy,” “Market,” “Structure and Network,” “Innovation Culture” and “Project
Management Process”. These factors has been named based on the grouping of the manifest variables in their corresponding factors, which
aligned better with the groupings established in the CWA 15899 standard. The five factors represented 68.55% of total variance extracted (see
Table 5).
Table 5
Innovation Capability scale assessment analysis.
a
Factors CFA Loadings EFA Loadings
Loadings t- 1 2 3 4 5
b
value
Strategy (a = 0.90, AVE = 0.64)
C12: Innovation projects are based on the general company strategy 0.79 44.73 0.69
C21: Excellent knowledge on the competitive market environment 0.78 38.50 0.77
C22: Precise definition creation in advance of developing tasks and goals 0.84 61.39 0.71
C23: Communication of information needed for innovation projects is exceptionally frank, 0.82 60.26 0.61
transparent and honest
C24: Mistakes made during innovation projects are viewed as opportunities to systematically learn 0.80 42.66 0.62
and improve
C30: The innovation vision is considered during strategic decision-making 0.78 36.40 0.58
C31: Innovation projects follow a documented innovation process that considers all areas of activity 0.76 32.40 0.52
Market (a = 0.86, AVE = 0.59)

C18: Relationships with customers, suppliers, etc. are maintained in anticipation of future market 0.81 40.29 0.65
needs
C19: To meet future demands, customers are included in the entire process of product/service 0.78 36.51 0.74
development
C20: Feedback such as complaints and suggestions are systematically reviewed and acted upon 0.82 47.07 0.78
C33: Through innovation, the company has acquired greater market shares than its competitors 0.78 41.74 0.72
C35: As part of the innovation process, market-oriented distribution channels are identified at an 0.65 19.09 0.54
early stage
C36: By taking into account various factors throughout the product development process, a diverse 0.76 32.37 0.65
range of products is produced
Structure and Network (a = 0.82, AVE = 0.73)

C11: A budget is allocated for innovation projects that is not directly funded by customer orders 0.87 74.00 0.79
C16: External business partners who meet the particular project requirements are sought 0.82 30.17 0.74
C17: Correspondence with external research sources is maintained to remain aware of relevant 0.88 65.09 0.79
technological and research developments
Innovation Culture (a = 0.79, AVE = 0.61)

C3: Management bodies demonstrate high willingness to engage in new ventures (openness to new 0.87 75.01 0.64
markets and technologies, etc.)
C7: Employees are free to present ideas or suggestions at any time 0.75 22.87 0.77
C15: Capability exists for employees with different backgrounds to work together in innovation 0.81 30.42 0.58
project teams
C29: Project team members treat one another with trust and respect 0.69 20.87 0.73
Project Management Process (a = 0.77, AVE = 0.68)

C4: Innovation project risks are controlled systematically using methods and tools such as SWOT 0.79 23.48 0.61
analysis, etc.
C27: Innovation projects are nearly always carried out on planned time and budget schedules 0.83 50.09 0.85
C28: Clearly defined and precise criteria are used to evaluate and select potential projects for 0.86 43.45 0.62
implementation
Note: CFA = Confirmatory Factor Analysis; EFA = Exploratory Factor Analysis; a: Total variance extracted through the five factors equals 68.55%; Rotation: Varimax normalized;
The Kaiser-Meyer-Olkin (KMO) measure = 0.86; Approx. Chi-Square = 1014.95; df = 253 with a significance of p< 0.000; b: all t-values are significant at P < 0.001
5.2. Reliability and validity of the proposed scales
The dimensionality of the scales was further confirmed by performing a range of Confirmatory Factor Analyzes (CFA). All items retained
from the EFA were constrained to load their respective factors. The results of the CFA presented in Table 5 demonstrate that all item loadings
were greater than 0.5 on their respective constructs. This confirms individual item reliability (Sanzo et al., 2003). The convergent validity of the
individual items was confirmed by loading values greater than 0.7 (Carmines and Zeller, 1979), with the exception of item C35, which was
slightly lower. However, because it was so close to the threshold value, this item was retained in accordance with the relaxed criterion suggested
by Petnji et al. (2011).
The measurement items for the remaining latent variables of the model were also examined through groupings of CFA as presented in Table
6.
Table 6 demonstrates that individual item reliability and validity in relation to the relevant factors were confirmed with the exception of items
TI1, TI3 and IP2, which were nevertheless retained because they were close to the threshold value of 0.7. Furthermore, the dimensionality and
internal consistency of each scale was assessed based on Cronbach’s Alpha (Cronbach,1951) and Average Variance (AVE) composite reliability
criteria. The results in Tables 5 and 6 show that Cronbach’s Alpha ranged from 0.75 for “Product” to 0.90 for “Strategy” and was greater than the
minimum accepted threshold of 0.7
Table 6
Measurement items for the remaining latent variables in the model.
a b
Factors Loadings t-value SD
Types of Innovation
c
Product (a = 0.75, AVE = 0.67)
TI1: Product Innovation 0.67 15.53 0.04
TI6: Incorporation of existing technology 0.92 92.06 0.01
TI7: Development of new technology for the market 0.84 39.88 0.02
Marketing and organization (a = 0.71, AVE = 0.63)

TI3: Service Innovation 0.67 8.91 0.08
TI4: Organization Innovation 0.88 29.28 0.03
TI5: Marketing Innovation 0.82 24.72 0.03
Innovation Performance (2007–2010)

Innovation Performance (a = 0.87, AVE = 0.62)
IP1: Investment in R&D projects 0.78 38.8 0.02
IP2: Approximate deviation from the forecasted budget 0.64 19.91 0.03
IP3: Number of innovative ideas registered and evaluated 0.79 46.36 0.02
IP4: Number of innovative projects launched 0.87 66.19 0.01
IP5: Number of innovative projects ongoing 0.85 87.23 0.01
IP6: Number of innovation projects completed and/or successfully implemented 0.77 30.29 0.03
Business Performance (2007–2010)

Business Performance (a = 0.78, AVE = 0.70)
BP1: Employees 0.85 62.47 0.01
BP2: R&D employees 0.82 47.75 0.02
BP3: Annual turnover 0.84 46.32 0.02
Financial Crisis (2007–2010)

FC: The economic crisis has not reduced the level of investment dedicated to company innovation projects. 1.00 – –
UNE 166002 (SIMS)

The company possesses the UNE 166002 innovation management standard certification. 1.00 – –
a All t-values are significant at: p< 0.000.

b SD = Standard Error.
c a = Cronbach’s Alpha and AVE = Average Variance Extracted.
(Nunnally and Bernstein, 1994). Moreover, the AVE value was greater than the threshold value of 0.5 (Fornell and Larcker, 1981), indicating
high internal consistency. Overall, the strength and psychometric properties of the scales underpinning our model demonstrate that the scales can
be used with confidence.
5.3. Testing the hypotheses in the conceptual model
Structural equation modeling (SEM) was conducted and path modeling was assessed by means of Partial Least Squares using the PLS
Version 2.0 software package. “Innovation Capability” and “Type of Innovation” were modeled as second order constructs to achieve a general
understanding of the model. Again, the factors structure congruence as shown in Table 5 and 6 were further confirmed. In addition, the
Cronbach’s Alpha and the AVE were also higher than the acceptable cut-off value as is shown Table 7.
A discriminatory validity test was also performed to confirm that each factor was uniquely represented as a separate dimension, following
criteria provided by Fornell and Larcker (1981). The results presented in Table 8 demonstrate strong evidence for discriminatory validity
between the constructs. Thus, the linear inter-factor correlation test for each construct with the other constructs was less than the square root of
the AVE value (highlighted in bold italic) for each scale.
Table 7
Model fit with Innovation Capability and Types of Innovation loaded as the second order.
AVE Composite Reliability R Square Cronbach’s Alpha Commonality Redundancy

Business Performance 0.70 0.87 0.47 0.78 0.70 0.03
Financial Crisis 1.00 1.00 1.00 1.00
Innovation Capability 0.53 0.94 0.43 0.94 0.43 0.07
Innovation Performance 0.62 0.91 0.26 0.87 0.62 0.15
Types of Innovation 0.58 0.77 0.09 0.77 0.38 0.03
UNE 166002 1.00 1.00 1.00 1.00
Table 8
Inter-factor correlations between main constructs, and the extracted square root of AVE.
1 2 3 4 5 6
1. Business Performance 0.84
2. Financial Crisis 0.44 1.00
3. Innovation Capability 0.31 0.46 0.66
4. Innovation Performance 0.66 0.63 0.49 0.79
5. Type of Innovation 0.07 0.30 0.47 0.32 0.62
6. UNE 166002 0.23 0.23 0.37 0.32 0.01 1.00
The Goodness-of-Fit index (GoF) for the overall model was examined in accordance with recommendations by Tenenhaus et al. (2004). The
model GoF accounting for both the explained variance and commonality was 0.41, indicating a significantly high fit index (Tenenhaus et al.,
2004). Finally, the hypothesized relationships between different constructs were tested for confidence, and these results are summarized and
presented in Table 9.
As predicted, types of innovation factor positively and directly influences Innovation Capability (H1) and Business Performance (H2). In
contrast, Innovation Capability is not directly influencing Business Performance (H3) but appears instead to be mediated by the Innovation
Performance factor (H7 + H8). Moreover, the results of the study show that the Financial Crisis directly affects Innovation Capability (H4) and
Types of Innovation (H6). Against all odds, the Financial Crisis did not appear to affect Business Performance (H5) directly. In addition,
Innovation Capability directly and positively influences Innovation Performance (H7), which in turn directly and positively influences Business
Performance (H8). Similarly, the SIMS (UNE 166002) directly and positively impacts Innovation Capability (H9a) and Business Performance
( H9c). However, it is unexpectedly not influencing Innovation Performance (H9b) directly. Therefore, it is apparent that the path may instead be
mediated by the IC factor (H9a + H7).
Finally, the moderating effects in the model were assessed following procedures outlined by Baron and Kenny (1986). The UNE 166002
moderating effect (H9d) was assessed by examining its influence on the relationship strength and/or direction between the Financial Crisis
predictor variable and the Innovation Capability outcome variable. The results of the analysis were confirmed through the discovery of a
significant moderating effect. Thus, the relationship strength between the predictor and the moderator was signi ficant at P-value < 0.001.
Similarly, the results that predicted the moderating role of Financial Crisis on the relationship between Innovation Capability and Innovation
Performance were also confirmed (H10). Fig. 5 below summarizes the results of testing the hypotheses.
The direction of the causal relationships defined in H9a, H9b and H9c may be open to debate, because the companies that implemented the
standard may have been already more innovative before the SIMS implementation, with a higher levels of Innovation Capability and Innovation
Performance than companies that have not implemented it. Moreover, Business Performance could be a causative factor in implementing the
standard, as those companies would more easily be able to afford the cost of implementation and certi fication than those companies with lower
levels of Business Performance. Many authors have discussed the parallel dilemma in the case of ISO 9001, including, for example, Häversjö
(2000), Heras et al. (2002), Dick et al. (2008), Heras et al. (2011) and Heras and Boiral (2013). However, the different results found do not make
a final conclusion possible. Of course, no papers discuss this issue in relation to this specific innovation standard.
Nevertheless, as the questionnaire for this study was constructed on the basis of relative development, by asking them about becoming better
or worse for each item of Business Performance and Innovation Performance during the previous 4 years, and as in this period was during most
of the companies implemented the standard, it can be assumed that responses mainly indicate a comparison of the company's status before and
after implementation. This fact suggests that the direction of the relationships between SIMS and IC, IP, and BP are mainly in the direction
previously expected. Moreover, the survey included an extra question only for the certified companies asking about the motivations for the SIMS
implementation. They
Table 9
Hypotheses results for the structural model.
a
Hypotheses Path coefficient SE t-value P-value Conclusion
H1 Types of Innovation !Innovation Capability 0.39 0.04 11.23 0.000 Accepted***
H2 Types of Innovation !Business Performance 0.18 0.04 4.19 0.000 Accepted***
H3 Innovation Capability ! Business Performance 0.05 0.04 1.23 0.222 Rejected
H4 Financial Crisis ! Innovation Capability 0.27 0.03 8.67 0.000 Accepted***
H5 Financial Crisis ! Business Performance 0.05 0.05 1.06 0.293 Rejected
H6 Financial Crisis ! Types of Innovation 0.30 0.04 7.52 0.000 Accepted***
H7 Innovation Capability ! Innovation Performance 0.43 0.04 12.12 0.000 Accepted***
H8 Innovation Performance ! Business Performance 0.67 0.05 14.40 0.000 Accepted***
H9a UNE 166002 ! Innovation Capability 0.31 0.04 8.57 0.000 Accepted***
H9b UNE 166002 ! Innovation Performance 0.01 0.04 0.35 0.727 Rejected
H9c UNE 166002 ! Business Performance 0.16 0.04 3.65 0.000 Accepted***
a (*) (**) (***)
Standard Error; Significant at two tail: P-value < 0.05; P-value <0.01 and P-value < 0.001
Financial
Criisis
H4 H6 H5
H10
Types of
Innovation H2
H1
Innovation H3 Business
Capability Performance
H7
Innovation H8
Performance
H9d
H9a H9c
H9b
SIMS
UNE 166002
Supported relationship
Supported (moderating effect)
Not supported
Fig. 5. Research analysis framework results (SIMSMA2 results).
declared that the most important motivations for the implementation of the standard were to systematize and enhance ef ficiency of the innovation
process of the company and to make more innovation projects (71.5%), rather than to improve corporate image (19%), or because of the fact that
implementation and certification costs were supported by public subsides (9.5%). These responses provide outstanding arguments for assuming
that the directions of the relationships are mainly as proposed.
6. Discussion and conclusions
The literature includes previous studies on the well known standards for quality management, which examine whether standardization and
innovation and/or creativity are compatible through consideration of the ISO 9000 and TQM standards (Kondo, 2000; Prajogo and Sohal, 2006).
These studies, however, were not conducted based on standards specifically designed to manage innovation but to serve quality management
systems, and thus contradictory paths could appear in the results.
Therefore, given that standards exist to help companies more intensively and effectively engage in innovation, as is the case of the standard
UNE 166002 in Spain (AENOR, 2006), and because these standards already have a proven track record of expansion (Mir et al., 2012, 2014), it
is time to study the impact that these systems have on companies empirically. This partly responds to a suggestion made in a case study by Mir
and Casadesús (2011a,b), which called for future research that examines a larger number of companies. The current empirical study has taken an
additional step towards investigating the impact of SIMSs frameworks on companies. The study is based on a 2011 survey that was distributed to
347 companies, which represents a pioneering undertaking in this field. Within the sample of companies examined, two distinct groups are
identifiable; the first consists of companies with UNE 166002 (SIMS) certification while the second group of companies does not have this
credential.
In summary, the Standardized Innovation Management System (SIMS) has a signifi cant positive relationship (H9a) to company
Innovative Capability (IC) and (BP) Business Performance (H9c). Furthermore, SIMS is related to crisis, as it has a
moderating effect (H9d) over (H4), which relates Innovation Capability to crisis. According to the literature, economic crises have a negative
effect on company innovative capability (Lawson and Samson, 2001; Archibugi et al., 2013) and the present study contributes to this knowledge
by finding that SIMS is a moderating factor in the relationship between crisis and innovative capability. Supporting this point, some comments
from respondents of companies without any SIMS implemented should be noted:
“We need to focus on innovation management” or

“Just in 2010 we canceled all the innovation projects, leaving this area of business parked in standby because of the lack of personnel and
time (economic environment)” or
“Today the R&D department has no authority at all in the organization hierarchy; it depends on the technical department that constrains
initiatives and developments. Moreover, there is no specific budget for R&D projects” or
“There is a lot of will for innovation and there are resources for innovation but here exists deep cultural sediment with low tolerance to
criticism. Complacency doesn’t boost innovation; neither do the archaic management business models”
All of these comments focus on the need for an innovation management system, the lack of priority given to innovation processes, the lack of
innovation authority/leadership and the absence of innovation capability and culture. One might expect that if these less innovative companies
had implemented a SIMS, many weaknesses expressed could have been addressed taking into account the results of this study.
Apart from the analysis of the proposed hypotheses represented in the SIMSMA2 model, the study has obtained interesting results from the
Innovation Capability scale assessment analysis, through which five of the nine dimensions established by the European Workshop Agreement
CWA 15899 used in the survey to determine the values of Inovation Capability (IC) were confirmed. The dimensions that were confirmed were
Innovation Culture, Strategy, Structure and Network, Market and Project Management. Agreement was also found for at least four of the seven
major elements proposed by Lawson and Samson (2001), namely vision and strategy, organizational intelligence (related to market),
organizational structures and systems, and culture and climate. Similarly, agreement was found for four of the seven innovation fundamentals
proposed in the “7S Model” of Drejer (2008) which includes market, strategy, organization culture, employee motivation, cross-functional
learning, knowledge management, and internal and external networks. Although certain dimensions in the latter model did not result in a perfect
match, including certain elements such as motivation and relationships between employees, certainly the Innovation Culture factor in this model
should be considered similarly. Finally, it is worth noting that when reducing the scale for IC dimension, 23 of the 37 items proposed by the
CWA 15899 guidelines are found to meet the criterion of validity used in the survey for the IC dimension.
Consistent and significant results are found using this model that can be extrapolated for future studies in other countries or for other
geographical scopes. From the results of this study, it can be concluded that SIMSs, that are standards specifically designed for company
innovation process management, such as the UNE 166002, strongly and positively influence innovative capability and business performance. It
can also been concluded that innovative capability is related to innovation performance in accordance with prior studies ( Lawson and Samson,
2001; Saunila and Ukko, 2012), and this last factor is in turn related to business performance, supporting studies of Franko (1989), Geroski et al.
(1993) and Saunila and Ukko (2012). In addition, the dimension Types of Innovation (TI) is positively related to IC and BP. Finally, although IC
does not appear to influence BP directly, the results in the path modeling suggest that IC have to be mediated by IP factor which do influence BP.
The results of the current study are relevant as they show empirically for the first time that significant relationships exist between SIMS, IC
and BP, showing that SIMS impacts IC and BP significantly and also confirms the findings of prior literature on IC ! IP and IP !BP relationships.
Furthermore, the results of this study contribute to the former academic debate about whether standardization has negative ( Zairi, 1994; Wright
et al., 2012) or positive effects on creativity and innovation performance (Kondo, 1996, 2000; Prajogo and Sohal, 2004, 2006; Prajogo and Hong,
2008) by arguing that while previous standardized management systems such as ISO 9001 had contradictions regarding its effects on innovation,
SIMSs, that are specially developed focusing on Innovation, have clearly strong positive effects on innovation capability and business
performance of firms.
While these results will be beneficial for many companies, they may be most helpful for small and medium enterprises (SMEs), as this
knowledge could assist such companies to enter into the culture of innovation while effectively structuring the complex and often chaotic
innovation process. SIMSs guidelines should be researched further to determine whether these frameworks are more beneficial for SMEs and to
isolate the economic sectors in which these systems are found to be the most effective. Future studies should also examine how SIMSs
guidelines integrate with other management systems as well as the drawbacks and benefits of previously implemented QMS structures such as
ISO 9001.
There are some limitations in this study. One of the main limitation of this study is the geographic focus on one country, Spain. However, it is
also a strength given that Spain is pioneering in this kind of standards and is the only country that has compiled data on SIMS-certified
companies over time (Mir et al., 2012, 2014). In contrast, while the standard BS 7000-1:2008 is valid in the UK, it is not certi fiable and thus no
list of certified companies is available. In Portugal, only 150 companies are certified under the Portuguese standard NP4457:2007 (Peetri et al.,
2013), and the majority of certifications occurred during the last two years. Future studies focusing on broader geographical regions should be
conducted when sufficient data for such research becomes available. Another limitation of this study is the sample size. It has been the dif ficulty
to gather a
greater number of responses, as companies appeared to be reluctant to participate in the study especially because of exposure during the financial
crisis.
It would be interesting to distinguish between the impact of the innovation system itself (internal) and the impact of the certificate (external)
but, unfortunately, data collection included only certified companies. These data do not make it possible to analyze the impact in any companies
that have implemented the standard but did not apply for certification. However, there is another indirect way to analyze this point. If we
consider that the implementation of ISO 9001 is a similar case, and the only one that has been extensively studied, it is clear that many papers
demonstrate that although the motivation to implement the ISO 9001 standard is usually external, the benefits are mainly internal (Leung et al.,
1999; Casadesus and Karapetrovic, 2005a,b). This is to say, the internal effect (improving quality, reducing costs) is more important than the
external effect of having the certificate (marketing, sales). On the assumption that the two cases are similar, it can be anticipated, but obviously
not demonstrated, that the effect of the analyzed standard could be similar. Furthermore, the motivation for SIMS implementation may also
suggests that the impact of the implementation of the SIMS (internal) would be greater than the impact of having the certi ficate (external)
because the main motivation to implement the SIMS is found to be internal (71,5%) rather than external (28,5%) as expressed by the companies
that have achieved the UNE 166002 certificate. This approach can also be exemplified by one of the companies that achieved this certificate. The
respondent of this survey wrote a short comment:
“Even though the R&D department survives, the resources that we have are lower due to the crisis, our production capacity and the income
has been reduced, but we still go ahead”.
This comment, together with other responses to the survey, suggests that the SIMS implementation for these speci fic companies have been
beneficial, at least internally, as they can survive in a deep crisis by working with fewer resources and performing innovation activities more
efficiently (internal benefits).
It is worth noting that, by the perspective of dynamic capabilities, the SIMS may possess a duality. On the one hand, it may be viewed as an
organizational DC itself (Lawson and Samson, 2001) for the management of the innovation resources, which are usually VRIN resources, and on
the other hand, according to the results of this study, it may be seen as an enabler to improve firm IC which is also in fact another kind of DC
(Eisenhardt and Martin, 2000). Consequently, as IC and DC are widely recognized as sources of competitive advantage, the importance of SIMS
influence on them deserves further attention for research, either regional SIMSs or the International ISO 50501 when it is published.
Innovation management, may be seen not only as skills and abilities or processes and routines, but also as a set of organizational capabilities,
and by applying SIMS, the firm personnel can learn how to manage innovation process more efficiently and apply continuous improvements on
the innovation process. Consequently, it is not surprising that a standardized system such as the UNE 166002 has signi ficant effects on firm
innovation capability and business performance. Thus, it suggests that the competitive advantage gained through the new DC adopted (SIMS),
and its benefits to the firm IC, have resulted in the superior business performance (Newbert, 2007; Lin and Wu, 2012).
The results of the study lead to new research questions. For example, are the requirements of the SIMS to be certi fiable beneficial or
detrimental?, companies that implemented the SIMS but did not apply for a certi ficate have better or worse results?, companies that implmented
non-certifiable SIMS have better results?. Are similar trends occurring in other countries with similar published SIMSs? At the European level,
would the establishment of the European SIMS be beneficial? Does the existence of prior management systems before SIMS structures, such as
ISO 9000 or ISO 14000, facilitate or inhibit SIMS implementation and integration? As proposed by Matias and Coelho (2011), is it entirely
necessary that new total management systems focus on innovation? Is there instead a need for a more general ISO that satisfies all stakeholders?
Considering the various key processes of management: innovation, environment, quality, safety, energy, social responsibility and so on, would a
general ISO be feasible and useful? All companies would certainly prefer to become more competitive without necessarily having to adhere to so
many standards that must be integrated and paid for.
Studies should also be conducted on implementation results and the co-existence of SIMS structures with other management systems
(quality, environment, etc.) to analyze the complexities of their implementation and to test whether these systems are truly compatible and
adaptable in the way that is explicitly described in the UNE 166002. Future studies may also evaluate the European standard CEN-TS 16555-
1:2013 that was recently published, though this framework is still in its initial stages, and approximately four years of implementation will be
necessary before sufficient data may be gathered.
We are currently at an inflection point in the field of innovation management and standardized management systems, and this empirical study
is the first to introduce SIMSs frameworks to this field. As was empirically demonstrated in this research, this new agent plays an important role
that cannot be ignored. Further research should thus examine the factors that promote the SIMSs in the form of national and European standards,
and the development of future implementation taking the form of ISO standard on innovation management, as is currently being undertaken by
the ISO/TC 279 Technical Committee. When those standards have been published, their impact on companies should be tested empirically.
Finally, we can speculate about the future of innovation standards. For example, the upcoming ISO standard for innovation management will
be published in guideline format like the British BS 7000-1:2008 rather than as a certifiable standard like the Spanish case (Mir and Casadesús,
2011a,b). This may reflect some interest, as the case of the European standard CEN-TS 16555-1:2013, in avoiding that governments require
standard certification to access to public subsidies for
innovation projects. ORGALIME (2009), the European Engineering Industries Association argue that it is completely up to companies how they
shape their innovation processes, considering that innovation is fundamentally a business-driven decision which should not be restricted by
regulatory requirements and bureaucratic certification measures, and the use of innovation management guidelines should remain voluntary and
not linked to selection criteria for public grants. The dilemma whether innovation standards have to be certifiable or not is open to debate, but the
results of this study are clear and conclude that standardized innovation management systems enhance innovation capability and business
performance of companies even in a crisis period, and the standard analyzed in this study is a certifiable one.
Appendix A
ANNEX – Questions of the survey

Types of innovation
1- In your company, the types of innovation are mostly:
_Product Innovation*
_ Process Innovation*
_ Organizational Innovation*
_ Marketing Innovation*
_ Service Innovation**
_ Incorporation of existing technology***
_ Development of new technology to the market***
*According to Oslo Manual definitions
*
For companies in the service sector
*
According to UNE 166002:2006 definitions
Rate each option from 1 to 5 in importance (1 is not important, 5 is very important). Business
Performance and Innovation Performance During the last four years:
2- The total number of employees

3- The total number of employees in R&D&I
4- The annual turnover
5- The percentage of turnover due to innovations that had equal to or less than 3 years on the market
6- The investment in R&D&I
7- The approximate deviation from the proposed budget for R&D&I
8- The number of innovative ideas recorded and evaluated
9- The number of innovation projects initiated
10- The number of innovation projects in progress
11- The number of innovation projects completed and/or implemented successfully
12- The number of patent applications
The above questions (2–12) have 3 possible answers:
It has increased or improved.
It has remained stable.
It has decreased or worsened.
13- The financial crisis has not made us decrease neither the investments in, nor the number of, innovation projects because we believe that
innovation is a strategic key factor for long term survival.
The possible answers are:
Yes. I strongly agree.
Yes. I somewhat agree.
No. I somewhat disagree.
No. I strongly disagree.
Scoring 4, 3, 2 and 1 successively, according to Likert scale.
Innovative capability (questions 14–50)
There are 37 questions (not listed in this annex) set by the standard CEN Workshop Agreement: CWA 15899:2008 Standardization of an
innovation capability rating for SMEs. This European standard is designed to diagnose the innovative capability of firms. The reader is referred
to those 37 questions in the standard itself.
SIMS – Standardized Innovation Management System (UNE 166002)
51- Your company has implemented and certificated an innovation management system according the requirements of the standard UNE
166002.
Possible answers: Yes/No.
Extra questions
- Overall business performance of the past four years have improved Possible
answers: Yes/No.
Motivation: Factors influencing the implementation of the UNE 166002 when the answer of question number 51 is “Yes”:
- Implementation subsided
- Systematize and improve the efficiency of the innovation process
- Enhance corporate image
- Developing more innovative projects
- Sort highest to lowest factors in importance.
Profile of the respondent: Position in company

- General management
- General or departmental direction
- Intermediate direction
- Employee with low responsabilities
Choose an answer.
Profile of the respondent: Level of education
- University degree
- Medium studies (bachelor's degree or similar)
- Basic studies
Choose an answer.
References
AENOR, 2006. UNE 166002:2006–R&D&I Management: R&D&I Management System Requirements. Spanish Association for Standardization and Certification,
Madrid, Spain.
Acs, Z.J., Audretsch, D.B., 1989. Patents as a measure of innovative activity. Kyklos 42 (2), 171 –180.
Archibugi, D., Pianta, M., 1996. Measuring technological change through patents and innovation surveys. Technovation 16 (9), 451 –468.
Archibugi, D., Filippetti, A., Frenz, M., 2013. Economic crisis and innovation: is destruction prevailing over accumulation? Res. Policy 42 (2), 303 –314.
Artz, K.W., Norman, P.M., Hatfield, D.E., Cardinal, L.B., 2010. A longitudinal study of the impact of R&D, patents, and product innovation on firm performance.
J. Prod. Innov. Manage. 27 (5), 725–740.
Atalay, M., Anafarta, N., Sarvant, F., 2013. The relationship between innovation and firm performance: an empirical evidence from Turkish automotive supplier industry. 2nd
International Conference on Leadership, Technology and Innovation Management, 3 April 2013 – ProcediaSoc. Behav. Sci. 75, 226–235.
Barney, J.B., 1986. Strategic factor markets: expectations, luck, and business strategy. Manage. Sci. 32 (10), 1231 –1241.
Baron, R.M., Kenny, D.A., 1986. The moderator–mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J. Pers.
Soc. Psychol. 51, 1173–1182.
Bossink, B.A.R., 2002. The strategic function of quality in the management of innovation. Total Qual. Manage. 13 (2), 195 –205.
CEN, 2008. CWA 15899:2008 Standardization of an Innovation Capability Rating for SMEs. European Committee for Standardization, Brussels.
CEN, 2013. CEN-TS 16555-1:2013 Innovation Management: Innovation Management System. European Committee for Standardization, Brussels.
CIS, 2010. The Community Innovation Survey Form. July 2010. European Commission, Brussels.
Calantone, R.J., Cavusgil, S.T., Zhao, Y., 2002. Learning orientation, firm innovation capability, and firm performance. Ind. Mark. Manage. 31 (6), 515–524.
Carmines, E.G., Zeller, R.A., 1979. Reliability and Validity Assessment. Sage Publications, London.
Casadesus, M., Karapetrovic, S., 2005a. An empirical study of the bene fits and costs of ISO 9001:2000 compared to ISO 9001/2/3:1994. Total Qual. Manage.
Bus. Excel. 16 (1), 105–121.
Casadesus, M., Karapetrovic, S., 2005b. Has ISO 9000 lost some of its lustre? A longitudinal impact study. Int. J. Oper. Prod. Manage. 25 (6), 580 –596. Cho, H., Pucik, V., 2005.
Relationship between innovativeness, quality, growth, pro fitability, and market value. Strateg. Manage. J. 26 (6), 555 –570. Correa, L. Christian, Yepes, V., Pellicer, E., 2007.
Factores determinantes y propuestas para la gestión de la innovación en las empresas constructoras. Revista
ingeniería de construcción 22 (1), 5–14.
Cronbach, L.J., 1951. Coefficient alpha and the internal structure of tesis. Psychometrika 16, 297 –334.
Dick, P.M. Gavin, Heras, I., Casadesus, M., 2008. Shedding light on causation between ISO 9001 and improved business performance. Int. J. Oper. Prod.
Manage. 28 (7), 687–708.
Drejer, A., 2008. Are you innovative enough? Int. J. Innov. Learn. 5 (1), 1 –17.
EFQM, 2005. The EFQM Framework for Innovation. European Foundation for Quality Management, Brussels.
Eisenhardt, K., Martin, J., 2000. Dynamic capabilities. what are they? Strateg. Manage. J. 21, 1105 –1121.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res. 28 (1), 39 –50.
Franko, L.G., 1989. Global corporate competition: who's winning, who's losing, and the R&D factor as one reason why. Strateg. Manage. J. 10 (5), 449 –474.
Geroski, P., Machin, S., Van Reenen, J., 1993. The profitability of innovating firms. Rand J. Econ. 24 (2), 198–211.
Geroski, P., 1997. How persistently do firms innovate? Res. Policy 26, 33–48.
Gunday, G., Ulusoy, G., Kılıç, K., Alpkan, L., 2011. Effects of innovation types on firm performance. Int. J. Prod. Econ. 133 (2), 662–676.
Häversjö, T., 2000. The financial effects of ISO 9000 registration for Danish companies. Manage. Audit. J. 15 (1), 47 –52.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis with Readings. Prentice-Hall, Englewood Cliffs, NJ.
Han, J.K., Kim, N., Srivastaka, R.K., 1998. Market orientation and organizational performance: is innovation missing link? J. Mark. 62, 30 –45. Helfat, C., Finkelstein, S.,
Mitchell, W., Peteraf, M., Singh, H., Teece, D., Winter, S., 2007. Dynamic Capabilities. Understanding Strategic Change in
Organizations. Blackwell Publishing, Oxford.
Heras, I., Boiral, O., 2013. ISO 9001 and ISO 14001: Towards a research agenda on management system standards. Int. J. Manage. Rev. 15 (1), 47 –65.
Heras, I., Gavin, P.M., Casadesús, M., 2002. ISO 9000 registration ’s impact on sales and profitability: a longitudinal analysis of performance before and after accreditation. Int. J.
Qual. Reliab. Manage. 19 (6), 774–791.
Heras, I., Molina, J.F., Dick, G.P., 2011. ISO 14001 certification and financial performance: selection-effect versus treatment-effect. J. Clean. Prod. 19 (1), 1 –12.
Jayawarna, D., Holt, R., 2009. Knowledge and quality management: an R&D perspective. Technovation 29 (11), 775 –785.
Jayawarna, D., Pearson, A.W., 2001. The role of ISO 9001 in managing the quality of R&D activities. TQM Mag. 13 (2), 120 –128.
Kanji, G.K., 1996. Can total quality management help innovation? Total Qual. Manage. 7 (1), 3–9 .
Keogh, W., Bower, D.J.,1997. Total quality management and innovation: a pilot study of innovative companies in the oil and gas industry. Total Qual. Manage.
129 (5), 562–569.
Kondo, Y., 1996. Are creative ability and work standardization in contradictory relationship? Train. Qual. 4 (3), 35 –39.
Kondo, Y., 2000. Innovation versus standardization. TQM Mag. 12 (1), 6–10.
Laforet, S., 2011. A framework of organisational innovation and outcomes in SMEs. Int. J. Entrepr. Behav. Res. 17 (4), 380 –408.
Law, N., 2010. Application of the IUMSS methodology in an R&D-oriented nanotechnology setting. Master Thesis. Department of Mechanical Engineering.
University of Alberta, Alberta. Canada.
Lawson, B., Samson, D., 2001. Developing innovation capability in organisations: a dynamic capabilities approach. Int. J. Innov. Manage. 5 (3), 377 –400.
Leung, H.K.N., Chan, K.C.C., Lee, T.Y., 1999. Costs and benefits of ISO 9000 series: a practical study. Int. J. Qual. Reliab. Manage. 16 (7), 675–691.
Lin, Y., Wu, L., 2012. Exploring the role of dynamic capabilities in firm performance under the resource-based view framework. J. Bus. Res. 67 (3), 407 –413.
Mathur-De-Vré, R., 1997. Approach to quality system in research and development. Accredit. Qual. Assur. 2 (2), 63 –68.
Mathur-De-Vré, R., 2000. The scope and limitations of a QA system in research. Accredit. Qual. Assur. 5 (1), 3 –10.
Matias, J.C.O., Coelho, D.A., 2011. Integrated total quality management: beyond zero defects theory and towards innovation. Total Qual. Manage. Bus. Excel.
22 (8), 891–910.
Mir, M., Bernardo, M., 2012. Integration of innovation management system standards within existing management systems: a proposed guideline. In: Bernardo, M. (Ed.),
Quality Management and Beyond: the Current Situation and Future Perspectives. UdG Publicacions. GITASP 8, Documenta Universitaria, Girona. Spain, pp. 121–134.
Mir, M., Casadesús, M., 2008. UNE 166002:2006: estandarizar y sistematizar la i + d + i. Dyna Ingeniería e Industria 83 (6), 325 –331.
Mir, M., Casadesús, M., 2011a. Normas para la gestión de la innovación: un análisis comparativo. Dyna Ingeniería e Industria 86 (1), 49 –58.
Mir, M., Casadesús, M., 2011b. Standardised innovation management systems: a case study of the Spanish standard UNE 166002:2006. Innovar 21 (40), 171 – 187.
Mir, M., Llach, J., Marimon, F., 2012. Pattern of diffusion of the UNE 166002 innovation management standard: an exploratory analysis. 6 Th International Conference on
Industrial Engineering and Industrial Management – XVI Congreso De Ingeniería De Organización (CIO), 18 –20 July, Vigo, Spain, pp. 538 – 545 Industrial Engineering:
Overcoming the Crisis. Book of Full Papers.
Mir, M., Llach, J., Marimon, F., 2014. Diffusion of the UNE 166002 innovation management standard: a forecast model approach towards internationalization.
TMQ – Tech. Methodol. Qual. 1 (5), 106–125.
Mir, M., 2012. Gestión de la Innovación y la calidad: Rumbo hacia un binomio excelente. Forum Calidad 235 (12), 36 –43.
Neely, A.D., 1998. Measuring Business Performance. Economist Books, London.
Neely, A., 2005. The evolution of performance measurement research: developments in the last decade and a research agenda for the next. Int. J. Oper. Prod.
Manage. 25 (12), 1264–1277.
Newbert, S.L., 2007. Empirical research on the resource-based view of the firm: an assessment and suggestions for future research. Strateg. Manage. J. 28 (2), 121–146.
Nunnally, J.C., Bernstein, I.H., 1994. Psychometric Theory. McGraw-Hill, New York.
OECD, 2005. Oslo Manual Guidelines for Collecting and Interpreting Innovation Data, 3rd ed. OECD Publishing, Paris.
ORGALIME, 2009. Innovation management: a business choice. [online] availabe from <http://www.orgalime.org/position/innovation-management-business-choice>
(3.08.15).
Paunov, C., 2012. The global crisis and firms’ investments in innovation. Res. Policy 41 (1), 24–35.
Pedroza, A., Ortiz, S., Samaniego, A., 2013. Análisis exploratorio del sistema de gestión de la tecnología, según la norma mexicana NMX GT 003 IMNC.
ALTEC conference. XV Congreso Latino-Iberoamericano de Gestâo de Tecnologia, 27 –31 October, Porto, Portugal.
Peetri, M., Passos, A., Xavier, A., 2013. Understanding the benefits of standardizing innovation management. ALTEC conference. XV Congreso Latino- Iberoamericano de
Gestâo de Tecnologia, 27–31 October, Porto, Portugal.
Pellicer, E., Yepes, V., Correa, C.L., Marinez, G., 2008. Enhancing R&D&i through standardization and certi fication: the case of the Spanish construction industry. Ingenieria de
Construcción 23 (2), 112–121.
Pellicer, E., Yepes, V., Rojas, R., 2010. Innovation and competitiveness in construction companies. J. Manage. Res. 10 (2), 103 –115.
Pellicer, E., Correa, C.L., Yepes, V., Alarcon, L.F., 2012. Organizational improvement through standardization of the innovation process in construction firms.
Eng. Manage. J. 24 (2), 40–53.
Pellicer, E., Yepes, V., Correa, C.L., Alarcon, L.F., 2014. Model for systematic innovation in construction companies. J. Constr. Eng. Manage. 140 (4) B4014001-1/ 8.
Perdomo-Ortiz, J., González-Benito, J., Galende, J., 2006. Total quality management as a forerunner of business innovation capability. Technovation 26 (10), 1170–1185.
Petnji, L.H., Marimon, F., Casadesus, M., 2011. Customer’s loyalty and perception of ISO 9001 in online banking. Ind. Manage. Data Syst. 111 (8), 1194 –1213.
Porter, M.E., 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. Free Press, New York.
Prajogo, D.I., Hong, S.W., 2008. The effect of TQM on performance in R&D environments: a perspective from South Korean firms. Technovation 28 (12), 855– 863.
Prajogo, D.I., Sohal, A.S., 2003. The relationship between TQM practices, quality performance, and innovation performance. Int. J. Qual. Reliab. Manage. 20
(8), 901–918.
Prajogo, D.I., Sohal, A.S., 2004. The multidimensionality of TQM practices in determining quality and innovation performance: an empirical examination.
Technovation 24 (6), 443–453.
Prajogo, D.I., Sohal, A.S., 2006. The integration of TQM and technology/R&D management in determining quality and innovation performance Omega. Int. J.
Manage. Sci. 34 (3), 296–312.
Roberts, R., 1998. Managing innovation: the pursuit of competitive advantage and the design of innovation intense environments. Res. Policy 27 (2), 159 –175. Roberts, P.W.,
1999. Product innovation, product-market competition and persistent pro fitability in the U.S. pharmaceutical industry. Strateg. Manage. J. 20
(7), 655–670.
Robins, M.M., Scarll, S.J., Key, P.E., 2006. Quality assurance in research laboratories. Accredit. Qual. Assur. 11 (5), 214 –223.
Romero-Cuevas, M., Reviriego-Barquilla, F., Marín-Martínez, M., Rodríguez-de- Fonseca, F., 2010. Gestión documental en un grupo de investigación biomédica. El
profesional de la Información 19 (3), 300–307.
Rossetto, S., 1995. Quality and Innovation: a conceptual model of their interaction. Total Qual. Manage. 6 (3), 221 –230.
Santos-Vijande, M.L., Alvarez-Gonzalez, L.I., 2007. TQM and firms performance: an EFQM excellence model research based survey. Int. J. Bus. Sci. Appl.
Manage. 2 (2), 21–41.
Sanzo, M.J., Santos, M.L., Vazquez, R.A., lvarez, L.I., 2003. The effect of market orientation on buyer- seller relationship satisfaction. Ind. Mark. Manage. 32 (4), 327–345.
Saunila, M., Ukko, J., 2012. A conceptual framework for the measurement of innovation capability and its effects. Baltic J. Manage. 7 (4), 355 –375.
Schumpeter, J.A., 1934. The Theory of Economic Development. Harvard University Press, USA.
Sitkin, S.M., Sutcliffe, K.M., Schroeder, R.G., 1994. Distinguishing control from learning in total quality management: a contingency perspective. Acad.
Manage. Rev. 19 (3), 537–564.
Teece, D.J., Pisano, G., Shuen, A., 1997. Dynamic capabilities and strategic management. Strateg. Manage. J. 18 (7), 509 –533.
Teece, D.J., 2007. Explicating dynamic capabilities: the nature and microfundations of (sustainable) enterprise performance. Strateg. Manage. J. 28 (13), 1319–1350.
Tenenhaus, M., Amato, S., Esposito, V., 2004. A global goodness-of-fit index for PLS structural equation modelling. XLII SIS Scientific Meeting Proceedings, CLEUP, Padova,
pp. 739–742.
Therrien, P., Doloreux, D., Chamberlin, T., 2011. Innovation novelty and (commercial) performance in the service sector: a Canadian firm-level analysis.
Technovation 31, 655–665.
Valcarcel, M., Rios, A., 2003. Quality assurance in analytical laboratories engaged in research and development activities. Accredit. Qual. Assur. 8 (2), 78 –81. Wright, C., Sturdy,
A., Wyliec, N., 2012. Management innovation through standardization: consultants as standardizers of organizational practice. Res. Policy
41 (3), 652–662.
Yepes, V., Pellicer, E., Alarcon, L.F., Correa, C.L., 2016. Creative innovation in Spanish construction firms. J. Prof. Issues Eng. Educ. Pract. 142 (1) 04015006-1/10.
Zairi, M., 1994. Innovation or innovativeness? Results of a benchmarking study. Total Qual. Manage. 5 (3), 27 –43.
Zollo, M., Winter, S.G., 2002. Deliberate learning and the evolution of dynamic capabilities. Org. Sci. 13 (3), 339 –351.

1 s2.0 S0923474816300352 Main

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

1 s2.0 S0923474816300352 Main

Transféré par

Droits d'auteur :

Formats disponibles

International Journal of Contemporary Hospitality Management

E-commerce performance in hospitality and

To copy this document: permissions@emeraldinsight.com

Access to this document was granted through an Emerald subscription provided by

*Related content and download information correct at time of download.

technique to identify key sources to uncover the E-commerce performance

networking business-related technology (Berkus, 2013), for example, is

framework governing E-commerce performance would emerge. Specifically, this

3. E-commence performance background

A temporal pattern emerged while researchers were closely examining the

3.1 Marketing and competition benefits of E-commerce

enterprises (SMEs) because it can enable efficient collaboration

3.2 Essential benefits of E-commerce that support strategy and

3.3 Business efficiency benefits of E-commerce

performance at a higher level (Barua and Mukhopadhyay, 2000).

4. E-commerce performance in hospitality and tourism

4.1 Marketing and competition benefits of E-commerce

processing system at one time. SABRE Technology Group introduced

4.2 Essential benefits of E-commerce that support strategy and

planning among partners of supply chains (Quayle, 2002), E-commerce was

stakeholder groups when addressing issues and questions related to E-commerce

Because E-commerce has become a necessity (Hua et al., 2015), E-commerce

as achieving competitive advantages by targeting niche markets, could also be

adoption as E-commerce evolves to incorporate more variations and dynamics

synthesizing 155 recent articles, this study proposes a framework of E- on could

- the organization’s internal optimal performance; and

- flows between the organization and its market environment,

The proposed framework was used to identify specific gaps in the

business adoption and its impact on relational quality in travel agency

the success of radical innovation projects within established firms: der

Enterprise Development, Vol. 9, pp. 331-348.

information technology: a study of electronic data interchange”,

economy”, Journal of Business and Industrial Marketing, Vol. 23 No. 6, emerging

Zhu, K. and Kraemer, K.L. (2002), “E-commerce metrics for net-enhanced

- SainaghiRuggero, Ruggero Sainaghi, BaggioRodolfo, Rodolfo Baggio, PhillipsPaul, Paul Phillips,

of Hospitality Management 73, 125-137. [Crossref]

Marketing & Management . [Crossref]

Electronic Commerce: Theory and Practice

Keywords: Electronic commerce, Internet, E-commerce advantages and disadvantages

information, electronic auctions, direct marketing to consumers. Electronic commerce can be

Table 1. Electronic Commerce Concept

E-business can be supported by all market participants (Misevičiūtė, 2001):

b E-commerce Advantages and Disadvantages

Table 2. E-commerce Advantages and Disadvantages

4. Statistical Data on E-commerce in Lithuania

- Nanehkaran, Y., A. (2013). An Introduction To Electronic Commerce. International journal of scientific

E-Commerce: A Short History Follow-up on Possible Trends

Article · October 2017

Valdeci Ferreira dos Santos Leandro Ricardo Sabino

Carlos Alberto Gonçalves

Neuromessenger View project

The user has requested enhancement of the downloaded file.

E-Commerce: A Short History Follow-up on Possible Trends

Published by Sciedu Press 130 ISSN 1923-4007 E-ISSN 1923-4015

Published by Sciedu Press 131 ISSN 1923-4007 E-ISSN 1923-4015

Table 1. Types of transactions possible in electronic commerce

Published by Sciedu Press 132 ISSN 1923-4007 E-ISSN 1923-4015

2.3 Main Tools in Electronic Commerce

Table 2. Advantages of electronic commerce in marketing approach

Published by Sciedu Press 133 ISSN 1923-4007 E-ISSN 1923-4015

Interactivity and trustworthiness for on-demand information

2.5 Challenges and Trends of Electronic Commerce

Published by Sciedu Press 134 ISSN 1923-4007 E-ISSN 1923-4015