Vous êtes sur la page 1sur 11

PERGAMON International Journal of Project Management 18 (2000) 93±103


Strategic project management and strategic behaviour

Tony Grundy
Cran®eld School of Management, Cran®eld, Bedford MK44 0AL, UK


Strategic projects are crucial to the implementation of strategies. Besides the analytical diculties of managing strategic
projects these are perhaps overshadowed by behavioural diculties. Research into the strategic behaviour at BT has identi®ed
several techniques for managing the behavioural issues facing strategic projects more e€ectively. These techniques include: cause
of behaviour analysis, personal and strategic agenda analysis, behavioural scenarios and diculty, energy and frustration over
time curves. # 2000 Elsevier Science Ltd and IPMA. All rights reserved.

1. Introduction agers will easily recognise that the conduct of strategy

itself is a battleground, given the considerable turbu-
In ``Strategy Implementation and Project lence which surrounds both external and internal stra-
Management'' [1] I argued that there were close a- tegic moves. Strategic projects, however well
nities between strategy implementation and project intentioned, become easily bu€eted by strategies which
management. In that article, I imported a number of are highly emergent and unpredictable. The more di-
tools from strategic management, value management cult of the in¯uences are frequently behavioural in
and from organizational change to enrich project man- nature. There appears to be great merit in incorporat-
agement techniques. ing techniques for surfacing behavioural issues in pro-
These included: ject management, especially for the more strategic
. The ®ve forms of strategyÐand the strategy mix But ®rst we need to de®ne `strategic behaviour'.
. The notion of a set of strategic projects (the `stra- This is de®ned as:
tegic project set')
. `Fishbone' or root cause analysis ``The cognitive, emotional and territorial interplay
. How±How analysis of managers within (or between) groups when the
. From±To analysis agenda relates to strategic issues.'' [2]
. Force ®eld analysis
. Stakeholder analysis
. Attractiveness±implementation diculty Our de®nition of strategic behaviour stresses the
. Assumption analysis/uncertainty extent to which cognitive, emotional and territorial
. Importance±urgency analysis perspectives and agendas of managers are interwoven.
We are thus more able to understand those aspects of
These techniques deal primarily with the more ana- strategy implementation which are perhaps less easily
lytical aspects of strategy implementation. Recent discussible by managers. This diculty could be due
research (in collaboration with Robin Wensley at either to the fact that they involve power (whether
Warwick Business School) has led me to the con- manifested through o€ensive or defensive behaviour,
clusion that equally important are the behavioural or through alliances) or because of emotional sensi-
aspects of strategy implementation. Practising man- tivities.

0263-7863/00/$20.00 # 2000 Elsevier Science Ltd and IPMA. All rights reserved.
PII: S 0 2 6 3 - 7 8 6 3 ( 9 8 ) 0 0 0 7 6 - 3
94 T. Grundy / International Journal of Project Management 18 (2000) 93±103

Two examples of strategic behaviour include: viour more e€ectively than before. The study produced
some most interesting and positive behavioural shifts
. A ®nancial services company is trying to reduce its suggesting that there is at least some potential else-
business complexity from ®fteen to ®ve strategic where for increasing the degree of harmony of strategic
business units. One of the directors raises semi-spur- behaviour. This shift may or may not directly result in
ious reasons for retaining one business unit even measurably improved team performance (as other vari-
though the balance of evidence is clearly against ables are likely to intrude). However, at least this
remaining in it. The debate gets messier and messier, study gives us a ®rst stage in exploring how strategic
with less and less agreement and more and more behaviour might ultimately impact on individual and
frustration. The director coils his arms and legs and business performance.
slumps into defensive non-verbal behaviour. He has Indeed, the BT team found that a number of tools
not, and will not reveal his underlying personal (contained in this paper) were extremely useful for
agenda which is to save the embarrassment of going helping them cope with their own behavioural issues,
back to the managers who have put their very trust as they set about crystallising these strategic projects.
in him, to defend their case, and their jobs. This paper is split into three main parts:
. A retail company is undertaking scenario develop-
ment. One manager invited because of his technical . A quick overview of past literature on strategic
knowledge becomes more and more grumpy. He is behaviour (drawing out implications for strategic
moody because the subject matter is too broad to projects)
help him on his own very speci®c functional issues. . Techniques for understanding strategic behaviour
The rest of the team confront him with the problem, . Illustrations from the BT research
and one says: ``If you really don't see this as adding . Lessons and conclusion.
value to you personally, feel free to leave''. He did,
and the team then began to make progress again.
2. A quick overview of past literature on strategic
The two examples above illustrate the profound behaviour
e€ect of behaviour on the evaluation of strategy. One
might indeed characterise the ®rst example of one of Prior to this research there have been relatively few
``Managers Behaving Badly'', except that would be too ®eld studies explicitly focusing on strategic behaviour.
normative. Certainly the word `dysfunctional' seems For example, Johnson [3] says:
applicable to the ®rst example and partially also to the
second example. ``It is perhaps surprising that . . .there are so few sys-
Whilst strategy formulation is clearly of some temic (systematic) studies of the way in which the in-
emotional and territorial sensitivity, this is likely to be teraction of individuals contributes to strategic
greatly magni®ed during strategy implementation. For decision-making.''
during strategic action, strategy creates turbulence in
the everyday fabric of the organisation. Frequently the
main vehicle for that strategic action is the Strategic As mentioned in my previous paper, [1] it is widely
Project, and it is precisely here that much behavioural accepted that managers tend to make strategic de-
turbulence (or `BT') is felt. cisions in an incremental fashion, often making the
Coincidentally, our research site was British end result look somewhat disjointed, with limited
Telecom (BT). The study focused on the strategic logic. [4] Strategic projects su€er from the same di-
behaviour of a senior team within BT whose remit was culty, making it arduous to manage interdependencies
to understand the implications (market and technologi- or to establish coherent programmes to steer im-
cal) of major changes in BT's external and internal en- plementation over time.
vironment. This key department was charged with Management decision-making has been characterised
de®ning strategic projects which would then form a as having `bounded rationality' [4] or focuses on `mud-
central plank of BT's technological migration. As these dling through'. [5] In its extreme form, `decisions' can-
projects had a multi-business impact they were fre- not easily be detected at all. [6] Even when they are
quently fraught with complexityÐnot merely at a ter- made, they are often subject to a half-hearted or `weak
ritorial level but also organisationally. implementation'. [7]
The study examined the team's patterns of strategic Although organisational learning might hopefully
behaviour whilst discussing these strategic projects come to the rescue, [8] sadly this is very likely to
over a period of several months. This was part of an become bogged down in defensive routines, [9] es-
action research process, the main point of which was pecially where strategic projects threaten existing terri-
to see if managers could harness their strategic beha- torial barriers and existing organisational mind-sets.
T. Grundy / International Journal of Project Management 18 (2000) 93±103 95

Fig. 1. The surface strategic project iceberg.

The consequence of this is that a `rational' approach 3. Techniques for understanding strategic behaviour
to strategy implementation, along the lines of my pre-
vious paper [1] is abandoned in favour of muddling Our techniques for understanding the strategic beha-
through. [10] Unfortunately, the muddling through can viour associated with strategic projects break into
result in profound strategic errors and three levels of behavioural diagnosis, agenda analysis
misjudgements [10] due to `bias' or `gross omissions' and dynamic analysis (Fig. 3). In each case we critique
(for instance, in considering alternatives). the potential problems and diculties which might
In e€ect, conventional project management thus arise when implementing themÐtogether with their
focuses on a very small part of the total `strategic pro- possible solution.These techniques include:
ject iceberg' (see Fig. 1), particularly its:
3.0.1. Behavioural diagnosis
. rational tasks
. resources . Cause of behaviour (`COBRA') analysis
. timescales. . Piranha analysisÐfor behavioural problems
. Importance±in¯uence analysis.
Although strategic project management tries to
address the ®t of the project with both external and in-
3.0.2. Agenda analysis
ternal strategic breakthroughs, and strategic vision (the
apex of the iceberg) it still does not address the deeper . Personal and strategic agenda (`PASTA' factor
more behavioural issues (see Fig. 2). analysis)Ðand sub-personalities.

Fig. 2. The underlying strategic project iceberg.

96 T. Grundy / International Journal of Project Management 18 (2000) 93±103

Fig. 3. Analysing strategic behaviour.

3.0.3. Dynamic analysis . Crucially important root causes are missed either
because these causes are taken-for-granted, hidden
. Behavioural scenarios and `wishbone' analysis or mis-de®ned.
. Diculty over time, energy over time and frustra- . Having done the analysis no-one is prepared to use
tion over time curves. it because of its sensitivity, and due to the lack of
knowledge and interpersonal skill in raising sensitive
issues of this kind.

3.1. Cause of behaviour (`COBRA') analysis Facilitation or self-challenge of the process is here
very helpful: for example, one can always ask the ques-
Cause of behaviour (`COBRA') analysis applies root tion: ``Are these root causes both a necessary and su-
cause or ®shbone analysis [1] to behavioural issues. cient condition of creating this problem?''
Although initially this may not seem like a very major
innovation, in practice this can be an extremely power- 3.2. Piranha analysisÐfor behavioural problems
ful technique for getting hold of the less tangible fac-
tors which may be at work in frustrating a strategic A particular strategic project might be rife with
project (see Fig. 4). behavioural problems. Here it is possible to apply a
COBRA analysis (so named because of the slipperi- re®nement of the Fishbone technique by showing a
ness of behavioural drivers) can help to understand: number of small problems on the same sheet (Fig. 6).
This is called `Piranha' analysis because each problem
. why a project team as a whole is behaving in an on its own, whilst being non-life-threatening to the
apparently dysfunctional way, or project nevertheless eats in a major way into its likely
. the dysfunctional behaviour of a particular individ- success. (A shoal of piranhaÐactually a very small
ual either within or outside the project team. ®shÐis apparently sucient to strip a human being
down to a skeleton in next to no time.)
Once the COBRA analysis has been done, it is then
A potential concern here might be that we have mis-
possible to prioritise which area to intervene with, for
de®ned the various sub-problems that revolve around
example by prioritising those behavioural drivers
the main symptom. To genuinely get their arms
which are:
around and underneath the total problem, managers
. attractive to deal with need to be very careful in ensuring that there are no
. not too dicult to deal with. major dimensions of the problem missing. For
example, one major insurance company faced a huge
perhaps using a more formal attractiveness±implemen- resource constraint in dealing with its Year 2000 IT
tation diculty analysis [1] (Fig. 5). problem. The piranha analysis highlighted not just
Potential drawbacks of COBRA analysis might one, but several sub-problems. One was that there was
include: a severe shortage of IT skills in the marketplace. But a
second problem was the cumbersome way in which IT
. Managers attribute the causes of behavioural pro- projects were managed anyway in the organisationÐ
blem to merely further symptoms rather than to showing itself as an `unable to prioritise e€ectively'
their ultimate root causes. symptom. A third dimension was added by the fact
T. Grundy / International Journal of Project Management 18 (2000) 93±103 97

Fig. 4. Cause of behaviour (COBRA) analysis.

that the company was the last place at which good IT actually mean by `in¯uence'. Ideally managers should
sta€ would choose to work. Each one of these symp- expose their reasons for judging `low' versus`high' in-
toms requires its own, very separate, ®shbone. ¯uence.

3.3. Importance±in¯uence analysis 3.4. Personal and strategic agendas

Whilst COBRA and Piranha analysis are essentially Coupled with the need to in¯uence behaviours (and
initial diagnostic tools, Importance±in¯uence analysis underlying agendas) we may need to dig deeper into
helps us to prioritise behavioural interventions. Fig. 7 those agendas for speci®c individuals. My earlier
helps us to identify which behavioural issues it is most paper [1] dealt with the more visible positions of key
bene®cial to address (especially in the north-east quad- stakeholders. Personal and strategic agenda analysis
rant). But also Fig. 7 may help to challenge thinking allows us to go much deeper.
about behavioural issues in the south-eastÐhow can Called the `PASTA' factor (because of our ®ndingÐ
the team get more in¯uence over these? at BT that both personal and strategic agendas are
Diculties which might arise with the importance± often inextricably intertwinedÐ-almost like spaghetti)
in¯uence grid include, for example, not having clarity this analysis can yield some major insights. Fig. 7
about what is most important versus least important shows a force ®eld analysis of one individual's per-
criteria. Or, managers might not be clear what they sonal and strategic agenda. The `enabling' forces are

Fig. 5. Attractiveness/implementation diculty ("AID") analysis.

98 T. Grundy / International Journal of Project Management 18 (2000) 93±103

ful for another manager (or managers) to tentatively

suggest or hypothesise that someone `might have cer-
tain X, Y and Z agendas'. The awkwardness of doing
this is signi®cantly alleviated if the enquiring man-
ager(s) explicitly say that they are having the equival-
ent of an `out-of-body experience'. This is usually
taken in a humorous vein, thus enabling the strategic
intervention to become ¯uid again.
Before we leave the topic of PASTA factors, it is
also useful to see that an individual may not always
have a coherent set of agendas. The very same individ-
ual may pursue one strategy one day, another strategy
on another day (sometimes called `stratophrenia'). [11]
A useful way of getting one's mind around this is to
use the notion of `sub-personalities'. For example, a
particular stakeholder might have several mini person-

Fig. 6. Piranha analysis. . A personality (A) that wants to be seen as being

`very clever'Ðactually, the cleverest in the team.
now called the `attractors' (what turns that stakeholder . A personality (B) which seeks to protect at all cost
on) whilst the constraining forces are called the `repel- what has worked in the past, especially where the
lers' (or what turns that stakeholder o€). person has put their own stamp of approval on it.
In order to get a good handle on someone's . A personality (C) (contrary to A and B) which is
`PASTA' factors, it is advisable to simulate the `out- actually quite helpful and supportive of change.
of-body experience' (OBE) which entails one imagin-
In doing the PASTA factor analysis you would need
ingÐand identifying withÐthe feelings, thoughts,
here to weigh up the likelihood of a particular sub-per-
habits and everyday concerns (and history) of that key
sonality coming to the fore in this situation.
stakeholder. PASTA factor analysis is an essential pre-
lude to creating behavioural scenarios for your stra-
tegic project. 3.5. Behavioural scenarios and `wishbone' analysis
One resistance to using PASTA factor analysis is
likely to be the discomfort which managers experience Behavioural scenarios involve semi-structured story-
when re¯ecting on their own agendas. Here it is help- telling about how the future of the strategic project

Fig. 7. Importance±in¯uence analysis.

T. Grundy / International Journal of Project Management 18 (2000) 93±103 99

Fig. 8. Personal and strategic agendas (PASTA factors).

may develop. A ®rst technique [1] is to plot some key . The ®rst, second or third order behavioural conse-
assumptions about the behavioural in¯uences in the quence of an event within or outside the strategic
project. Fig. 9 illustrates these, also identifying where project.
certain behavioural assumptions which might have
seemed to have been either of less importance or lesser Finally, you might identify one ideal behavioural
uncertainty can quickly move into the south-east of state of the project (or its behavioural vision) and then
the uncertainty gridÐits `danger zone'. try to map out all the behavioural factors which would
Where one or more assumptions occur in this need to be aligned in order to deliver that behavioural
danger zone, it is at that point that a particular beha- outcome. This can be drawn as a `®shbone'-like pic-
vioural scenario can be drawn out. One scenario is ture, except this time starting from the left-hand side
that the project team leader resigns and is replaced by of the page and working forward into the future.
another team leader who then immediately conducts Because the goal of this picture is not to diagnose a
an entire review of the strategic project with the result problem but to create an opportunity, and because it
being a radical change in its direction. deals with the future rather than the past, we call this
Behavioural scenarios can be re®ned by story-telling. technique behavioural wishbone analysis (Fig. 10) [9].
For example, one can pick out `transitional beha- Wishbone analysis is a powerful way of making sure
vioural events' that will lead us from the current state that not only the necessary, but also the sucient con-
of the strategic project to one which is quite di€erent. ditions are created to achieve a behavioural outcome.
Or, one might start o€ with a particular future and With wishbone analysis the main area to guard
then work backwards to de®ne the kind of behavioural against is that however imaginative managers are they
story-line which might lead up to that scenario (as in may fail to capture the one or two factors which still
the 1990s ®lm ``Back to the Future''). need to be aligned but currently are not. Wishbone
analysis is most e€ective when accompanied by scen-
A potential problem with using behavioural scen-
ario story-telling. This ensures that all the things
arios is that it is very possible that the scenario turns
which need to go right do go right.
out to be completely o€-beam. The very nature of
scenarios makes the possibility of mis-judging the
future a signi®cant risk. However the alternativesÐ
either of not looking into the behavioural future or 3.6. Behavioural diculty, energy and frustration over
extrapolating from the behavioural pastÐdo not seem time curves
viable. If there is real doubt that one particular scen-
ario fails to tease-out the main behavioural turning So far (with the exception of behavioural scenarios)
points, then develop a second or even a third one we have looked primarily at the deeper drivers of stra-
(Fig. 9). tegic behaviour associated with strategic projects
Or one might use the metaphor model of the uncer- rather than with the dynamics of those behaviours.
tainty tunnel [12] to understand: The following curves were piloted by companies
including Amersham International and Nokia.
. The behavioural antecedents of the projects. Particularly at Nokia these curves have proved es-
. The factors which might amplify or dampen beha- pecially helpful in encouraging managers to anticipate
vioural change a€ecting the project. future behavioural diculty.
100 T. Grundy / International Journal of Project Management 18 (2000) 93±103

Fig. 9. Uncertainty and importance mapping.

Fig. 11 plots behavioural diculty and energy over nitive revelation to uncover these patterns. This re-
time for a typical strategic project. Note that as the inforces the impression that despite being extolled to
behavioural diculty goes up, the energy of the team do the contrary, managers involved in managing stra-
frequently does down. This reduction in energy is due tegic projects manage from the `now' rather than
partly to the decline in energy as the team gets less `backwards from the future'.
enthusiastic generally, and also as a direct response to Clearly, these curves can help to ¯esh out the beha-
cumulative behavioural diculty. (Clearly, if our ear- vioural scenarios which are covered in the earlier sec-
lier diagnostic tools are not used, and also the tools tion.
from our earlier paper [1] are neglected, this is highly Finally, if we want to extend this dynamic thinking
likely to occur.) from the `D', `E', `F' (diculty, energy and frustration
Fig. 11 helps us to explain why many strategic pro- over time curves) to A to F we can also draw curves
jects become bogged down, especially three to six for:
months after inception.
Finally, Fig. 12 draws together the two strands of . AÐactivity (i.e. its level) over time
energy over time and frustration over time curves. . BÐbelief (in project success) over time
Again, as frustration mounts, this again saps the . CÐconfusion (i.e. its level) over time.
energy of the team unless it can somehow re-energise
itself or reduce its frustrations (or both). The major concern with applying the techniques of
Interestingly, managers experimenting (especially at plotting diculty, energy and frustration over time is
Nokia) with drawing these curves found it quite a cog- likely to reside in potential subjectivity. The way to

Fig. 10. Behavioural wishbone analysis. Fig. 11. Diculty and energy over time curves.
T. Grundy / International Journal of Project Management 18 (2000) 93±103 101

resolve this (besides getting the separate and/or inde-

pendent input from di€erent managers) is to ask prob-
ing questions like: ``Why do you think it will be so
dicult at time T?'' This in turn might invite a ®sh-
bone analysis.

4. Illustrations from the BT research

The action research process in the BT team involved

a number of
Fig. 12. Energy and frustration over time curves.
1. Initial interviews with the team.
2. Observation at an initial strategic workshop where
the managers interacted in their usual style.
3. Debrief interviews with managers, re¯ecting on than holding attention at a more `strategic level'.
their behaviourÐand what was considered func- Subsequent feedback showing the level and dynamics
tional versus dysfunctional by them (and why). of discussion (see Fig. 13) helped the team to under-
4. Feedback by the researcher to the team on what its stand the dysfunctional e€ects of excessive `picking
key behavioural patterns seemed to be (with discus- apart' behaviour. Fig. 13 shows the discussion going
sion). through high, medium or low levels of generality, as
5. Facilitation by the researcher of a further strategic categorised by the researcher. Although this was not,
workshop sessionÐwould/could the team shift its and could not be, coded with precision, it did prove
behaviour? helpful in discerning who in the team had the greatest
6. Observation of a ®nal strategic workshop session tendency to `rabbit hole' the discussions.
where managers had the chance to try out di€erent The team found it especially helpful to refer to
behavioural patterns to those which were manifest Fig. 13 which contrasted strategic thinking from `rab-
in their previous state. bit-hole thinking'. As discussion proceeded the team
7. Feedback of the overall ®ndings (and discussion). members were able to re¯ect ``are we actually in the
helicopter in our thinking, or have we gone down a
The overall process took place over a four-month rabbit hole?''
periodÐin order to allow any shifts in behaviour to Another important ®nding came out of Phases 5)
occur and to allow data to be digested. and 6). When the team had a narrower focus of atten-
The main ®ndings of the research [2] were as fol- tion and also when it was more able to share the cog-
lows: nitive maps and assumptions of key individuals, there
First, the BT managers themselves were frequently seemed to be far greater momentum and harmony in
frustrated at the slow progress of key strategic issues its behaviours. So the more cognitive clarity exists
and projects. Their discussions were often too open- within a team, ceteris paribus, the less behavioural tur-
ended, inconclusive and di€use to produce more bulence is likely to exist.
focused outputs. Their interactions were swayed by a
number of behavioural factors which appeared to han-
dicap their e€ectiveness. These ®ndings were generated
by the phases 1) to 3) of the research.
The cause-of-behaviour analysis (COBRA) which
was a central part of the feedback to managers is
shown in Fig. 14. This (tailored) ®shbone analysis
proved to be a very powerful intervention in the man-
agement team (research phase 4), as there was appar-
ently unanimous agreement `to see if we can try
something better'.
In the facilitated session (phase 5) the team again
ran into diculties as old behaviours (not surprisingly)
persisted. A most interesting ®nding was the tendency
of key members of the team to focus on more micro-
scopic issues (and often going o€ at a tangent) rather
102 T. Grundy / International Journal of Project Management 18 (2000) 93±103

Another interesting ®nding was that certain be- them incorporating several of the techniques into
havioural drivers had the tendency to govern the cog- future team sessions.
nitive subject matter of the managers. The main
drivers of this were the personal agendas of individuals.
These agendas contained a mix of emotional, territor-
ial and cognitive elements which were all highly inter-
5. Lessons and conclusion
mingled. So, for example, one manager held certain
views about the attractiveness of a particular technol-
Strategic projects have become a vital way of brid-
ogy with BT's strategy. His cognitive assumptions
ging strategic and operational management, but their
were imbued with considerable feeling. But at the same
very success is impeded by a lack of both analytical
time those views represented a particular `ideas terri-
and behavioural techniques. Whilst the analytical tech-
tory' which he had staked out as being `right and
niques have already been addressed, [11] arguably it is
appropriate for BT'.
equally important (and probably more dicult) to
This notion of `ideas territory' is likely to be very
manage the behavioural drivers impacting on strategic
helpful elsewhere, especially when a particular strategic
projects. Hopefully managers of strategic projects
project is complex and where a particular idea of `how
everywhere can now experiment successfully with those
it will work or not work' can come to dominate the
practical techniques for channelling strategic behaviour
wider project team's thinking. And to challenge the
for their projects.
dominant ideas territory is to make as much of a pol-
From research at BT (and related studies) a number
itical challenge as it is to challenge the speci®c in¯u-
of behavioural techniques now exist which provide
ence and power of an individual.
managers with some very practical and in-depth ways
Finally, when managers were able to self-regulate
of managing the strategic behaviour associated with
their own behaviour and cognitive interplay, some
strategic projects. A focus on strategic behaviour needs
very notable shifts in behaviours occurred. One
championing in the organisation as it represents a
particularly interesting change occurred when using
major shift in how things are done. This would require
the importance-in¯uence grid. As the BT managers
not merely input from internal human resources facili-
began to discuss what issues they had high versus low
tators or from outside but also leadership executive or
in¯uence over, they became remarkably calmer and
equivalent. For it is up to him/her ultimately if there is
more ¯uid. Instead of operating at a predominantly
to be a new openness and incisiveness in the inter-
low and microscopic level of debate they began to
action within senior teams.
glide e€ortlessly up and down through the levels. In
their own judgement, they delivered a good deal more
(and in a shorter period of time) than in their previous
To summarise, then, we found that: References

. Cause of behaviour analysis (COBRA) was a power- [1] Grundy AN. Strategy implementation and project management.
ful and e€ective way of managers' re¯ecting on past International Journal of Project Management 1998;16(1):43±50.
behaviour, feeding into real behavioural shift. [2] Grundy AN. Harnessing strategic behaviour. Financial Times,
. The strategic thinking/helicopter picture helped pro- [3] Johnson GJ. Strategic change and the management process.
vide some useful visual prompts to focus on certain Oxford: Basil Blackwell, 1986.
behaviours more and on other behaviours less. [4] Quinn JB. Strategies for changeÐlogical incrementalism.
. Narrowing the focus of strategic attention helped to Illinois: Richard D. Urwin, 1980.
gain more cognitive clarity which, in turn, calmed [5] Cyert RM, March JH. A behavioural theory of the ®rm.
Englewood Clift, NJ: Prentice Hall, 1963.
much of the behavioural turbulence. This was also [6] Mintzberg H. Patterns in strategy formation. Management
facilitated by the use of the importance-in¯uence Science 1978, 934±48.
grid which helped to provide a stabilising, cognitive [7] Hickson DH, Butler RJ et al. Top decisions: strategic decision-
focus. making in organisations. Oxford: Basil Blackwell, 1986.
. Re¯ection on personal agendas helped the team to [8] Senge P. The ®fth discipline: the art and practise of the learning
organisation. Doubleday, 1990.
challenge the dominant `ideas territory' which might [9] Argyris C. Strategy, change and defensive routines. Pitman,
have hindered more innovative thinking. 1986.
[10] Janis IL. Crucial decisions. New York: Free Press, Macmillan,
The team felt generally that the action research had 1989.
been a very fruitful process, both for re¯ecting on and [11] Grundy AN. Breakthrough Strategies for Growth. Pitman,
shifting their behaviours, and also for providing more 1995.
[12] Grundy AN. Exploring strategic ®nancial management.
shared structure in their cognitive debate. This led to Englewood Cli€s, NJ: Prentice Hall, 1998.
T. Grundy / International Journal of Project Management 18 (2000) 93±103 103

Dr Tony Grundy is a Senior Lecturer

in Strategic Management at Cran®eld
School of Management, and Director,
Cambridge Corporate Development.
He has worked with BP, ICI and
KPMG, and is author of six books
on strategy, ®nance and change
management, including Breakthrough
Strategies for Growth (Pitman, 1995),
Exploring Strategic Financial Manage-
ment (1998) (Prentice Hall) and
Harnessing Strategic Behaviour (1998)
(FT Publishing).