Keller 1993

Conceptualizing, Measuring, and Managing Customer-Based Brand Equity
Author(s): Kevin Lane Keller

Source: Journal of Marketing, Vol. 57, No. 1 (Jan., 1993), pp. 1-22
Published by: American Marketing Association
Stable URL: http://www.jstor.org/stable/1252054 .
Accessed: 09/09/2013 02:26
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
American Marketing Association is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Marketing.
http://www.jstor.org
This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

All use subject to JSTOR Terms and Conditions
Kevin Lane Keller
Conceptualizing, and
Measuring,
ManagingCustomer-Based
Brand
Equity
The author presents a conceptual model of brand equity from the perspective of the individual consumer.
Customer-based brand equity is defined as the differential effect of brand knowledge on consumer re-
sponse to the marketing of the brand. A brand is said to have positive (negative) customer-based brand
equity when consumers react more (less) favorably to an element of the marketing mix for the brand
than they do to the same marketing mix element when it is attributed to a fictitiously named or unnamed
version of the product or service. Brand knowledge is conceptualized according to an associative network
memory model in terms of two components, brand awareness and brand image (i.e., a set of brand
associations). Customer-based brand equity occurs when the consumer is familiar with the brand and
holds some favorable, strong, and unique brand associations in memory. Issues in building, measuring,
and managing customer-based brand equity are discussed, as well as areas for future research.
M UCH attentionhas been devoted recently to the or divestiturepurposes. Several different methods of
concept of brandequity (Aaker and Biel 1992; brandvaluation have been suggested (Barwise et al.
Leuthesser 1988; Maltz 1991). Brandequity has been 1989; Wentz 1989). For example, InterbrandGroup
viewed from a variety of perspectives (Aaker 1991; has used a subjectivemultiplierof brandprofits based
Farquhar1989; Srivastavaand Shocker 1991; Tauber on the brand's performancealong seven dimensions
1988). In a general sense, brandequity is defined in (leadership, stability, market stability, interational-
terms of the marketingeffects uniquely attributableto ity, trend, support, and protection);GrandMetropol-
the brand-for example, when certain outcomes re- itan has valued newly acquiredbrandsby determining
sult from the marketingof a product or service be- the difference between the acquisitionprice and fixed
cause of its brand name that would not occur if the assets. Simon and Sullivan (1990) define brandequity
same productor service did not have that name. in termsof the incrementaldiscountedfuturecash flows
There have been two general motivations for thatwould resultfrom a producthavingits brandname
studying brandequity. One is a financiallybased mo- in comparisonwith the proceeds that would accrue if
tivation to estimate the value of a brand more pre- the same productdid not have thatbrandname. Based
cisely for accountingpurposes (in terms of asset val- on the financial market value of the company, their
uationfor the balancesheet) or for merger,acquisition,
estimation technique extracts the value of brand eq-
uity from the value of a firm's other assets.
KevinLaneKeller is Associate
Professor
of Marketing
andFletcherJones A second reason for studying brandequity arises
Faculty Scholarfor1992-1993,GraduateSchoolof Business,
Stanford
Univerity.Thisarticlewaswritten
whiletheauthorwasVisitingProfes- from a strategy-basedmotivationto improve market-
sor at the AustralianGraduateSchoolof Management,Universityof ing productivity. Given higher costs, greatercompe-
NewSouthWales,Sydney,Australia. HethanksDavidAaker,Sheri tition, and flatteningdemand in many markets,firms
Bridges, DeborahMacinnis,JohnRoberts,JohnRossiter,
RichardStae- seek to increase the efficiency of their marketingex-
lin,JenniferAaker,andtheanonymous JMreviewers fordetailed,
con-
structivecomments. penses. As a consequence, marketersneed a more
thoroughunderstandingof consumerbehavioras a ba-
Journal of Marketing
Vol. 57 (January 1993), 1-22 Customer-Based
BrandEquity
/ 1

sis for making better strategic decisions about target in memory. Then the conceptof customer-basedbrand
marketdefinition and productpositioning, as well as equity is considered in more detail by discussion of
better tactical decisions about specific marketingmix how it can be built, measured, and managed. After
actions. Perhapsa firm's most valuable asset for im- the conceptual framework is summarized, areas for
proving marketingproductivityis the knowledge that future researchare identified.
has been createdabout the brandin consumers' minds
from the firm's investmentin previous marketingpro-
grams. Financialvaluationissues have little relevance Brand Knowledge
if no underlyingvalue for the brandhas been created
or if managersdo not know how to exploit that value Background
by developing profitablebrandstrategies. A brandcan be defined as "a name, term, sign, sym-
The goal of this article is to assist managersand bol, or design, or combinationof them which is in-
researcherswho are interestedin the strategicaspects tendedto identify the goods and services of one seller
of brandequity. Specifically, brandequity is concep- or groupof sellers and to differentiatethem from those
tualized from the perspective of the individual con- of competitors"(Kotler 1991; p. 442). These individ-
sumerand a conceptualframeworkis providedof what ual brand components are here called "brandidenti-
consumers know about brandsand what such knowl- ties" and their totality "the brand."Some basic mem-
edge implies for marketingstrategies.Customer-based ory principles can be used to understandknowledge
brandequityis definedas the differentialeffect of brand about the brand and how it relates to brand equity.
knowledge on consumerresponse to the marketingof The importanceof knowledge in memoryto consumer
the brand. That is, customer-basedbrand equity in- decision making has been well documented (Alba,
volves consumers' reactionsto an element of the mar- Hutchinson, and Lynch 1991). Understandingthe
keting mix for the brandin comparisonwith their re- content and structureof brandknowledge is important
actions to the same marketingmix element attributed because they influence what comes to mind when a
to a fictitiously named or unnamed version of the consumer thinks about a brand-for example, in re-
product or service. Customer-basedbrandequity oc- sponse to marketingactivity for that brand.
curs when the consumeris familiarwith the brandand Most widely acceptedconceptualizationsof mem-
holds some favorable, strong, and unique brand as- ory structureinvolve some type of associative model
sociations in memory. formulation(Anderson 1983; Wyer and Srull 1989).
Conceptualizingbrand equity from this perspec- For example, the "associativenetworkmemorymodel"
tive is useful because it suggests both specific guide- views semantic memory or knowledge as consisting
lines for marketing strategies and tactics and areas of a set of nodes and links. Nodes are stored infor-
where research can be useful in assisting managerial mation connected by links that vary in strength. A
decision making. Two importantpoints emerge from "spreadingactivation"process from node to node de-
this conceptualization.First, marketersshould take a termines the extent of retrieval in memory (Collins
broad view of marketingactivity for a brandand rec- and Loftus 1975; Raaijmakers and Shiffrin 1981;
ognize the various effects it has on brandknowledge, Ratcliff and McKoon 1988). A node becomes a po-
as well as how changesin brandknowledgeaffect more tentialsource of activationfor othernodes eitherwhen
traditionaloutcome measures such as sales. Second, external informationis being encoded or when inter-
marketersmust realize that the long-term success of nal informationis retrievedfrom long-termmemory.
all future marketingprogramsfor a brand is greatly Activation can spread from this node to other linked
affected by the knowledge about the brandin memory nodes in memory.When the activationof anothernode
thathas been establishedby the firm's short-termmar- exceeds some threshold level, the informationcon-
keting efforts. In short, because the content and struc- tained in that node is recalled. Thus, the strengthof
ture of memory for the brand will influence the ef- association between the activatednode and all linked
fectiveness of futurebrandstrategies, it is critical that nodes determines the extent of this "spreadingacti-
managers understandhow their marketing programs vation" and the particularinformationthat can be re-
affect consumer learning and thus subsequent recall trieved from memory. For example, in consideringa
for brand-relatedinformation. soft drink purchase, a consumer may think of Pepsi
The next section provides a conceptualizationof because of its strong associationwith the productcat-
brandknowledge by applyingsome basic memory no- egory. Consumerknowledge most strongly linked to
tions. Brand knowledge is defined in terms of two Pepsi should also then come to mind, such as per-
components,brandawarenessand brandimage. Brand ceptions of its taste, sugar and caffeine content, or
awarenessrelates to brandrecall and recognition per- even recalled images from a recent advertisingcam-
formanceby consumers. Brandimage refers to the set paign or past productexperiences.
of associationslinked to the brandthatconsumershold Consistent with an associative network memory
2 / Journalof Marketing,January1993

model, brand knowledge is conceptualized as con- be a member of the consideration set (Baker et al.
sisting of a brand node in memory to which a variety 1986; Nedungadi 1990), the handful of brands that
of associations are linked. Given this conceptualiza- receive serious consideration for purchase. Second,
tion, the key question is, what properties do the brand brand awareness can affect decisions about brands in
node and brand associations have? As developed here, the consideration set, even if there are essentially no
the relevant dimensions that distinguish brand knowl- other brand associations. For example, consumers have
edge and affect consumer response are the awareness been shown to adopt a decision rule to buy only fa-
of the brand (in terms of brand recall and recognition) miliar, well-established brands (Jacoby, Syzabillo, and
and the favorability, strength, and uniqueness of the Busato-Schach 1977; Roselius 1971). In low involve-
brand associations in consumer memory. These di- ment decision settings, a minimum level of brand
mensions are affected by other characteristics of and awareness may be sufficient for product choice, even
relationships among the brand associations. For ex- in the absence of a well-formed attitude (Bettman and
ample, factors related to the type of brand association Park 1980; Hoyer and Brown 1990; Park and Lessig
(such as its level of abstraction and qualitative nature) 1981). The elaboration likelihood model (Petty and
and the congruity among brand associations, among Cacioppo 1986) suggests that consumers may base
others, affect the favorability, strength, and unique- choices on brand awareness considerations when they
ness of brand associations. To simplify the discus- have low involvement, which could result from either
sion, emphasis is placed on the brand name compo- a lack of consumer motivation (i.e., consumers do not
nent of the brand identities, defined as "that part of a care about the product or service) or a lack of con-
brand which can be vocalized" (Kotler 1991, p. 442), sumer ability (i.e., consumers do not know anything
though other components of the brand identities (e.g., else about the brands). Finally, brand awareness af-
brand logo or symbol) are considered also. fects consumer decision making by influencing the
formation and strength of brand associations in the
Brand Awareness
brand image. A necessary condition for the creation
The first dimension distinguishing brand knowledge of a brand image is that a brand node has been estab-
is brand awareness. It is related to the strength of the lished in memory, and the nature of that brand node
brand node or trace in memory, as reflected by con- should affect how easily different kinds of informa-
sumers' ability to identify the brand under different tion can become attached to the brand in memory.
conditions (Rossiter and Percy 1987). In other words,
how well do the brand identities serve their function? Brand Image
In particular, brand name awareness relates to the
likelihood that a brand name will come to mind and Though brand image long has been recognized as an
the ease with which it does so. Brand awareness con- importantconcept in marketing (e.g., Gardnerand Levy
sists of brand recognition and brand recall perfor- 1955), there is less agreement on its appropriate def-
mance. Brand recognition relates to consumers' abil- inition (Dobni and Zinkhan 1990). Consistent with
definitions by Herzog (1963) and Newman (1957),
ity to confirm prior exposure to the brand when given
the brand as a cue. In other words, brand recognition among others, and an associative network memory
model of brand knowledge, brand image is defined
requires that consumers correctly discriminate the brand
as having been seen or heard previously. Brand recall here as perceptions about a brand as reflected by the
relates to consumers' ability to retrieve the brand when brand associations held in consumer memory. Brand
associations are the other informational nodes linked
given the product category, the needs fulfilled by the
to the brand node in memory and contain the meaning
category, or some other type of probe as a cue. In
other words, brand recall requires that consumers cor- of the brand for consumers. The favorability, strength,
and uniqueness of brand associations are the dimen-
rectly generate the brand from memory. The relative
sions distinguishing brand knowledge that play an im-
importance of brand recall and recognition depends on
the extent to which consumers make decisions in the portant role in determining the differential response
store (where they potentially may be exposed to the that makes up brand equity, especially in high in-
volvement decision settings. Before considering those
brand) versus outside the store, among other factors
(Bettman 1979; Rossiter and Percy 1987). Brand rec- dimensions, it is useful to examine the different types
of brand associations that may be present in consumer
ognition may be more important to the extent that
product decisions are made in the store. memory.
Brand awareness plays an important role in con- Types of brand associations. Brand associations
sumer decision making for three major reasons. First, take different forms. One way to distinguish among
it is important that consumers think of the brand when brand associations is by their level of abstraction (Alba
they think about the product category. Raising brand and Hutchinson 1987; Chattopadhyay and Alba 1988;
awareness increases the likelihood that the brand will Johnson 1984; Russo and Johnson 1980)-that is, by
Customer-Based
BrandEquity/ 3

how much informationis summarizedor subsumedin type of activity (formal or informal), among other as-
the association. Along this dimension, brand associ- pects. User and usage image attributescan also pro-
ations can be classified into three majorcategories of duce brandpersonalityattributes.Plummer(1985) as-
increasing scope: attributes, benefits, and attitudes. serts that one component of brand image is the
Severaladditionaldistinctionscan be madewithinthese personality or characterof the brand itself. He sum-
categories according to the qualitative nature of the marizes research demonstratingthat brands can be
association. characterized by personality descriptors such as
Attributesare those descriptive featuresthat char- "youthful," "colorful," and "gentle." These types of
acterizea productor service-what a consumerthinks associations seem to arise most often as a result of
the productor service is or has and what is involved inferences about the underlying user or usage situa-
with its purchase or consumption. Attributescan be tion. Brand personality attributes may also reflect
categorized in a variety of ways (Myers and Shocker emotions or feelings evoked by the brand.
1981). Here, attributesare distinguishedaccordingto Benefits are the personal value consumers attach
how directly they relate to productor service perfor- to the productor service attributes-that is, what con-
mance. Product-related attributes are defined as the sumers think the productor service can do for them.
ingredients necessary for performing the product or Benefits can be furtherdistinguishedinto three cate-
service function sought by consumers. Hence, they gories accordingto the underlyingmotivationsto which
relate to a product's physical composition or a ser- they relate (Park, Jaworski, and Maclnnis 1986): (1)
vice's requirements.Product-relatedattributesvary by functional benefits, (2) experientialbenefits, and (3)
product or service category. Non-product-related at- symbolic benefits. Functional benefits are the more
tributes are defined as external aspects of the product intrinsic advantages of product or service consump-
or service that relate to its purchaseor consumption. tion and usually correspondto the product-relatedat-
The four main types of non-product-relatedattributes tributes.These benefits often are linked to fairly basic
are (1) price information, (2) packaging or product motivations, such as physiological and safety needs
appearanceinformation,(3) user imagery (i.e., what (Maslow 1970), and involve a desire for problemre-
type of person uses the product or service), and (4) moval or avoidance(Fennell 1978; Rossiterand Percy
usage imagery (i.e., where and in what types of sit- 1987). Experientialbenefitsrelate to what it feels like
uations the productor service is used). to use the productor service and also usually corre-
Because product-relatedattributesare more com- spond to the product-relatedattributes.These benefits
monly acknowledged, only non-product-relatedattri- satisfy experiential needs such as sensory pleasure,
butes are elaboratedhere. The price of the productor variety, and cognitive stimulation. Symbolic benefits
service is considered a non-product-relatedattribute are the more extrinsic advantagesof product or ser-
because it representsa necessary step in the purchase vice consumption. They usually correspondto non-
process but typically does not relate directly to the product-related attributesand relateto underlyingneeds
product performanceor service function. Price is a for social approvalor personal expression and outer-
particularly important attribute association because directedself-esteem. Hence, consumersmay value the
consumers often have strong beliefs about the price prestige, exclusivity, or fashionabilityof a brandbe-
and value of a brandand may organize their product cause of how it relates to their self-concept (Solomon
category knowledge in terms of the price tiers of dif- 1983). Symbolic benefits should be especially rele-
ferent brands(Blattbergand Wisniewski 1989). Sim- vant for socially visible, "badge"products.
ilarly, packaging is considered part of the purchase Brand attitudes are defined as consumers' overall
and consumptionprocess but, in most cases, does not evaluationsof a brand(Wilkie 1986). Brandattitudes
directly relate to the necessary ingredientsfor product are importantbecause they often form the basis for
performance. User and usage imagery attributescan consumer behavior (e.g., brandchoice). Though dif-
be formed directly from a consumer's own experi- ferent models of brandattitudeshave been proposed,
ences and contactwith brandusersor indirectlythrough one widely accepted approachis based on a multiat-
the depiction of the targetmarketas communicatedin tribute formulation in which brand attitudes are a
brand advertising or by some other source of infor- function of the associated attributesand benefits that
mation (e.g., word of mouth). Associations of a typ- are salient for the brand. Fishbein and Ajzen (1975;
ical branduser may be based on demographicfactors Ajzen and Fishbein 1980) proposed what has been
(e.g., sex, age, race, and income), psychographicfac- probably the most influential multiattributemodel to
tors (e.g., according to attitudestoward career, pos- marketing (Bettman 1986). This expectancy-value
sessions, the environment, or political institutions), model views attitudesas a multiplicativefunction of
and other factors. Associations of a typical usage sit- (1) the salient beliefs a consumerhas about the prod-
uation may be based on the time of day, week, or uct or service(i.e., the extentto whichconsumersthink
year, the location (inside or outside the home), or the the brandhas certainattributesor benefits) and (2) the

evaluative judgment of those beliefs (i.e., how good Not all associations for a brand, however, will be
or bad it is that the brandhas those attributesor ben- relevantand valued in a purchaseor consumptionde-
efits). cision. For example, consumers often have an asso-
Brand attitudes can be related to beliefs about ciation in memory from the brand to the product or
product-relatedattributesand the functional and ex- package color. Though this association may facilitate
perientialbenefits, consistent with work on perceived brandrecognition or awareness or lead to inferences
quality (Zeithaml 1988). Brand attitudescan also be about product quality, it may not always be consid-
related to beliefs about non-product-relatedattributes ered a meaningful factor in a purchase decision.
and symbolic benefits (Rossiterand Percy 1987), con- Moreover, the evaluations of brandassociations may
sistent with the functional theory of attitudes (Katz be situationally or context-dependentand vary ac-
1960; Lutz 1991), which maintainsthat attitudescan cording to consumers' particulargoals in their pur-
serve a "value-expressive"function by allowing in- chase or consumption decisions (Day, Shocker, and
dividuals to express their self-concepts. Because it is Srivastava 1979). An association may be valued in
difficult to specify correctly all of the relevant attri- one situationbut not another(Millerand Ginter1979).
butes and benefits, researchersbuilding multiattribute For example, speed and efficiency of service may be
models of consumer preferencehave included a gen- very importantwhen a consumer is under time pres-
eral componentof attitudetowardthe brandthat is not sure but may have little impact when a consumer is
capturedby the attributeor benefit values of the brand less hurried.
(Park 1991; Srinivasan 1979). Moreover, as noted
previously, researchalso has shown that attitudescan Strength of brand associations. Associations can
be formedby less thoughtfuldecisionmaking(Chaiken be characterizedalso by the strengthof connection to
the brandnode. The strengthof associations depends
1986; Petty and Cacioppo 1986)-for example, on the
basis of simple heuristics and decision rules. If con- on how the informationentersconsumermemory(en-
sumerslack either the motivationor ability to evaluate coding) and how it is maintainedas partof the brand
the product or service, they may use signals or "ex- image (storage). Strength is a function of both the
trinsic cues" (Olson and Jacoby 1972) to infer product amount or quantityof processing the informationre-
ceives at encoding (i.e., how much a person thinks
or service quality on the basis of what they do know
aboutthe information)and the natureor qualityof the
aboutthe brand(e.g., productappearancesuch as color
or scent). processing the informationreceives at encoding (i.e.,
the mannerin which a person thinks about the infor-
Thus, the different types of brand associations
mation).Forexample,the levels- or depth-of-processing
making up the brandimage include product-relatedor approach(Craik and Lockhart1972; Craik and Tulv-
non-product-relatedattributes; functional, experien- ing 1975; Lockhart, Craik, and Jacoby 1976) main-
tial, or symbolic benefits; and overall brandattitudes. tains that the more the meaning of informationis at-
These associations can vary accordingto their favor- tended to during encoding, the strongerthe resulting
ability, strength, and uniqueness. associations in memory will be. Thus, when a con-
Favorability of brand associations. Associations sumer actively thinks about and "elaborates"on the
differ accordingto how favorablythey are evaluated. significanceof productor service information,stronger
The success of a marketingprogramis reflectedin the associations are created in memory. This strength,in
creationof favorablebrandassociations-that is, con- turn,increasesboth the likelihoodthat informationwill
sumers believe the brand has attributesand benefits be accessible and the ease with which it can be re-
that satisfy their needs and wants such that a positive called by "spreadingactivation."
overall brandattitudeis formed. Cognitive psychologists believe memory is ex-
MacKenzie (1986) summarizesresearchevidence tremely durable, so that once informationbecomes
suggesting that the "evaluativejudgment"component stored in memory its strength of association decays
of expectancy-value models of attitude (i.e., con- very slowly (Loftus and Loftus 1980). Though "avail-
sumer perceptionsof the favorabilityof an attribute) able" and potentially retrievable in memory, infor-
is both conceptually and empirically related to attri- mation may not be "accessible" and easily retrieved
bute importance.Specifically, attributeimportancehas withoutstronglyassociatedremindersor retrievalcues
been equatedwith polarityof attributeevaluation(Ajzen (Tulving and Psotka 1971). Thus, the particularas-
and Fishbein 1980; Fishbeinand Ajzen 1975). In other sociations for a brand that are salient and "come to
words, consumersare unlikely to view an attributeor mind" depend on the context in which the brand is
benefit as very good or bad if they do not also con- considered. The larger the numberof cues linked to
sider it to be very important.Hence, it is difficult to a piece of information,however, the greaterthe like-
create a favorable association for an unimportantat- lihood thatthe informationcan be recalled(Isen 1992).
tribute. Uniqueness of brand associations. Brand associ-
Customer-Based
BrandEquity
/ 5

ations may or may not be sharedwith othercompeting America or some other marketleader to be the best
brands. The essence of brandpositioning is that the example of a bank.
brand has a sustainable competitive advantage or Because the brand is linked to the product cate-
"unique selling proposition"that gives consumers a gory, some product category associations may be-
compelling reason for buying that particularbrand come linked to the brand, either in terms of specific
(Aaker 1982; Ries and Trout 1979; Wind 1982). These beliefs or overall attitudes.Productcategory attitudes
differences may be communicatedexplicitly by mak- can be a particularlyimportantdeterminantof con-
ing direct comparisons with competitors or may be sumer response. For example, if a consumer thinks
highlighted implicitly without stating a competitive banksare basically "unfriendly"and "bad,"he or she
point of reference. Furthermore,they may be based probablywill have similarlyunfavorablebeliefs about
on product-relatedor non-product-relatedattributesor and attitudetowardany particularbank simply by vir-
functional, experiential, or image benefits. tue of its membershipin the category. Thus, in almost
The presenceof stronglyheld, favorablyevaluated all cases, some productcategory associationsthat are
associations that are unique to the brand and imply linked to the brandare sharedwith otherbrandsin the
superiorityover other brands is critical to a brand's category. Note that the strengthof the brandassoci-
success. Yet, unless the brandhas no competitors,the ations with the product category is an important
brand will most likely share some associations with determinant of brand awareness (Nedungadi and
otherbrands.Sharedassociationscan help to establish Hutchinson 1985; Ward and Loken 1986).
category membership(Maclnnis and Nakamoto 1991) Competitive overlap with other brandsassociated
and define the scope of competition with other prod- with the productcategorydoes have a downside, how-
ucts and services (Sujanand Bettman 1989). Research ever, in terms of possible consumer confusion. For
on noncomparablealternatives (Bettman and Sujan example, Keller (1987) and Burke and Srull (1988)
1987; Johnson 1984; Park and Smith 1989) suggests have shown that the numberof competingbrandsad-
that even if a branddoes not face direct competition vertising in a productcategory can affect consumers'
in its productcategory,andthusdoes not shareproduct- ability to recall communicationeffects for a brandby
related attributeswith other brands, it can still share creating"interference" in memory.Keller (1991b) also
more abstract associations and face indirect compe- showed that though these interferenceeffects can pro-
titionin a morebroadlydefinedproductcategory.Thus, duce lower brandevaluations, they can be overcome
though a railroadmay not compete directly with an- throughthe use of ad retrievalcues-that is, distinc-
other railroad, it still competes indirectly with other tive ad execution informationthat is present when a
forms of transportation,such as airlines, cars, and consumer actually makes a brandevaluation(e.g., at
buses. the point of purchase).
A productor service categorycan be characterized Interaction among characteristics of brand asso-
also by a set of associations that include specific be- ciations. The level of abstractionand qualitativena-
liefs about any memberin the category in additionto ture of brand associations should affect their favora-
overall attitudestoward all members in the category. bility, strength,and uniqueness. For example, image-
These beliefs include many of the product-relatedat- relatedattributes,such as user type or usage situation,
tributes for the relevant brands, as well as more de- may easily create unique associations. Abstract as-
scriptive attributesthat do not necessarily relate to sociations (e.g., benefits and especially attitudes), in
product or service performance(e.g., the color of a contrast, tend to be inherently more evaluative be-
product, such as red for ketchup). Certain attributes cause of the embedded meaning they contain. Be-
or benefits may be considered "prototypical"and es- cause of this evaluative nature, abstractassociations
sential to all brands in the category, and a specific tend to be more durableand accessiblein memorythan
brand may be considered an "exemplar"that is most the underlying attributeinformation(Chattopadhyay
representative of the productor servicecategory(Cohen and Alba 1988). In fact, brandattitudesmay be stored
andBasu 1987;NedungadiandHutchinson1985;Rosch and retrieved in memory separatelyfrom the under-
and Mervis 1975; Ward and Loken 1986). For ex- lying attributeinformation(Lynch, Mamorstein,and
ample, consumersmight expect a runningshoe to pro- Weigold 1988).
vide supportand comfort, be built well enough to last One importantreason for considering brand atti-
throughrepeatedwearings, and so on, and they may tudes to be a brand association is that they can vary
believe that Nike or some other leading brand best in strength(Farquhar1989). Attitudestrengthhas been
representsa runningshoe. Similarly,consumersmight measured by the reaction time needed to evaluative
expect a bank to offer a variety of checking and sav- queries about the attitudeobject (Fazio et al. 1986).
ings accounts, provide branchand electronic delivery Individualswho can evaluatean attitudeobjectquickly
services, and so on, and they may consider Bank of are assumed to have a highly accessible attitude.Re-

searchhas shown thatattitudesformedfrom directbe- information-though the unexpectedness of infor-
havior or experience are more accessible than atti- mation inconsistent in meaning with the brandsome-
tudesbasedon informationor indirectformsof behavior times can lead to moreelaborateprocessingandstronger
(Fazio and Zanna 1981). Highly accessible brandat- associationsthaneven consistentinformation(Houston,
titudes are more likely to be activated spontaneously Childers, and Heckler 1987; Myers-Levy and Tybout
upon exposureto the brandand guide subsequentbrand 1989; Wyer and Srull 1989). That is, consumersmay
choices (Bergerand Mitchell 1989; Fazio, Powell, and have expectations as to the likelihood that a product
Williams 1989). or service has a particularassociationgiven that it has
Figure 1 summarizes the dimensions of brand some otherassociation(Bettman,John, and Scott 1986;
knowledge. Sujan 1985). These expectations should affect con-
sumers' ability to learn new brand information. For
Congruence of brand associations. The favora- example, if a running shoe has a brand association
bility and strengthof a brandassociation can be af- with "very durable and long-lasting," presumablyit
fected by other brand associations in memory. Con- would be easier to establish an association with "all
gruence is defined as the extent to which a brand weather"than with "stylish." As noted subsequently,
association shares content and meaning with another these expectationsalso may result in the formationof
brandassociation. The congruence of brandassocia- inferredbrand associations. Thus, the strengthof an
tions should affect (1) how easily an existing associ- association should depend on how its content relates
ation can be recalled and (2) how easily additional to the content of other associations for the brand.
associations can become linked to the brandnode in The congruence among brand associations deter-
memory. In general, informationthat is consistent in mines the "cohesiveness" of the brand image-that
meaning with existing brand associations should be is, the extent to which the brandimage is character-
more easily learned and rememberedthan unrelated ized by associationsor subsetsof associationsthatshare
FIGURE 1
Dimensions of Brand Knowledge
Customer-Based
BrandEquity/ 7

meaning. The cohesiveness of the brand image may brand with the response to the same marketingof a
determine consumers' more holistic or gestalt reac- fictitiously named or unnamedversion of the product
tions to the brand. Moreover, a "diffuse" brand im- or service. Brand knowledge is defined in terms of
age, where there is little congruenceamong brandas- brandawareness and brand image and is conceptual-
sociationsfor consumers,can presentseveral potential ized accordingto the characteristicsand relationships
problemsfor marketers.First, consumersmay be con- of brandassociationsdescribedpreviously. Consumer
fused as to the meaning of the brand and, because response to marketing is defined in terms of consumer
they do not have as much informationto which new perceptions, preferences, and behavior arising from
informationcan be easily related,new associationsmay marketingmix activity (e.g., brandchoice, compre-
be weakerand possibly less favorable(Heckler,Keller, hension of copy points from an ad, reactionsto a cou-
and Houston 1992). Moreover, because any one as- pon promotion, or evaluations of a proposed brand
sociation shares little meaning with other associa- extension).
tions, brandassociationsmay be more easily changed Thus, accordingto this definition, a brand is said
by competitive actions. Finally, anotherproblemwith to have positive (negative) customer-based brand eq-
a diffuse brand image is the greater likelihood that uity if consumers react more (less) favorably to the
consumerswill discount or overlook some potentially product, price, promotion, or distribution of the brand
relevant brand associations in making brand deci- than they do to the same marketing mix element when
sions. For example, research on "part-listcuing ef- it is attributed to a fictitiously named or unnamed ver-
fects" has shown that recall of informationcan inhibit sion of the product or service. Favorable consumer
and lower the recall of other informationfrom mem- responseand positive customer-basedbrandequity, in
ory (Alba and Chattopadhyay1985a,b, 1986; Hoch turn, can lead to enhanced revenue, lower costs, and
1984; Keller 1991a). Hence, only some of the poten- greaterprofits. Brandknowledge is centralto this def-
tially retrievablebrand associations actually may be inition. In particular,the favorability, strength, and
recalled when the brand image is not cohesive and uniquenessof the brandassociationsplay a criticalrole
consistent. in determiningthe differentialresponse. If the brand
is seen by consumersto be the same as a prototypical
version of the productor service in the category, their
Customer-Based Brand Equity
response should not differ from their response to a
As noted, brandequity has been defined in a variety hypotheticalproductor service; if the brandhas some
of ways, depending on the particularpurpose. Be- salient, unique associations, those responses should
cause the goal of this article is to facilitate the de- differ. The actual natureof how the responses differ
velopment of more effective marketingstrategiesand depends on consumers' evaluations of these associa-
tactics, the focus is on brandeffects on the individual tions, as well as the particularmarketingmix element
consumer. The advantage of conceptualizing brand under consideration.Thus, establishing brandaware-
equity from this perspective is that it enables man- ness and a "positive brand image" (i.e., favorable,
agers to consider specifically how their marketing strong, and unique brand associations) in consumer
programimproves the value of their brands. Though memorycreatesdifferenttypes of customer-basedbrand
the eventual goal of any marketingprogramis to in- equity, depending on what marketingmix element is
crease sales, it is first necessary to establish knowl- under consideration. A brief discussion highlighting
edge structuresfor the brand so that consumers re- some relevant considerations for each of these ele-
spond favorably to marketingactivity for the brand. ments follows.
The preceding section provides a detailed framework Fundamentally,high levels of brandawarenessand
of brand knowledge. In this section, that framework a positive brandimage should increasethe probability
is used to consider in more detail how knowledge af- of brandchoice, as well as producegreaterconsumer
fects consumer response to the marketingof a brand (and retailer) loyalty and decrease vulnerability to
by defining customer-basedbrand equity and exam- competitivemarketingactions.Thus, the view of brand
ining how it is built, measured, and managed. loyalty adopted here is that it occurs when favorable
beliefs and attitudes for the brand are manifested in
Defining Customer-Based Brand Equity
repeat buying behavior. Some of these beliefs may
Customer-based brand equity is defined as the differ- reflect the objective reality of the product, in which
ential effect of brand knowledge on consumer re- case no underlyingcustomer-basedbrandequity may
sponse to the marketing of the brand. Three impor- be present, but in other cases they may reflect favor-
tantconceptsare includedin the definition:"differential able, strong, and unique associations that go beyond
effect," "brandknowledge," and "consumerresponse the objective reality of the product(Park 1991).
to marketing." Differential effect is determined by High levels of brandawarenessand a positivebrand
comparing consumer response to the marketingof a image also have specific implicationsfor the pricing,
8 / Journalof Marketing,
January1993

distribution, and promotion activities related to the Choosing brand identities. To see how the initial
brand.First, a positive image should enable the brand choice of the brandidentities can affect brandequity,
to command larger margins and have more inelastic consider the choice of a brandname. A varietyof cri-
responses to price increases. The most importantas- teria have been suggested for the selection of a brand
pect of the brand image that affects consumer re- name (e.g., Aaker1991;Kotler1991;Robertson1989).
sponses to prices is probably overall brand attitude. They generally can be classified accordingto whether
Consumers with a strong, favorable brand attitude they help enhance brand awareness or facilitate the
should be more willing to pay premiumprices for the linkage of brandassociations.
brand (Starr and Rubinson 1978). Similarly, a posi- Alba and Hutchinson(1987) give an extensive dis-
tive image should result in increasedconsumersearch cussion of psychological principlesthat can be useful
(Simonson, Huber, and Payne 1988) and a willing- in understanding how the choice of a nameaffectsbrand
ness to seek out distributionchannels for the product recall and recognition processes. Some criteriaoften
or service. Finally, high levels of brandawarenessand noted by otherresearchersare thatbrandnames should
a positive brand image can increase marketingcom- be simple, familiar, and distinctive, along the follow-
municationeffectiveness. All aspects of the brandim- ing lines. To enhancethe likelihood of successful pro-
age are relevant in determiningconsumerresponse to cessing at encoding, the brand name should be easy
advertisingand promotion. For example, several au- to comprehend,pronounce,and spell. In fact, market
thorsnote thatadvertisingresponseand decay patterns researcherssometimes evaluate the "flicker percep-
are a function of consumers' attitudes and behavior tion" of brandnames (i.e., how quickly a brandname
towardthe brand(Ray 1982;Rossiterand Percy 1987). can be perceived and understoodwhen exposed only
They maintainthat consumerswho are positively pre- for an instant) to assess consumer learning of candi-
disposed toward a brand may require fewer ad ex- date brand names (Dolan 1985). To improve con-
posures to meet communicationobjectives. Similarly, sumer learningof the brand,mnemonic factors (e.g.,
one could argue that strong attributeor benefit asso- One-A-Day) and vivid words are often employed that
ciationsfor the brandrequireless reinforcementthrough have rich evaluativeor experientialimagery(Robertson
marketingcommunications. 1987; but see Myers-Levy 1989). Similarly, the use
In these differentways, customer-basedbrandeq- of a familiar word should be advantageousbecause
uity is enhanced by creating favorable response to much informationis present in memory to which the
pricing, distribution, advertising, and promotion ac- namerelates.Finally, a distinctiveword is often sought
tivity for the brand. Moreover, a familiarbrandwith to attractattentionand reduce confusion among com-
a positive brand image can also yield licensing op- peting brands.
portunities (i.e., the brand name is used by another These different choice criteria for a brand name
firm on one of its products)and supportbrandexten- are not necessarily mutually compatible, and it may
sions (i.e., a firm uses an existing brandname to in- be difficult to choose names that are simple, familiar,
troducea new productor service),two importantgrowth and distinctive. Moreover, factors affecting the ease
strategies for firms in recent years. Licensing can be with which a brandname is recalled differ from fac-
a valuable source of royalty income, as evidenced by tors affecting the ease with which a brand name is
the substantialmerchandisingefforts in recent years, recognized. For example, past researchsuggests that
and typically has been employed when brandassoci- high frequency words (accordingto conventionaluse
ations have strong user imagery or brandpersonality in language) are easier to recall than low frequency
attributes.A more substantialinvestmentand risk pro- words, but low frequencywords may be easier to rec-
file for the company, however, is requiredwith brand ognize thanhigh frequencywords (Gregg 1976; Lynch
extensions. Because of theirpotentiallylasting effects and Srull 1982). This finding suggests that choosing
on consumer knowledge and the effectiveness of fu- a familiarword representinga well-known concept or
ture marketingactivity, brandextensions are consid- some othercommonobjector propertyas a brandname
ered in more detail in the section on managing may facilitate brandrecall, but that choosing a more
customer-basedbrandequity. unusual or distinctive word may facilitate brandrec-
ognition. Deciding whether to emphasize recall or
Building Customer-Based Brand Equity
recognition propertiesin choosing a brandname de-
Building customer-based brand equity requires the pends on managerialprioritiesconcerning the extent
creationof a familiarbrandthathas favorable, strong, of consumers' in-storeprocessing for the product,the
and unique brandassociations. This can be done both natureof the competitive environment,and so on.
throughthe initial choice of the brandidentities, such The choice of a brand name may also affect the
as the brandname, logo, or symbol, and throughthe favorability, strength, and uniqueness of brandasso-
integrationof the brand identities into the supporting ciations. The suggestiveness or meaningfulnessof the
marketingprogram. brand name should affect how easily brand associa-
Customer-Based
BrandEquity/ 9

tions are created. The brandname can be chosen to product or service specifications themselves are the
suggest semantically (1) the productor service cate- primary basis for the product-relatedattributeasso-
gory or (2) importantattributesor benefits within that ciations and determinea consumer's fundamentalun-
category.The first considerationshould enhancebrand derstanding of what the product or service means.
name awareness and the identificationwith the prod- Similarly, the pricingpolicy for the branddirectlycre-
uct category. The second consideration affords two ates associations to the relevantprice tier or level for
importantbenefits. First, even in the absence of any the brandin the productcategory, as well as its cor-
marketingactivity, the semanticmeaningof a sugges- respondingprice volatility or variance(e.g., in terms
tive brand name may enable consumers to infer cer- of the frequency and magnitudeof discounts).
tain attributesand benefits. For example, consumers The marketingcommunicationefforts by the firm,
could assume on the basis of the names alone that in contrast, afford a flexible means of shaping con-
Daybreak cereal is wholesome and natural, Chief sumerperceptionsof the productor service. At times,
laundrydetergentremoves tough stains, and Diamond marketersmay have to translateattributesinto their
toothpaste whitens and brightens teeth. Second, a correspondingbenefits for consumers throughadver-
suggestive brand name may facilitate marketingac- tising or other forms of communication. Marketing
tivity designedto link certainassociationsto the brand. communicationsalso may be helpful in creating user
Ideally, the brandname can be effectively supported and usage imagery attributes.The strength of brand
through marketingcommunicationsand a distinctive associations from communicationeffects depends on
slogan that ties together the brand name and its po- how the brand identities are integratedinto the sup-
sitioning. porting marketing program-for example, the posi-
Similar choice criteria apply to the other brand tion and prominenceof the brandidentities in a tele-
identities, such the brandlogo or symbol. Moreover, vision ad (Keller 1992). Though delaying brand
another importantobjective is to choose the various identificationuntil the end of a television commercial
brandidentitiesto be mutuallyreinforcingso thatthey may increase attention levels during commercial ex-
interactpositively to satisfy these criteria. Neverthe- posure, resultingin many communicationeffects being
less, althoughthe judicious choice of brandidentities stored in memory (e.g., ad execution and brandclaim
can contributesignificantly to customer-basedbrand information, as well as affective and cognitive re-
equity, the primaryinputcomes from supportingmar- sponses to that information),it may also produceweak
keting activities for the brandand the variousproduct, links from these effects to the brand. Finally, word-
price, advertising, promotion, and distributiondeci- of-mouth and other social influences also play an im-
sions, as discussed next. portantrole, especially for user and usage imageryat-
Developing supporting marketing
tributes.
programs.
Marketing programs are designed to enhance brand Leveraging secondary associations. The defini-
awarenessand establishfavorable, strong, and unique tion of customer-basedbrand equity does not distin-
brandassociations in memory so that consumerspur- guish between the sources of brandbeliefs (Fishbein
chase the productor service. Brand awareness is re- and Ajzen 1975)-that is, whetherbeliefs are created
lated to brandfamiliarity.Alba and Hutchinson(1987) by the marketeror by some other source of influence
define brand familiarity as the number of product- such as referencegroups or publicity. All that matters
relatedexperiencesthat have been accumulatedby the is the favorability, strength, and uniquenessof brand
consumer (throughproductusage, advertising, etc.). associations which, combined with brandawareness,
Greaterbrandfamiliarity,throughrepeatedexposures can produce differential consumer response to the
to a brand, should lead to increasedconsumerability marketingof a brand. Nevertheless, it is worthwhile
to recognize and recall the brand. Thus, the appro- to consider in greater depth how belief associations
priate marketingstrategyto increase brandawareness about the attributesand benefits of the brandarise.
and familiarityis clear from the definition-anything One way belief associations are created is on the
that causes the consumer to "experience"or be ex- basis of direct experience with the productor service.
posed to the brand has the potential to increase fa- A second way is by informationabout the productor
miliarity and awareness. Frequent and prominent service communicated by the company, other com-
mentions in advertising and promotion vehicles can mercial sources, or word of mouth. Of the two, direct
intrusively increase consumer exposure to the brand, experience may create strongerassociations in mem-
as can event or sportssponsorship,publicity,and other ory given its inherent self-relevance (Hertel 1982).
activities. These episodic memory traces (Tulving 1983) may be
Favorable, strong, and unique associationscan be especially importantfor user and usage image attri-
createdby the marketingprogramin a varietyof well- bute associations. A third importantway that belief
established ways that are only highlighted here. The associations are created is on the basis of inferences
January1993

from some existing brandassociations. That is, many choose a hybrid or sub-brandstrategy whereby they
associations are assumed to exist for the brand be- combine their company name with individual brand
cause it is characterizedby otherassociations.The type names (e.g., Kellogg's Corn Flakes and Courtyardby
and strengthof inferencing are a function of the cor- Marriott).The latter two types of brandingstrategies
relationsperceived by consumersamong attributesor should facilitate access to consumers'overall attitudes
benefits (Ford and Smith 1987; Huber and McCann towardthe company. The sub-brandstrategyoffers an
1982). For example, some consumers in certain cat- additionalpotentialbenefit in that it can allow for the
egories may infer a high level of product or service creation of more specific brandbeliefs.
quality from a high price, as well as infer specific Similarly, a brand may be associated with its
attributesor benefits such as prestige and social sta- "countryof origin" (i.e., the country in which the
tus. Dick, Chakravarti,and Biehal (1990) referto these company makes the product or provides the service)
types of inferences as based on "probabilisticconsis- in such a way thatconsumersinfer specific beliefs and
tency." They note that "evaluative consistency" in- evaluations (Erickson, Johansson, and Chao 1984;
ferences may also occur, as when consumersinfer the Hong and Wyer 1989, 1990). For example, French
favorabilityof a brandattributeor benefit on the basis wines, Germanautomobiles, and Japaneseelectronics
of their overall brand attitude or their evaluation of probablyall benefit from such inferences. Finally, the
some other perceived attributeor benefit. distributionchannels for a product may also create
Another type of inferredassociation occurs when secondaryassociations. Consumerscan form "brand"
the brand association itself is linked to other infor- images of retailers (Jacoby and Mazursky 1984) on
mation in memory that is not directly related to the the basis of theirproductassortment,pricingand credit
productor service. Because the brandbecomes iden- policy, quality of service, and so on. These store im-
tified with this other entity, consumersmay infer that ages have associationsthat may be linked to the prod-
the brand shares associations with that entity, thus ucts they sell (e.g., prestigeandexclusivityvs. bargain-
producingindirector "secondary"links for the brand. drivenand mass appeal). Similartypes of images may
These secondary associations may lead to a transfer be formed for catalogs and other forms of direct mar-
of global associations such as attitude or credibility keting.
(e.g., expertise, trustworthiness,and attractiveness) The final two types of secondaryassociations oc-
or more specific attributesand benefits related to the cur when the primarybrandassociations are for user
product or service meaning. Secondary associations and usage situation attributes,especially when they
may arise from primaryattributeassociations related are for a particular personor event. Considerthe case
to (1) the company, (2) the countryof origin, (3) the in which advertisingcreates an associationbetween a
distributionchannels, (4) a celebrity spokespersonor brand and a celebrity endorser (Rossiter and Percy
endorsorof the productor service, or (5) an event. 1987). As a result, other associationsfor the celebrity
The first three types of secondaryassociations in- may become related to the brand. Ideally, one such
volve "factualsources"for the brand(i.e., who makes association would be a favorable attitudetoward the
it, where it is made, and where it is purchased).This celebrity-for example, a well-known person could
informationis almost always potentially available to lend credibility to product or service claims because
consumers,but its strengthof associationwith the brand of his or her expertise, trustworthiness,or attractive-
depends on the emphasis it receives. First, the brand ness. Additionally, more specific beliefs may be in-
may vary by the extent to which it is identified with volved (Kahle and Homer 1985; McCracken 1989).
a particularcompany. Establishinga connection with Thus, consumers have images of celebrity endorsors
a company may cause existing associations for that in their minds as a result of observing the celebrities
company to become secondary associations for the in their own field of endeavoror as a result of media
brand (e.g., perceptions of company reputationand coverage. A celebrity invariablyhas some personality
credibility). The branding strategy adopted by the attributeassociations,as well as possiblysome product-
company making the productor providingthe service relatedattributeassociations, that may become linked
is the most importantfactor affecting the strengthof to the brand. Similarly, a brandmay also become as-
the company's associationwith the brand.Three main sociatedwith a particularevent. Again, thatevent may
brandingstrategies are possible (Kotler 1991). First, be characterizedby a set of attributeand attitudeas-
companies may choose individual brand names for sociationsin memory. When the brandbecomes linked
different products and services without any explicit with the event, some of these associations with the
mentionof the company(e.g., Procter& Gamblewith event may become indirectlyassociatedwith the brand.
Tide, Bold, Dash, Cheer, Gain, Oxydol, and Duz Finally, as noted previously, identification with the
laundrydetergents). Second, companies may choose product category itself can also result in inferences
their name for all of their productsor services (e.g., producingsecondary associations.
General Electric and Heinz). Third, companies may Secondarybrandassociations may be importantif
Customer-Based
BrandEquity
/ 11

existing brandassociationsare deficient in some way. the characteristicsand relationshipsamong brandas-
In other words, secondary associations can be lever- sociations. Because any one measure typically cap-
aged to create favorable, strong, and unique associ- tures only a particularaspect of brand knowledge,
ations that otherwise may not be present. Choosing to multiple measures must be employed to capture the
emphasize the company or a particularperson, place, multidimensionalnatureof brandknowledge.
or event should be based on consumers' awarenessof Brandawarenesscan be assessedeffectivelythrough
that entity, as well as how the beliefs and attitudes a variety of aided and unaidedmemorymeasures(see
about the entity can become linked to the brand (see Srull 1984 for a review) that can be applied to test
chapter11 of Rossiterand Percy 1987 for an excellent brandrecall and recognition. For example, brandrec-
discussion). Such a strategymakes sense if consumers ognition measures may use the actual brandname or
already have associations for the company, person, some perceptuallydegradedversionof the brandname
place, or event that are congruentwith desired brand (Alba and Hutchinson 1987). Brand recall measures
associations. For example, consider a countrysuch as may use different sets of cues, such as progressively
New Zealand, which is known for having more sheep narrowly defined product category labels. Besides
thanpeople. A New Zealandsweatermanufacturer that correctness,the ease of recall and recognitionperfor-
its
promotes product on the basis of its New Zealand mancecan be assessedwith more subtlemeasuressuch
wool presumablycould more easily establish strong as response latencies to provide a fuller picture of
and favorable brand associations because New Zea- memoryperformancewith respect to the brand(Fazio
land may alreadymean "wool" to many people. Sec- 1987). Brandrecall can also be coded in terms of the
ondarybrandassociationsmay be risky, however, be- orderof recall to capturethe extent to which the name
cause some control of the brand image is given up. is "top of mind" and thus stronglyassociatedwith the
The company, person, place, or event that makes up productcategory in memory.
the primarybrand association will undoubtedlyhave There are many ways to measure the characteris-
a host of associations of which only some smaller set tics of brand associations (i.e., their type, favorabil-
will be of interestto the marketer.Managingthe transfer ity, and strength). Qualitativetechniquescan be em-
process so that only the relevant secondary associa- ployed to suggest possible associations. For example,
tions become linked to the brand may be difficult. free associationtasks can be used wherebyconsumers
Moreover, these images may change over time as con- describe what the brandmeans to them in an unstruc-
sumers learn more about the entity, and new associ- tured format, either individually or in small groups.
ations may or may not be advantageousfor the brand. Specifically, consumers might be probed in terms of
"who, what, when, where, why, and how" types of
Measuring Customer-Based Brand Equity questionsaboutthe brand.Projectivetechniques(Levy
Thereare two basic approachesto measuringcustomer- 1978, 1981, 1985) such as sentence completion, pic-
based brandequity. The "indirect"approachattempts ture interpretation,and brandpersonalitydescriptors
to assess potential sources of customer-basedbrand may also be useful, especially if consumers are un-
equity by measuring brand knowledge (i.e., brand willing or otherwise unable to express their feelings.
awareness and brand image). The "direct"approach These indirect measures, however, may not ade-
attemptsto measurecustomer-basedbrandequity more quately capture the favorability or strength of asso-
directly by assessing the impact of brandknowledge ciations, and more directmeasuresoften are necessary
on consumerresponseto differentelementsof the firm's to provideadditionalinformation.For example, Ajzen
and Fishbein (1980) give a detaileddescriptionof how
marketing program. The indirect and direct ap-
beliefs and evaluations of attributesand benefits can
proaches to measuring customer-basedbrand equity
are complementaryand should be used together. The be scaled and how attitudescan be measuredthrough
indirectapproachis useful in identifyingwhat aspects a structuredformat, providingan illustrativeexample
of brandknowledgecause the differentialresponsethat in a consumer setting. As noted previously, response
creates customer-basedbrand equity; the direct ap- time measuresof attitudeshave been used as a proxy
for attitudestrength.
proach is useful in determiningthe natureof the dif-
ferential response. Though detailed descriptions and Relationships among brand associations can be
measuredby two general approaches:(1) comparing
critiquesof the many specific techniquesbehind these
two approachesare beyond the scope of this article the characteristicsof brandassociations in some way
(see Aaker 1991 for additional discussion), it is and (2) directly asking consumersfor informationrel-
worthwhile to highlight them briefly. evant to the congruence, competitive overlap, or le-
verage for the brand associations. Congruence is the
Indirect approach. The first approachto measur- extent to which brand associations are shared. Con-
ing customer-based brand equity, measuring brand gruence can be assessed by comparingthe patternof
knowledge, requires measuringbrand awareness and associations across consumersto determinewhich as-
January1993

sociations are common or distinctive. Additionally, brandattribution.Past researchof this type has shown
consumers could be asked directly their conditional thatknowledge of the brandaffects consumerpercep-
expectations for attribute,benefit, or attitudeassoci- tions, preferences, and choices for a product (e.g.,
ations (i.e., the likelihood that a product or service Allison and Uhl 1964; Jacoby, Olson, and Haddock
has one association given that it has another). 1971). Blind tests could be used to examine consumer
Competitive overlap of brand associations is the response to other elements of the marketingmix such
extent to which brand associations are linked to the as proposed pricing, promotion, and channels of dis-
product category (i.e., identification) and are or are tributionchanges.
not sharedwith other brands(i.e., uniqueness). Iden- One importantconsideration with the direct ap-
tification can be assessed by examining how con- proachis the experimentalrealismthatcan be achieved
sumersrespondto brandrecall tasks with productcat- when some aspect of the marketingprogramis attrib-
egory or some othertype of cues. Uniquenessof brand uted to a fictitiously namedor unnamedversion of the
associations can be assessed by comparingthe char- productor service. Detailed concept statementscan be
acteristicsof associationsof the focal brand(i.e., their employed in some situations when it may be other-
type, favorability, and strength) with the character- wise difficult for consumersto examine or experience
istics of associationsfor competing brands.Addition- the marketingmix element withoutbeing awareof the
ally, consumerscouldbe askeddirectly(1) how strongly brand. Thus, concept statementsmay be useful in as-
they identify the brandwith the productcategory and sessing customer-basedbrandequity when consumers
(2) what they consider to be the unique and shared make a product choice or evaluate a change in the
aspects of the brand. Multivariatetechniques such as productor servicecomposition,judge a proposedbrand
multidimensionalscaling also can be employed (Aaker
extension, or respond to a proposed price or distri-
and Day 1986). bution change. Assessing customer-basedbrand eq-
Leverage is the extent to which other brand as- uity with marketingcommunicationspresentsa bigger
sociations linked to a brand association become sec-
challenge with the direct approach(e.g., consumerre-
ondary associations for the brand. Leverage can be sponse to a proposed new advertisingcampaign). In
assessed by comparingthe characteristicsfor the par- this case, storyboardsand animaticor photomaticver-
ticularcompany, person, place, event, or productcat- sions of an ad could be used ratherthan a finished ad
egory with those characteristicsfor the focal brandac- to allow for the necessarydisguiseof the brand.Though
cording to their type, favorability, and strength. this approachshould work well with "informational"
Additionally, consumerscould be asked directly what ads, it probablywould be less appropriatefor "trans-
inferences are made about the brand on the basis of
formational"ads emphasizing user, usage, or some
knowledge of the particular person, place, event, other type of imagery, in which productionvalues are
company, or productcategory. a criticalingredientin achievingcommunicationgoals
Direct approach. The second approach to mea- (Rossiter and Percy 1987).
suring customer-basedbrandequity, directly measur- Finally, anotherpotentiallyuseful approachfor di-
ing the effects of brandknowledge on consumer re- rectly assessing customer-basedbrandequity is con-
sponse to marketingfor the brand,requiresexperiments joint or tradeoffanalysis (Greenand Srinivasan1978,
in which one group of consumersrespondsto an ele- 1990; Green and Wind 1975). Conjoint analysis can
ment of the marketingprogramwhen it is attributed be used to explore the main effects of the brandname
to the brandand anothergroupof consumersresponds (i.e., differencesin preferenceor choice for the brand)
to that same element when it is attributedto a ficti- and interactioneffects between the brand name and
tiously named or unnamedversion of the productor other marketingmix elements such as price, product
service. By attributingthe marketingelement to an or service features, and promotionor channel choices
unfamiliaror anonymous product, consumers should (i.e., differences in perceptions for the brand). For
interpretit with respectto theirgeneralknowledgeabout example, Rangaswamy, Burke, and Oliva (1990) use
the productor service, as well as prototypicalproduct conjoint analysis to explore how brandnames interact
or service specifications and price, promotion, and with physical product features to affect the extenda-
distributionstrategies.Comparingthe responsesof the bility of brandnames to new productcategories. Note
two groups thus provides an estimate of the effects thatif conjointanalysisis employed,caremustbe taken
due to the specific knowledge about the brand that that consumers do not evaluate unrealistic product
goes beyond basic productor service knowledge. profiles or scenarios that violate their basic expecta-
The classic example of this approach is the so- tions for the productor brand(Park 1991; Srinivasan
called "blind" test in which consumers evaluate a 1979).
producton the basis of a description,examination,or Table 1 summarizesthe differentmeasuremental-
actualconsumptionexperience, either with or without ternativesfor customer-basedbrandequity.
Customer-BasedBrandEquity/ 13

TABLE 1
Measurement of Brand Knowledge Constructs Related to Customer-Based Brand Equitya
Construct Measure(s) Purpose of Measure(s)
Brand Awareness
Recall Correct identification of brand given Capture "top-of-mind" accessibility of
product category or some other type of brand in memory
probe as cue
Recognition Correct discrimination of brand as having Capture potential retrievability or
been previously seen or heard availability of brand in memory
Brand Image
Characteristics of brand associations
Type Free association tasks, projective Provide insight into nature of brand
techniques, depth interviews associations
Favorability Ratings of evaluations of associations Assess key dimension producing
differential consumer response
Strength Ratings of beliefs of association Assess key dimension producing
differential consumer response
Relationships among brand associations
Uniqueness Compare characteristics of associations Provide insight into the extent to which
with those of competitors (indirect brand associations are not shared with
measure) other brands; assess key dimension
Ask consumers what they consider to be producing differential consumer response
the unique aspects of the brand (direct
measure)
Congruence Compare patterns of associations across Provide insight into the extent to which
consumers (indirect measure) brand associations are shared, affecting
Ask consumers conditional expectations their favorability, strength, or uniqueness
about associations (direct measure)
Leverage Compare characteristics of secondary Provide insight into the extent to which
associations with those for a primary brand associations to a particular person,
brand association (indirect measure) place, event, company, product class, etc.
Ask consumers directly what inferences are linked to other associations,
they would make about the brand based producing secondary associations for the
on the primary brand association (direct brand
measure)
'This table describes the indirect approach of assessing potential sources of customer-based brand equity by measuring brand
knowledge. The direct approach to measuring customer-based brand equity involves measuring the effects of brand knowledge
on consumer response to marketing-for example, by conducting experiments in which one group of consumers respond to an
element of the marketing mix when it is attributed to the brand, and another group of consumers respond to the same marketing
mix element when it is attributed to a fictitiously named or unnamed version of the product or service.
Managing Customer-Based Brand Equity tentially can create value for the brand by improving
consumers' ability to recall or recognize the brand and/
According to the definition of customer-basedbrand
or by creating, maintaining, or changing the favora-
equity, no single numberor measure captures brand
bility, strength, or uniqueness of various types of brand
equity. Rather, brandequity should be thought of as associations. By influencing brand knowledge in one
a multidimensionalconcept that depends on (1) what
or more of these different ways, marketing activity
knowledge structuresare presentin the minds of con- can potentially affect sales.
sumers and (2) what actions a firm can take to capi-
Second, marketers should define the knowledge
talize on the potential offered by these knowledge structures that they would like to create in the minds
structures.Different firms may be more or less able of consumers-that is, by specifying desired levels of
to maximize the potentialvalue of brandaccordingto awareness and favorability, strength, and uniqueness
the type and nature of marketingactivities that they of product- and non-product-related attributes; func-
are able to undertake.Nevertheless, six generalguide- tional, experiential, and symbolic benefits; and over-
lines based on the preceding conceptual framework all attitudes. In particular, marketers should decide on
are presented here to help marketersbetter manage the core needs and wants of consumers to be satisfied
customer-basedbrandequity. by the brand. Marketers should also decide the extent
First, marketersshould adopta broadview of mar- to which it is necessary to leverage secondary asso-
keting decisions. Marketingactivity for a brand po- ciations for the brand-that is, link the brand to the

company, product class, or particularperson, place, marketers should also conduct controlled experi-
or event in such a way that associations with those ments. Consumer knowledge of competitive brands
entities become indirect or "secondary"associations should be similarlytrackedto provide informationon
for the brand. their sources of customer-basedbrandequity. Exper-
Third, marketersshould evaluate the increasingly iments with consumer response to marketingactivity
largenumberof tacticaloptionsavailableto createthese for competitive brands can also provide a useful
knowledge structures,especially in terms of various benchmark-for example, to determine the unique-
marketingcommunicationalternatives.For example, ness of brandassociations.
the recent growth of "nontraditional"media, promo- Finally, marketersshould evaluate potential ex-
tions, and other marketingactivity (e.g., sports and tension candidates for their viability and possible
event sponsorship; in-store advertising; "minibill- feedbackeffects on core brandimage. Given theirpo-
boards"in transitvehicles, on parkingmeters, and in tentialimportanceto long-termbrandvalue, brandex-
other locations; and productplacementin movies and tension decisions are considered in detail in the rest
television shows) is appropriatefrom the perspective of this section from the perspectiveof customer-based
of customer-basedbrandequity. As noted previously, brandequity and other relevantresearch.
the manner in which a brand association is created Brandextensionscapitalizeon the brandimage for
does not matter-only the resulting favorability, the core productor service to efficiently inform con-
strength, and uniqueness. Thus, many of these new sumersand retailersaboutthe new productor service.
alternatives can offer a cost-effective means of af- Brandextensions can facilitate acceptanceof the new
fecting brandknowledge and thus sales, especially to product or service by providing two benefits. First,
the extent that they complementmore traditionalmar- awareness for the extension may be higher because
keting tactics. Regardless of which options are cho- the brand node is already present in memory. Thus,
sen, the entire marketingprogramshould be coordi- consumersshould need only to establish a connection
natedto createcongruentand strongbrandassociations. in memory between the existing brandnode and the
Different marketing tactics with the same strategic new product or service extension. Second, inferred
goals, if effectivelyintegrated,can createmultiplelinks associations for the attributes, benefits, and overall
to core benefits or other key associations, helping to perceived quality may be created. In other words,
produce a consistent and cohesive brandimage. Mar- consumers may form expectations for the extension
keters should judge the consistency and cohesiveness on the basis of what they alreadyknow aboutthe core
of the brandimage with the businessdefinitionin mind brand.These inferences can lower the cost of the in-
(Levitt 1960) and how well the specific attributesand troductorycampaignfor the extension-for example,
benefits that the productor service is intendedto pro- by increasing advertisingefficiency (Smith and Park
vide to consumers satisfy their core needs and wants 1992).
(Kotler 1991; Park, Jaworski, and Maclnnis 1986). Keller and Aaker (1992) review relevantliterature
Fourth,marketersshould take a long-termview of to provide a conceptualmodel of how consumersuse
marketing decisions. The changes in consumer their knowledge to evaluate a brandextension. They
knowledge about the brand from current marketing maintainthatextension evaluationswill dependon the
activity also will have an indirecteffect on the success salience of the core brand associations in the exten-
of future marketing activities. Thus, from the per- sion context, how relevantconsumersperceive this in-
spective of customer-basedbrand equity in making formationto be to theirextensionevaluations,and how
marketingdecisions, it is importantto consider how favorable inferred associations are in the extension
resulting changes in brandawarenessand image may context. In other words, extension evaluations will
help or hurt subsequentmarketingdecisions. For ex- depend on what kind of informationcomes to mind
ample, the use of sales promotionsinvolving tempo- aboutthe core brandin the extension context, whether
rary price decreases may create or strengthena "dis- this informationis seen as suggestive of the type of
count" association with the brand, with implications productor service that the brandextension would be,
for customer loyalty and responses to future price and whetherthis informationis viewed as good or bad
changes or non-price-orientedmarketing communi- in the extension context in comparisonwith compet-
cation efforts. itors.
Fifth, marketersshould employ trackingstudies to The salience or accessibility of the core brandas-
measure consumer knowledge structuresover time to sociations depends on their strength in memory, as
(1) detect any changes in the different dimensions of well as the retrieval cues provided by the extension
brandknowledge and (2) suggest how these changes context. Some associations may be salient when con-
might be relatedto the effectiveness of differentmar- sumers evaluate some extensions but not others. The
keting mix actions. To the extent that a more precise relevance of the salient core brand associations de-
assessment of customer-basedbrandequity is useful, pends, in part, on theirperceivedsimilarityto the pro-
Customer-Based
BrandEquity/15

posed extensionproductor service(Feldmanand Lynch implications, may be especially likely when the ex-
1988). When overall similarityis high, consumersare isting associationsfor the core brandare alreadyfairly
more likely to base theirextensionevaluationson their weak. For example, the successful introductionof the
attitudetowardthe core brand(Boush and Loken 1991; Miller Lite beer in the U.S. may have accentuated
Boush et al. 1987; Herr, Farquhar,and Fazio 1990). perceptionsof the flagship Miller High Life beer as a
Overall similarityjudgmentscould be made in differ- "less hearty"beer because thatperceptionhad already
ent ways (Loken and Ward 1990), though researchers been created in consumers' minds by its clear bottle
typically assume that they are a function of salient (in contrast to Budweiser's dark bottle). As another
shared associations between the core brand and the example of a potential dilution effect, successful ex-
extension product category. These similarity judg- tensions for brands with an exclusivity and prestige
ments could be based on product-relatedattributes,as image that effectively broadenthe target marketmay
well as non-product-related attributessuch as user type produce negative feedback effects on the brandfrom
or usage situation(Bridges 1990; Park, Milberg, and members of the original consumer franchise who re-
Lawson 1991). Whenoverallsimilarityis not very high, sent the market expansion. For example, the intro-
consumers are more likely to consider specific attri- duction of the lower priced Cadillac Cimaronmodel
butes and benefits involved. If relevant, the favora- is thought to have led to declines in image and sales
bility of inferredattributeand benefit beliefs will de- for the entire Cadillac division (Yovovich 1988).
pend on how they are valued in the extension context. Thoughthese differenttypes of dilutioneffects may
Though these evaluationswill generallycorrespondto occur, multiple productor service extensions may not
the favorability of the core brand associations, they be as harmfulto certain abstractassociations such as
can differ, and in fact be negative, even if the core brandattitudesand perceived quality. In other words,
brandassociationsthemselves are positive (Aakerand although the brand may not have the same specific
Keller 1990). Moreover, even if positive attributeand product or service meaning because of multiple ex-
benefit associations for the core brand lead to infer- tensions, consumers may still see the brand as rep-
ences of positive brand extension associations, in- resenting a range of productsor services of a certain
ferrednegative associationsmay still emerge (Bridges quality.
1990). Finally, when overall similarity is very low, An unsuccessfulbrandextension, in contrast, can
consumer evaluations also will be very low. harmthe core brandimage by creatingundesirableas-
When multiple product or service extensions are sociations. Such effects are most likely when there is
associatedwith the brand,the congruenceamong their little differencebetween the originalbrandand the ex-
associations becomes an importantdeterminantof the tension. For example, Sullivan (1990) conducted an
consistency and cohesiveness of the brand image. It econometric analysis that showed how the perceived
is often argued that an extension can help the core "suddenacceleration"problemof Audi's 5000 model
brandimage by improvingthe favorabilityand strength "spilled over" and reduced demand for its 4000 and
of associations and clarifying the business definition Quattromodels. RoedderJohnand Loken(1990) found
and core benefits for the brand. Aaker (1991) claims thatperceptionsof qualityfor a core brandin the health
that brandextensions helped to fortify the brandim- and beauty aids area decreased with the hypothetical
ages of WeightWatchersand Sunkist.KellerandAaker introductionof a lower quality extension in a similar
(1992) foundthatthe successfulintroductionof a brand productcategory (i.e., shampoo). Qualityperceptions
extension improved evaluations of a core brand that of the core brandwere unaffected,however, when the
originallywas perceivedto be of only averagequality, proposed extension was in a dissimilarproductcate-
although in their research setting consumers did not gory (i.e., facial tissue). Similarly, Keller and Aaker
have stronglyheld attitudestowardthe core brandand (1992) found that unsuccessfulinterveningextensions
the company adopted a family brandingstrategy that in dissimilar productcategories did not affect evalu-
raised the salience of its name (and thus perceptions ations of the core brand(also see Romeo 1990).
of its credibility). In summary, marketersshould evaluate potential
It has also been argued that successful brandex- extension candidatesfor their viability and their feed-
tensions may potentially harm the core brand image back effects on core brand image by (1) identifying
if they weaken existing associations in some way. If possibleextensioncandidateson the basis of core brand
a brand becomes associated with a disparate set of associations (especially with respect to brand posi-
products or services, product category identification tioning and core benefits) and overall similarityof the
and the correspondingproduct associations may be- extension to the brand, (2) evaluatingextension can-
come less strong. For example, Pepperidge Farm, didate potentialby measuringthe salience, relevance,
Cadbury, and Scott Paper have been accused of and favorabilityof core brandassociationsin the pro-
"overextending"by introducing too disparate prod- posed extension context and the favorability of any
ucts. Dilution effects, with potentially adverse profit inferred associations, and (3) considering the exten-
January1993

sion's potentialfeedbackeffects on the core brandim- term view of marketinga brand; specifying the de-
age and the favorability, strength, and uniqueness of sired consumer knowledge structuresand core bene-
core brandassociations. fits for a brand; considering a wide range of tradi-
tional and nontraditionaladvertising, promotion, and
other marketingoptions; coordinatingthe marketing
Discussion options that are chosen; conducting tracking studies
and controlled experiments;and evaluating potential
Summary extension candidates.
This article introducesthe concept of customer-based
brandequity, definedas the differentialeffect of brand Future Research Directions
knowledge on consumerresponse to the marketingof In the presentation of a conceptual framework of
the brand. A brandis said to have positive (negative) customer-basedbrand equity, several constructs and
customer-basedbrandequity if consumersreact more relationshipsare discussed. Consequently, additional
(less) favorably to an element of the marketingmix research is necessary both to refine this framework
for the brandthan -heydo to the same marketingmix and to suggest other implicationsfor marketingstrat-
element when it is attributedto a fictitiously named egies and tactics. Undoubtedly, much previous re-
or unnamedversion of the productor service. Brand search may be useful in this effort. Because this re-
knowledge is conceptualizedaccording to an associ- search was most likely conducted with a different
ative network memory model in terms of two com- purposein mind, however, additionalinsights may be
ponents, brandawarenessand brandimage (i.e., a set gained by considering it from the potentiallybroader
of brand associations). Brand awareness consists of perspective of customer-basedbrandequity. In clos-
brand recognition and brand recall. Brand associa- ing, some researchprioritiesfor building, measuring,
tions are conceptualized in terms of their character- and managing customer-basedbrandequity are iden-
istics by type (level of abstractionand qualitativena- tified.
ture), favorability,and strength,and in terms of their Thereare severalimportantresearchquestionsabout
relationship with other associations by congruence, how to build customer-basedbrandequity. First, bet-
competitive overlap (identification and uniqueness), ter choice criteriashould be establishedfor the brand
andleverage.Customer-based brandequityoccurswhen identities (brand name, logo, and symbol). For ex-
the consumer is aware of the brand and holds some ample, remarkablylittle empirical research has sys-
favorable, strong, and unique brand associations in tematically examined brand name considerationsas
memory. The differenttypes of customer-basedbrand they pertainto enhancingbrandawarenessand build-
equity are discussedby consideringthe effects of these ing favorable, strong, and unique brandassociations.
dimensions of brandknowledge on brandloyalty and Such research should recognize the numeroustrade-
consumerresponse to product, price, promotion, and offs in choice criteria by suggesting when certain
distributionstrategies. characteristicsof the brand identities should be em-
The article also explores some specific aspects of phasized. For example, memory retrieval consider-
this conceptualizationby considering how customer- ations that arise from associative strengthand part-list
based brandequity is built, measured, and managed. cueing theories in psychology imply that a meaning-
Buildingbrandequityrequirescreatinga familiarbrand ful, "suggestive"brandname may facilitateinitialpo-
name and a positive brandimage-that is, favorable, sitioning, but a nonsuggestive or neutralbrandname
strong, and unique brand associations. Strategies to may more effectively accommodatelater reposition-
build customer-basedbrand equity are discussed in ing. Supportof this hypothesiswould imply thatfirms
terms of both the initial choice of the brandidentities may be better off adopting more flexible branding
(brandname, logo, and symbol) and how the brand strategies, using more neutral brand names, if they
identities are supported by and integrated into the anticipateneeding to repositionthe brandlater. In de-
marketing program. Two basic approaches to mea- veloping contingency-basedchoice criteria,it also will
suring customer-basedbrandequity are outlined. The be necessaryto clarify the roles of variousbrandiden-
indirect approachmeasures brand knowledge (brand tities by consideringmore explicitlyhow brandnames,
awarenessand elements of brandimage) to assess the logos, symbols, slogans, and other trademarkscan
potentialsources of brandequity. The directapproach contribute differentially to building customer-based
measures the effects of the brandknowledge on con- brandequity. This line of researchcould consider vi-
sumerresponseto elements of the marketingmix. Ex- sual and verbalpropertiesof these brandidentitiesand
amples of both types of approachesare provided. Fi- how they might affect brand awareness and the fa-
nally, six guidelines for the managementof customer- vorability, strength, and uniqueness of brandassoci-
based brand equity are discussed. These guidelines ations.
emphasize the importanceof taking a broadand long- In termsof understandinghow the supportingmar-
Customer-Based
BrandEquity
/ 17

keting program builds customer-basedbrand equity, consistent and cohesive brandimage? This line of re-
two particularresearch directions could be pursued. search should clearly examine how traditionaland
First, factorsinfluencingthe favorability,strength,and nontraditionalmarketing options interact. Effective
uniquenessof brandassociations,a focus of much past strategies for integratingmarketing communications
research, should continue to be explored, but along in terms of advertising, promotion, publicity, direct
several different lines. Are certain types of associa- marketing,and packagedesign are especially needed.
tions inherently more favorable, stronger, or unique Forexample,how can advertisingbe coordinatedacross
in memory?Which types of associationsare more eas- broadcastand printmedia to enhancebrandawareness
ily created by a particularmarketingor communica- and strengthen brand associations? These research
tion mix element?Whichtypes of associationsare more studies might consider memory principles and theo-
likely to influence consumerresponse with respect to ries of encoding and retrieval.
a particularmarketingmix element? Finally, what are Also, how should the consistency and cohesive-
the tradeoffs involved in creating favorable, strong, ness of a brandimage be managedover consumerseg-
and unique brand associations? For example, it was ments(includinggeographicboundaries)andover time?
suggested previously that benefits can be more mem- A diffuse brandimage with weakerand less favorable
orable than attributeinformation, but attributesmay brand associations may be particularlyevident when
have to be communicatedto persuadeconsumersand a brand attempts to reposition itself (e.g., switching
create favorablebenefit associations. It was also sug- to a new targetmarket)(Heckler, Keller, and Houston
gested above that non-product-relatedor image attri- 1992). Are there ways in which a brandimage can be
butes, such as user type or usage situation, may create "flexible" and appeal to different consumer seg-
uniqueassociations,but undersome circumstancesthey ments? To manage the brandimage betterover time,
may not be favorably received or strongly linked to more precise guidelines as to the "indirect"effects of
the brandin memory. current marketing activity on the success of future
Second, the costs and benefits of leveraging sec- marketing activity are needed-for example, by
ondaryassociations should be explored. For example, achieving a betterunderstandingof how brandknowl-
how and underwhat conditions should a firm increase edge influences consumer response.
the salience of source factorsrelatedto the brand(i.e., Finally, broader implications of customer-based
the company, countryof origin, and distributionchan- brand equity should be explored by considering ag-
nel)? All of these source factors have their own set of gregation issues associated with brandknowledge ef-
associations. How do consumers merge or combine fects on marketsegments or the customerfranchiseas
these associations with other brand associations? In a whole, as opposed to effects on an individualcon-
other words, how do these source and brand images sumer. An aggregateanalysis also could consider the
interact?Another importantissue is when and how a implicationof customer-basedbrandequity for sales,
brandshould attemptto become associated with a par- market share, and profits. This more extended anal-
ticularpersonor event. Forexample,RossiterandPercy ysis should consider aspects of the company (e.g., its
(1987) offer the following criteriafor choosing a pre- strengthsand weaknesses) and the competitive nature
senter in advertising:(1) visibility, (2) credibility(ex- of its markets. Similarly, it may also be useful to in-
pertise and objectivity), (3) attraction(likability and corporatesome of the conceptsthatrelateto customer-
similarity), and (4) power. These criteria could be based brand equity to address other management
adaptedto address when and how a brandshould be- questionspertainingto branding-for example, to de-
come identified with an event. velop a financially based conceptualizationof brand
One importantresearchpriorityis to develop valid equity.
benchmarks for the direct approach to measuring
customer-basedbrand equity-that is, plausible de-
scriptions of the relevant activity (advertising, pro- Conclusions
motion, product, pricing, etc.) with no or fictitious The goal of this article is to present a conceptual
brandidentification.Anotheruseful contributionwould framework that would provide useful structure for
be to design efficient and effective approachesto con- managersdeveloping brandstrategiesand researchers
ducting trackingstudies. This would entail consider- studyingbrandequity. In particular,the articlebuilds
ing the pros and cons of different qualitative and a theoreticalfoundationbased on past researchin con-
quantitative approaches to measuring brand knowl- sumer behavior that should be helpful in addressing
edge of consumers. some of the new challenges in developing brandstrat-
Several research questions are relevant for man- egies that have arisen because of changes in the mar-
aging customer-basedbrand equity. What strategies keting environment (e.g., from the proliferationof
areeffective in creatingstrongbrandassociations?How brandextensions and the growth of new, alternative
can differentmarketingmix elements be integratedto promotionaland media alternatives).
create strong and congruentbrandassociations and a Though many of the ideas expressed in this con-
January1993

ceptual framework may be familiar to managers, its marketers to take better short-term and long-term mar-
value is in integrating these various notions to provide keting actions. Moreover, this framework may also
a more comprehensive picture of how marketers can suggest some considerations that have been otherwise
create value for a brand. For example, marketers may overlooked. Thus, this broader context can help man-
agree that they should take a broad and long-term view agers can make more insightful and informed brand
of marketing decisions for a brand, but how they should decisions.
do so may not be obvious. By recognizing that mar- For researchers, the value of the framework is in
keting activity can potentially enhance or maintain suggesting areas where managerial guidance is needed
consumers' awareness of the brand or the favorability, but academic guidelines are currently lacking. As sug-
strength, and uniqueness of various types of brand as- gested by the large number of suggested future re-
sociations, the customer-based brand equity frame- search directions identified, much work needs to be
work may provide the perspective that will enable done.
REFERENCES
Aaker, David A. (1982), "Positioning Your Product," Busi- Paul Marsh (1989), Accounting for Brands. London: Lon-
ness Horizons, 25 (May/June), 56-62. don Business School and the Institute for Chartered Ac-
(1991), Managing Brand Equity. New York: The countants in England and Wales.
Free Press. Berger, Ida E. and Andrew A. Mitchell (1989), "The Effect
and Alexander Biel, eds. (1992), Building Strong of Advertising on Attitude Accessibility," Journal of Con-
Brands. Hillsdale, NJ: Lawrence Erlbaum Associates. sumer Research, 16 (December), 280-8.
and George S. Day (1986), Marketing Research, Bettman, James R. (1979), An Information Processing Theory
3rd ed. New York: John Wiley & Sons, Inc. of Consumer Choice. Reading, MA: Addison Wesley Pub-
and Kevin Lane Keller (1990), "Consumer Eval- lishing Company.
uations of Brand Extensions," Journal of Marketing, 54 (1986), "Consumer Psychology," Annual Review
(January), 27-41. of Psychology, 37, 257-89.
Ajzen, Icek and Martin Fishbein (1980), Understanding Atti- and C. Whan Park (1980), "Effects of Prior
tudes and Predicting Social Behavior. Englewood Cliffs, Knowledge and Experience and Phase of the Choice Pro-
NJ: Prentice-Hall, Inc. cess on Consumer Decision Processes: A Protocol Analy-
Alba, Joseph W. and Amitava Chattopadhyay (1985a), "The sis," Journal of Consumer Research, 7 (December), 234-
Effects of Context and Part-CategoryCues on the Recall of 48.
Competing Brands," Journal of Marketing Research, 22 , Deborah Roedder John, and Carol A. Scott (1986),
(August), 340-9. "Covariation Assessment by Consumers," Journal of Con-
and (1985b), "The Effects of Part-List sumer Research, 13 (December), 316-26.
Cuing on Attribute Recall: Problem Framing at the Point and Mita Sujan (1987), "Effects of Framing on
of Retrieval," in Advances in Consumer Research, Vol. 12, Evaluation of Comparable and Noncomparable Alternatives
ElizabethC. Hirschmanand MorrisB. Holbrook, eds. Provo, by Expert and Novice Consumers," Journal of Consumer
UT: Association for Consumer Research, 410-13. Research, 14 (September), 141-54.
and (1986), "Salience Effects in Brand Blattberg, Robert C. and KennethJ. Wisniewski (1989), "Price-
Recall," Journal of Marketing Research, 23 (November), Induced Patterns of Competition," Marketing Science, 8
363-9. (Fall), 291-309.
and J. Wesley Hutchinson (1987), "Dimensions of Boush, David M. and Barbara Loken (1991), "A Process
Consumer Expertise," Journal of Consumer Research, 13 Tracing Study of Brand Extension Evaluations," Journal of
(March), 411-53. Marketing Research, 28 (February), 16-28.
and John G. Lynch, Jr. (1991), , Shannon Shipp, Barbara Loken, Ezra Gencturk,
"Memory and Decision Making," in Handbook of Con- Susan Crockett, Ellen Kennedy, Betty Minshall, Dennis
sumer Theory and Research, Harold H. Kassarjian and Misurell, Linda Rochford, and Jon Strobel (1987), "Affect
Thomas S. Robertson, eds. Englewood Cliffs, NJ: Pren- Generalization to Similar and Dissimilar Brand Exten-
tice-Hall, Inc., 1-49. sions," Psychology and Marketing, 4 (3), 225-37.
Allison, Ralph and Kenneth P. Uhl (1964), "Influencesof Beer Bridges, Sheri (1990), "A Schema Unification Model of Brand
Brand Identification on Taste Perception," Journal of Mar- Extensions," unpublished doctoral dissertation, Graduate
keting Research, 2 (August), 36-9. School of Business, Stanford University.
Anderson, John R. (1983), The Architecture of Cognition. Burke, Raymond R. and Thomas K. Srull (1988), "Compet-
Cambridge, MA: Harvard University Press. itive Interference and Consumer Memory for Advertising,"
Baker, William, J. Wesley Hutchinson, Danny Moore, and Journal of Consumer Research, 15 (June), 55-68.
Prakash Nedungadi (1986), "Brand Familiarity and Adver- Chaiken, Shelley (1986), "The Heuristic Model of Persua-
tising: Effects on the Evoked Set and Brand Preferences," sion," in Social Influence: The Ontario Symposium, Vol.
in Advances in Consumer Research, Vol. 13, Richard J. 4, Mark P. Zanna, E. Tory Higgins, and C. P. Herman,
Lutz, ed. Provo, UT: Association for Consumer Research, eds. Hillsdale, NJ. Lawrence Erlbaum Associates.
637-42.
Chattopadhyay, Amitava and Joseph W. Alba (1988), "The
Barwise, Patrick, ChristopherHigson, Andrew Likierman, and Situational Importance of Recall and Inference in Con-
Customer-Based
BrandEquity/ 19

sumer Decision Making," Journal of Consumer Research, Green, Paul E. and V. Srinivasan (1978), "Conjoint Analysis
15 (June), 1-12. in Consumer Research: Issues and Outlook," Journal of
Collins, Allan M. and Elizabeth F. Loftus (1975), "A Spread- Consumer Research, 5 (September), 102-23.
ing Activation Theory of Semantic Processing," Psycho- and (1990), "Conjoint Analysis in Con-
logical Review, 82, 407-28. sumer Research: New Developments and Directions,"
Craik, Fergus I. M. and Robert S. Lockhart (1972), "Levels Journal of Marketing, 54 (October), 3-19.
of Processing: A Framework for Memory Research," Jour- and Yoram Wind (1975), "New Ways to Measure
nal of Verbal Learning and Verbal Behavior, 11, 671-84. Consumers' Judgments," Harvard Business Review, 53
and Endel Tulving (1975), "Depth of Processing (July/August), 107-17.
and the Retention of Words in Episodic Memory," Journal Gregg, Vernon H. (1976), "WordFrequency, Recognition, and
of Experimental Psychology, 104 (3), 268-94. Recall," in Recall and Recognition, John Brown, ed. Lon-
Cohen, Joel B. and Kanul Basu (1987), "Alternative Models don: John Wiley & Sons, Inc.
of Categorization: Towards a Contingent Processing Heckler, Susan, Kevin Lane Keller, and Michael J. Houston
Framework," Journal of Consumer Research, 13 (March), (1992), "The Effects of Brand Repositioning on Ad Re-
455-72. call," working paper, Karl Eller School of Management,
Day, George S., Allan D. Shocker, and RajendraK. Srivastava University of Arizona.
(1979), "Customer-Oriented Approaches to Identifying Herr, Paul M., Peter H. Farquhar,and Russell H. Fazio (1990),
Product-Markets," Journal of Marketing, 43 (Fall), 8-19. "Extending Brand Equity to New Categories," working pa-
Dick, Alan, Dipankar Chakravarti,and Gabriel Biehal (1990), per, Graduate School of Business, Indiana University.
"Memory-Based Inferences During Consumer Choice," Hertel, Paula T. (1982), "Remembering Reactions and Facts:
Journal of Consumer Research, 17 (June), 82-93. The Influence of Subsequent Information," Journal of Ex-
Dobni, Dawn and George M. Zinkhan (1990), "In Search of perimental Psychology: Learning, Memory, and Cognition,
Brand Image: A Foundation Analysis," in Advances in 8 (6), 513-29.
Consumer Research, Vol. 17, Marvin E. Goldberg, Gerald Herzog, H. (1963), "Behavioral Science Concepts for Ana-
Gorn, and Richard W. Pollay, eds. Provo UT: Association lyzing the Consumer," in Marketing and the Behavioral
for Consumer Research, 110-19. Sciences, Perry Bliss, ed. Boston: Allyn and Bacon Inc.,
Dolan, Carrie (1985), "Concocting Zingy New Names Starts 76-86.
Turning Into a Business," Wall Street Journal. Hoch, Stephen J. (1984), "Availability and Interference in
Erickson, Gary M., Johny K. Johansson, and Paul Chao (1984), Predictive Judgment," Journal of Experimental Psychol-
"Image Variables in Multi-Attribute Product Evaluations: ogy: Learning, Memory, and Cognition, 10 (October), 629-
Country-of-Origin Effects," Journal of Consumer Re- 62.
search, 11 (September), 694-9. Hong, Sung-Tai and Robert S. Wyer (1989), "Effects of
Farquhar, Peter H. (1989), "Managing Brand Equity," Mar- Country-of-Origin and Product Attribute Information on
keting Research, 1 (September), 24-33. Product Evaluation: An Information Processing Perspec-
Fazio, Russell H. (1987), "Category-Brand Associations and tive," Journal of ConsumerResearch, 16 (September), 175-
Their Activation From Memory," unpublishedreport, Ogilvy 87.
Center for Research and Development, San Francisco, CA. and (1990), "Determinants of Product
, Martha C. Powell, and Carol J. Williams (1989), Evaluation: Effects of the Time Interval Between Knowl-
"The Role of Attitude Accessibility in the Attitude and Be- edge of a Product'sCountryof Origin and InformationAbout
havior Process," Journal of Consumer Research, 16 (De- Its Specific Attributes," Journal of Consumer Research, 17
cember), 280-8. (December), 277-88.
, David M. Sanbonmatsu, Martha C. Powell, and Houston, Michael J., Terry L. Childers, and Susan E. Heckler
Frank R. Kardes (1986), "On the Automatic Activation of (1987), "Picture-Word Consistency and the Elaborative
Attitudes," Journal of Personality and Social Psychology, Processing of Advertisements," Journal of Marketing Re-
50 (February), 229-38. search, 24 (November), 359-69.
and Mark P. Zanna (1981), "Direct Experiences Hoyer, Wayne D. and Steven P. Brown (1990), "Effects of
and Attitude-Behavior Consistency," in Advances in Ex- BrandAwareness on Choice for a Common, Repeat-Purchase
perimental Social Psychology, Vol. 14, Leonard Berkowitz, Product," Journal of Consumer Research, 17 (September),
ed. New York: Academic Press, Inc., 161-202. 141-8.
Feldman, Jack M. and John G. Lynch, Jr. (1988), "Self- Huber, Joel and John McCann (1982), "The Impact of Infer-
Generated Validity and Other Effects of Measurement on ential Beliefs on Product Evaluations," Journal of Market-
Belief, Attitude, Intention, and Behavior," Journal of Ap- ing Research, 18 (November), 428-41.
plied Psychology, 73 (August), 421-35. Isen, Alice M. (1992), "The Influence of Positive Affect on
Fennell, Geraldine (1978), "Consumer's Perceptions of the Cognitive Organization: Some Implications for the Influ-
Product-Use Situation," Journal of Marketing, 42 (April), ence of Advertising on Decisions About Products and
38-47. Brands," in Advertising Exposure, Memory, and Choice,
Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, In- Andrew A. Mitchell, ed. Hillsdale, NJ: Lawrence Erlbaum
tention, and Behavior: An Introduction to Theory and Re- Associates, in press.
search. Reading, MA: Addison-Wesley Publishing Com- Jacoby, Jacob and David Mazursky (1984), "Linking Brand
pany. and Retailer Images-Do the Potential Risks Outweigh the
Ford, Gary T. and Ruth Ann Smith (1987), "Inferential Be- Potential Benefits?" Journal of Retailing, 60 (2), 105-22.
liefs in Consumer Evaluations: An Assessment of Alter- , Jerry C. Olson, and Rafael A. Haddock (1971),
native Processing Strategies," Journal of Consumer Re- "Price, Brand Name, and Product Composition Character-
search, 14 (December), 363-71. istics as Determinants of Perceived Quality," Journal of
Gardner, Burleigh B. and Sidney J. Levy (1955), "The Prod- Applied Psychology, 55 (December), 570-9.
uct and the Brand," Harvard Business Review, 33 (March- , George J. Syzabillo, and Jacqeline Busato-Schach
April), 33-9. (1977), "InformationAcquisition Behavior in Brand Choice

Situations," Journal of Consumer Research, 3 (4), 209-16. That Influence Consumers' Evaluations of Brand Exten-
Johnson, Michael D. (1984), "Consumer Choice Strategies for sions," working paper, Karl Eller School of Management,
Comparing Noncomparable Alternatives," Journal of Con- University of Arizona.
sumer Research, 11 (December), 741-53. MacKenzie, Scott B. (1986), "The Role of Attention in Me-
Kahle, Lynne R. and Pamela M. Homer (1985), "Physical At- diating the Effect of Advertising on Attribute Importance,"
tractiveness of the Celebrity Endorsor: A Social Adaptation Journal of Consumer Research, 13 (September), 174-95.
Perspective," Journal of Consumer Research, 11 (4), 954- Maltz, Eliot (1991), "Managing Brand Equity: A Conference
61. Summary," Report #91-110. Cambridge, MA: Marketing
Katz, Daniel (1960), "The Functional Approach to the Study Science Institute.
of Attitudes,"Public Opinion Quarterly, 24 (Summer), 163- Maslow, Abraham H. (1970), Motivation and Personality, 2nd
204. ed. New York: Harper & Row Publishers, Inc.
Keller, Kevin Lane (1987), "Memory Factors in Advertising: McCracken, Grant (1989), "Who Is the Celebrity Endorsor?
The Effect of Advertising Retrieval Cues on Brand Eval- CulturalFoundations of the EndorsementProcess," Journal
uations," Journal of Consumer Research, 14 (December), of Consumer Research, 16 (December), 310-21.
316-33. Miller, K. E. and J. L. Ginter (1979), "An Investigation of
(1991a), "Cue Compatibility and Framing in Ad- Situational Variation in Brand Choice Behavior and Atti-
vertising," Journal of Marketing Research, 28 (February), tude, " Journal of Marketing Research, 16 (February), 111-
42-57. 23.
(1991b), "Memory and Evaluation Effects in Com- Myers, James H. and Allan D. Shocker (1981), "The Nature
petitive Advertising Environments," Journal of Consumer of Product-Related Attributes," in Research in Marketing,
Research, 16 (March), 436-76. Vol. 5, Jagdish Sheth, ed. Greenwich, CT: JAI Press, Inc.,
(1992), "Memory Retrieval Factors and Advertis- 211-36.
ing Effectiveness," in Advertising Exposure, Memory, and Myers-Levy, Joan (1989), "The Influence of a Brand Name's
Choice, Andrew A. Mitchell, ed. Hillsdale, NJ: Lawrence Association Set Size and Word Frequency on Brand Mem-
Erlbaum Associates, in press. ory," Journal of ConsumerResearch, 14 (September), 197-
and David A. Aaker (1992), "The Effects of Se- 207.
quential Introduction of Brand Extensions," Journal of and Alice M. Tybout (1989), "Schema Congruity
Marketing Research, 29 (February), 35-50. as a Basis for Product Evaluation," Journal of Consumer
Kotler, Philip H. (1991), Marketing Management: Analysis, Research, 16 (June), 39-54.
Planning, and Control, 8th ed. Englewood Cliffs, NJ: Nedungadi, Prakash (1990), "Recall and Consumer Consid-
Prentice-Hall, Inc. eration Sets: Influencing Choice Without Altering Brand
Leuthesser, Lance, ed. (1988), "Defining, Measuring and Evaluations," Journal of Consumer Research, 17 (Decem-
Managing Brand Equity: A Conference Summary," Report ber), 263-76.
#88-104. Cambridge, MA: Marketing Science Institute. and Wesley Hutchinson (1985), "The Prototypi-
Levitt, Theodore (1960), "MarketingMyopia," Harvard Busi- cality of Brands: Relationships With Brand Awareness,
ness Review, 38 (July/August), 45-56. Preference and Usage," in Advances in Consumer Re-
Levy, Sidney J. (1978), Marketing Behavior: Its Meaning for search, Vol. 12, Elizabeth C. Hirschman and Morris B.
Management. New York: AMACOM. Holbrook, eds. Provo, UT: Association for Consumer Re-
(1981), "Interpreting Consumer Mythology: A search, 489-503.
Structural Approach to Consumer Behavior," Journal of Newman, Joseph W. (1957), "New Insight, New Progress for
Marketing, 45 (Summer), 49-61. Marketing," Harvard Business Review, 35 (November-
(1985), "Dreams, Fairy Tales, Animals, and Cars," December), 95-102.
Psychology and Marketing, 2 (2), 67-81. Olson, Jerry C. and Jacob Jacoby (1972), "Cue Utilization in
Lockhart, Robert S., Fergus I. M. Craik, and Larry Jacoby the Quality Perception Process," in The Proceedings of the
(1976), "Depth of Processing, Recognition, and Recall," Third Annual Conference of the Association for Consumer
in Recall and Recognition, John Brown, ed., New York: Research, M. Venkatesan, ed. Iowa City, IA: Association
John Wiley & Sons, Inc. for Consumer Research, 167-79.
Loftus, Elizabeth F. and Gregory R. Loftus (1980), "On the Park, C. Whan, BernardJ. Jaworski, and Deborah J. Maclnnis
Permanence of Stored Information in the Human Brain," (1986), "Strategic Brand Concept-Image Management,"
American Psychologist, 35 (May), 409-20. Journal of Marketing, 50 (October), 621-35.
Loken, Barbara and James Ward (1990), "Alternative Ap- and V. Parker Lessig (1981), "Familiarity and Its
proaches to Understanding the Determinants of Typical- Impact on Consumer Biases and Heuristics," Journal of
ity," Journal of Consumer Research, 17 (September), 111- Consumer Research, 8 (September), 223-30.
26. , Sandra Milberg, and Robert Lawson (1991),
Lutz, Richard J. (1991), "The Role of Attitude Theory in Mar- "Evaluationof Brand Extensions: The Role of ProductLevel
keting," in Perspectives in Consumer Behavior, 4th ed., Similarity and Brand Concept Consistency," Journal of
Harold H. Kassarjianand Thomas S. Robertson, eds. Glen- Consumer Research, 18 (September), 185-93.
view, IL: Scott, Foresman and Company, 317-39. and Daniel C. Smith (1989), "Product-LevelChoice:
Lynch, John G., Jr., HowardMamorstein,and Michael Weigold A Top-Down or Bottom-Up Process?" Journal of Con-
(1988), "Choices From Sets Including Remembered Brands: sumer Research, 16 (December), 289-99.
Use of Recalled Attributes and Prior Overall Evaluations," Park, Chan-Su (1991), "Estimation and Prediction of Brand
Journal of Consumer Research, 15 (September), 169-84. Equities Through Survey Measurement of Consumer Pref-
and Thomas K. Srull (1982), "Memory and Atten- erence Structures," unpublished doctoral dissertation pro-
tional Factors in Consumer Choice: Concepts and Research posal, Graduate School of Business, Stanford University.
Methods," Journal of Consumer Research, 9 (June), 18- Petty, Richard E. and John T. Cacioppo (1986), Communi-
36. cation and Persuasion. New York: Springer-Verlag.
Maclnnis, Deborah J. and Kent Nakamoto (1991), "Factors Plummer, Joseph T. (1985), "How Personality Makes a Dif-
Customer-Based
BrandEquity/ 21

ference," Journal of Advertising Research, 24 (6), 27-31. Stimuli: A Symbolic Interactionism Perspective," Journal
Raaijmakers,J. G. W. and RichardM. Shiffrin (1981), "Search of Consumer Research, 10 (December), 319-29.
of Associative Memory," Psychological Review, 88, 93- Srinivasan, V. (1979), "NetworkModels for EstimatingBrand-
134. Specific Effects in Multi-Attribute Marketing Models,"
Rangaswamy, Arvind, Raymond Burke, and Terence A. Oliva Management Science, 25 (January), 11-21.
(1990), "Brand Equity and the Extendibility of Brand Srivastava, Rajendra and Allan D. Shocker (1991), "Brand
Names," Working Paper No. 90-019, The Wharton School, Equity: A Perspective on Its Meaning and Measurement,"
The University of Pennsylvania. working paper, GraduateSchool of Business, University of
Texas at Austin.
Ratcliff, Roger and Gail McKoon (1988), "A Retrieval Theory
of Priming in Memory," Psychological Review, 95 (3), 385- Srull, Thomas K. (1984), "Methodological Techniques for the
408. Study of Person Memory and Social Cognition," in Hand-
book of Social Cognition, Vol. 2, Robert S. Wyer and
Ray, Michael L. (1982), Advertising and Communication Thomas K. Srull, eds. Hillsdale, NJ: Lawrence Erlbaum
Management. Englewood Cliffs, NJ: Prentice-Hall, Inc. Associates, 1-72.
Ries, Al and Jack Trout (1979), Positioning: The Battle for Starr, Martin K. and Joel R. Rubinson (1978), "A Loyalty
Your Mind. New York: McGraw Hill, Inc.
Group Segmentation Model for Brand Purchasing Simula-
Robertson, Kim R. (1987), "Recall and Recognition Effects tion," Journal of Marketing Research, 15 (August), 378-
of Brand Name Imagery," Psychology and Marketing, 4 83.
(1), 3-15. Sujan, Mita (1985), "Consumer Knowledge: Effects on Eval-
(1989), "Strategically Desirable Brand Name Char- uation Strategies Mediating Consumer Judgments," Jour-
acteristics," Journal of Consumer Marketing, 6 (Fall), 61- nal of Consumer Research, 12 (June), 31-46.
71. and James R. Bettman (1989), "TheEffects of Brand
Roedder John, Deborah and BarbaraLoken (1990), "Diluting Positioning Strategies on Consumers' Brand and Category
Brand Equity: The Impact of Brand Extensions," working Perceptions: Some Insights From Schema Research," Jour-
paper, Carlson School of Management, University of Min- nal of Marketing Research, 26 (November), 454-67.
nesota. Sullivan, Mary W. (1990), "Measuring Image Spillovers in
Romeo, Jean B. (1990), "The Effect of Negative Information UmbrellaBrandedProducts,"Journal of Business, 63 (July),
on the Evaluation of Brand Extensions and the Family 309-29.
Brand," in Advances in Consumer Research, Vol. 18, Tauber, Edward M. (1988), "Brand Leverage: Strategy for
Rebecca H. Holman and Michael R. Solomon, eds. Provo, Growth in a Cost Controlled World," Journal of Advertis-
UT: Association for Consumer Research, 399-406. ing Research, 28 (August/September), 26-30.
Rosch, Eleanor and Carolyn B. Mervis (1975), "Family Re- Tulving, Endel (1983), Elements of Episodic Memory. Lon-
don: Oxford University Press.
semblances: Studies in the Internal Structure of Cate-
and Joseph Psotka (1971), "Retroactive Inhibition
gories," Cognitive Psychology, 7 (October), 573-605. in Free Recall: Inaccessibility of Information Available in
Roselius, Ted (1971), "Consumer Ranking of Risk Reduction
Memory Store," Journal of Experimental Psychology, 87
Methods," Journal of Marketing, 35 (January), 56-61. (1), 1-8.
Rossiter, John R. and Larry Percy (1987), Advertising and Ward, James and BarbaraLoken (1986), "The Quintessential
Promotion Management. New York: McGraw-Hill Book Snack Food: Measurement of Product Prototypes," in Ad-
Company. vances in Consumer Research, Vol. 13, Richard J. Lutz,
Russo, Edward J. and Eric J. Johnson (1980), "What Do Con- ed. Provo, UT: Association for Consumer Research, 126-
sumers Know About Familiar Products?" in Advances in 31.
Consumer Research, Vol. 7, Jerry C. Olson, ed. Ann Ar- Wentz, Laurel (1989), "WPP Considers Brand Valuation,"
bor, MI: Association for Consumer Research, 417-23. Advertising Age, 24.
Simon, Carol J. and Mary W. Sullivan (1990), "The Mea- Wilkie, William (1986), Consumer Behavior. New York: John
surement and Determinants of Brand Equity: A Financial Wiley & Sons, Inc.
Approach," working paper, Graduate School of Business, Wind, Yoram (1982), Product Policy: Concepts, Methods, and
University of Chicago. Strategy. Reading, MA: Addison-Wesley Publishing Com-
Simonson, Itamar, Joel Huber, and John Payne (1988), "The pany.
Relationship Between Prior Knowledge and Information Wyer, Robert S., Jr. and Thomas K. Srull (1989), "Person
Acquisition Order," Journal of Consumer Research, 14 Memory and Judgment,"Psychological Review, 96 (1), 58-
83.
(March), 566-78.
Yovovich, B. G. (1988), "WhatIs Your Brand Really Worth?"
Smith, Daniel C. and C. Whan Park (1992), "The Effects of Adweek's Marketing Week (August 8), 18-21.
Brand Extensions on Market Share and Advertising Effi-
Zeithaml, Valarie (1988), "Consumer Perceptions of Price,
ciency," Journal of MarketingResearch, 29 (August), 296- Quality, and Value: A Means-End Model and Synthesis of
313. the Evidence," Journal of Marketing, 52 (July), 2-22.
Solomon, Michael R. (1983), "The Role of Products as Social Reprint No. JM571100


Keller 1993

Transféré par

Informations du document

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

Keller 1993

Transféré par

Droits d'auteur :

Formats disponibles

Conceptualizing, Measuring, and Managing Customer-Based Brand Equity

Author(s): Kevin Lane Keller

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

This content downloaded from 128.83.63.20 on Mon, 9 Sep 2013 02:26:09 AM

Vous aimerez peut-être aussi