Moorman 1998 B

The Convergence of Planning and Execution: Improvisation in New Product Development
Author(s): Christine Moorman and Anne S. Miner

Source: Journal of Marketing, Vol. 62, No. 3 (Jul., 1998), pp. 1-20
Published by: American Marketing Association
Stable URL: http://www.jstor.org/stable/1251740 .
Accessed: 23/09/2013 16:38
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 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

All use subject to JSTOR Terms and Conditions
ChristineMoormanand Anne S. Miner
The Convergenceof Planning and

in New
Execution: Improvisation
Product Development
The field of marketing strategy often makes the important assumption that marketing strategy should occur by first
composing a plan on the basis of a careful review of environmental and firm information and then executing that
plan. However, there are cases when the composition and execution of an action converge in time so that, in the
limit, they occur simultaneously. The authors define such a convergence as improvisation and develop hypotheses
to investigate the conditions in which improvisation is likely to occur and be effective. The authors test these hy-
potheses in a longitudinal study of new product development activities. Results show that organizational improvi-
sation occurs moderately in organizations and that organizational memory level decreases and environmental
turbulence level increases the incidence of improvisation. Results support traditional concerns that improvisation
can reduce new product effectiveness but also indicate that environmental and organizational factors can reduce
negative effects and sometimes create a positive effect for improvisation. These results suggest that, in some con-
texts, improvisation may be not only what organizations actually practice but also what they should practice to
flourish.
he fields of marketing strategy in general and new lief that marketingstrategies should, in general, first be for-
product development in particularappear to assume mulated and then implemented.
that marketingstrategyshould occur by first compos- Much of the researchliteraturein marketingshares this
ing a plan on the basis of a careful review of environmental view. For example, a review of published articles cited in a
and firm informationandthen executing thatplan. In this ar- recent special issue of the Journal of MarketingResearchon
ticle, we question this assumptionby suggesting that there innovation and new products reveals a tendency to view
are cases when the composition and execution of an action planning as the accepted norm while acknowledging that
converge in time so that, in the limit, they occur simultane- business environmentsare becoming increasingly dynamic
ously. We define such a convergenceof compositionand ex- (Windand Mahajan1997). Studies thereand elsewhere sug-
ecution as improvisationand suggest that the narrowerthe gest thatproductdevelopmentcycle times are faster(Griffin
time gap between composing and performing(or planning 1997), failure rates lower (Cooper and Kleinschmidt 1986;
and implementation),the more that act is improvisational. Montoya-Weiss and Calatone 1994), financial returns
Marketingliterature,similar to many businesses fields, greater(Ittnerand Larcker1997; Song and Parry 1997), and
has paid relatively little attention to extemporaneous be- innovationlevels higher (Olson, Walker,and Ruekert 1995)
havior and has taken a ratherstrong stand in favor of mar- when companies take certainadvancedplanning steps.
keting planning, especially of a formal type (e.g., Our purposein this article is not to detractfrom the val-
Armstrong 1982; Sinha 1990). Although there have been ue of planning.We believe that planningis an importantas-
important exceptions in marketing and related literature pect of effective marketing management and decision
(Holbrook 1995; Hutt, Reingen, and Ronchetto 1988; making.However,priortheoryalso points to severalreasons
March and Simon 1958; Mintzberg 1994), a rational plan- why improvisationsometimes can be a valuable and effec-
ning norm has become deeply entrenched. Table I tive approachto marketingaction. First, improvisationcan
overviews a sample of marketingand productstrategytext- be an effective choice when a firm faces environmentaltur-
books that indicates a fairly uniform acceptance of the be- bulence that requiresaction in a time frame that is shorter
than a regularplanningcycle. For example, Egge (1986) de-
scribes how a salespersonmight improvisewhen immediate
action is requiredin the face of changing client demands;
Moorman
Christine isAssociate ofMarketing
Professor andAnneS.Miner
Graduate Uni-
SchoolofBusiness, Dickson (1997) suggests that fast learning and adaption
Professor
is Associate ofManagement,
Thisresearch
ofWisconsin-Madison. fromthesupport
benefited of without much advance planning are importantto firm sur-
versity
BobDrane,GaborKemeney,
PaulaBassoff,AricRindfleisch, Stephanie vival; Moormanand Miner (1995) describe how a team im-
Dixon,DavidRobinson,andRonaVelteandthecomments of PeterDick- provised a new product formula in response to a surprise
son,KathyEisenhardt,
MaryJoHatch,JanHeide,andThekla Ruraonpre- introductionof a competitive product;and Eisenhardtand
vious versionsof the article.NationalScience Foundation Grant Tabrizi (1995) find that an experiential strategy involving
SBR-9410419, andtheUniversity
ScienceInstitute,
theMarketing ofWis-
Fundhavesupported
consinSchoolofBusinessSabbatical thisresearch. improvisationworks better in the computer industry when
productdevelopment must operate within the high level of
Journal of Marketing
Vol. 62 (July 1998), 1-20 Improvisationin New ProductDevelopment/1

TABLE 1
Overview of How Marketing Strategy Textbooks View the Timing of Planning and Implementation
Aaker (1988) Figure2-1 provides an overview of the strategic market management process. This figure shows
external and self-analysis occurringfirst, followed by strategy identificationand selection, which
includes, in order: mission specification, strategic alternative identification, strategy selection,
implementation,and evaluation (p. 22).
Boyd and Walker The authors define strategy as "a fundamental pattern of present and planned objectives, re-
(1990) source deployments, and interactionsof an organization with markets, competitors, and other
environmental factors"(p. 43). This definitionemphasizes planning, or the setting of objectives,
as an integralpart of strategy formulation.
Boyd, Walkerand Exhibit1-7 depicts the marketingmanagement process as moving from analysis to the formula-
Larreche(1998) tion of strategic marketing programs, and finallyto the implementation and control of such pro-
grams (p. 16).
Cravens and Lamb "Strategicplanning is a continuingcycle of making plans, launching them, trackingperformance,
(1993) identifyingperformance gaps, and then initiatingproblem-solvingactions" (p. 688). The authors
provide eight comprehensive steps in preparingand implementing the strategic marketing plan
(Exhibit1, p. 688).
Dalrympleand "Once marketingplans have been prepared, they are used to guide field marketing activities for
Parsons (1990) the planning period."Marketingperformance should be monitored to compare results with the
goals of the marketingplan (p. 822).
Day (1990) Although Day incorporates both bottom-up and top-down approaches to decision making, the
implicitstrategy model in his book is that strategic planning precedes implementation and execu-
tion of the plan. This is evidenced in Figure2-1 and because implementation is not an important
part of the text; instead, the emphasis is on strategy development.
Dickson (1997) Defines improvisationas "impromptuaction" and notes (p. 37), "An organization's survival de-
pends on its abilityto learn and adapt quickly;in practice, this means that plans often must be
altered at the very time they are being implemented." He sums up his position on improvisation
when he notes (p. 399), "Insummary,the ying and the yang of product development is the plan-
ning disciplineof up-frontcontinuous environmentalanalysis, targeting/positioning,product speci-
fication,judicious stage-gate reviews of the emerging product's fit, feasibility, and estimated prof-
itabilitycombined with the creative improvisationfrom a team's many iterationsof prototype de-
sign and testing."
Jain (1997) Exhibit2-4 depicts the process of strategic marketing, which involves a clear separation of strat-
egy development and strategy implementation(p. 33).
Kerin and "Theselection of a course of action must be followed by development of a plan for its implemen-
Peterson (1995) tation. Simply deciding what to do willnot make it happen. The execution phase is critical,and
planning for it forces [you]to consider source allocation and timing questions" (pp. 35-36).
Kotler(1994) Figure 3-1 (p. 63) shows a linear process moving from planning to implementing to controlling.
The process has feedback loops from controllingback to implementationand planning.
Lehmann and "Ingeneral, the planning process works as shown in Figure 2-4. Whereas the collection and
Winer(1994) analysis of data and the development of product strategies takes place in a limited time frame,
there is no beginning or ending to the planning process as a whole. The formal part of the proc-
ess is followed by implementation,during which programssuch as distribution,promotion, adver-
tising, and the like are executed. Monitoringand evaluating both the performance of the plan
and changes in competition and customers in the external environment are also continuous
tasks. This informationfeeds back to the formal planning part of the process" (p. 31).
Peter and Donnelly "[T]heorganization gathers informationabout the changing elements of its environment.... This
(1998) informationis useful in aiding the organization to adapt better to these changes through the
process of strategic planning.The strategic plan(s) and supporting plan are then implemented in
the environment"(p. 10).
Quelch, Dolan, and 'The marketingprocess can be divided in several different ways. (Our)conceptualization of mar-
Kosnik(1993) keting tasks is: (1) marketing research, (2) marketing strategic formulation, (3) marketing plan-
ning, programming,and budgeting, and (4) marketingorganizationand implementation"(p. 9).
2 / Journalof Marketing,July 1998

uncertainty created by quickly changing markets and provisation is effective. As we have reviewed previously,
technologies. one research tradition is based on the belief that a lack of
Second, improvisationmight be prompted when plan- advance planning reduces the chances of a firm's success
ning has not providedall the details or tactics of implemen- (e.g., Cooper and Kleinschmit 1986, 1987). Otherresearch,
tation. Quinn's (1980, 1986) investigation of ten large however, has found success stories when organizations
organizations finds that these organizations refined their have improvised. For example, Mintzberg and McHugh
general strategic course incrementallyas new information (1985) detail the effective improvisations of the National
emerged from the environment.Likewise, researchershave Film Board of Canada;Preston(1991) describes a group of
reportedthat many tactical marketingdecisions are not in- managers that improvises an effective solution to a manu-
cluded in marketingplans (Cosse and Swan 1983; Sutton facturingplant strike; Eisenhardtand Tabrizi (1995) show
1990). that an experiential approach involving improvisation ac-
These reasons for improvisationare consistent with a celerates product development in the computer industry;
more behaviorally based view of strategy development, Weick (1993a) describes the effective improvisationssev-
which asserts that organizationallaction often occurs with- eral firefighters used to escape the firestorm at Mann
out much advance planning (Cohen, March, and Olsen Gulch; and Pascale's (1984) portraitof Honda's successful
1972; Cyert and March 1992; Pfeffer 1982; Weick 1979). introductionof its 50cc bikes into the U.S. marketinvolves
This research traditiontherefore tends to focus on the be- improvisation.
havioral dynamics that produceeffective action as opposed It is not our goal to suggest that improvisation inher-
to documenting whetherorganizationsadhere to normative ently is either helpful or harmfulfor organizations.Instead,
models of action (Burgelman 1983; Miner 1987, 1990; we study the literatureon improvisationto uncover the sys-
Mintzbergand McHugh 1985). tematic influence of various factors that affect whether im-
Several marketingscholars have followed in this tradi- provisation hurts or helps organizations. We seek to
tion and suggest that there is a gap between normativeand supplementcurrentknowledge regardingthe types of envi-
descriptive accountsof marketingstrategy(Anderson 1982; ronmentalconditions and organizationalcompetencies that
Day and Wensley 1983; Wind and Robertson 1983). For ex- might determinethe effectiveness of improvisation.There-
ample, Hutt, Reingen, and Ronchetto (1988) find that un- fore, our objectives are twofold: (1) to investigate the con-
planned, innovative new product activities, or autonomous ditions in which improvisationis likely to occur and (2) to
strategicbehavior(Burgelman1983), occur in organizations examine the conditions in which improvisationis likely to
and that marketingplays a key role in such activities by be effective.
virtue of its boundary-spanningand product-championing
behavior.
Therefore,there is a precedentboth in and outside mar- Improvisation
keting literaturefor examining the topic of unplanned,inno- Definition and Discrimination from Related
vative behavior.However, there are few cases of systematic Constructs
empirical investigationsof improvisationspecifically. Out- Our review of priorresearchsuggests that an assessment of
side marketing,researchhas tendedto focus on depthanaly- whether improvisation occurs requires looking not just at
sis of single improvisations by persons, groups, or whathappens,but also at the temporalorderin which it hap-
organizations (e.g., Hutchins 1991; Preston 1991; Weick pens (see also Moormanand Miner 1998). Observerstypi-
1993a). In marketing,researchhas been qualitative(Moor- cally assume thatcomposition or planningoccurs first and is
man and Miner 1995) or focused more broadly on innova- followed at a later time by implementationor execution. In
tion (Hutt, Reingen, and Ronchetto 1988).
improvisation,the time gap between these events narrowsso
We build on these importantobservationsby developing
that, in the limit, composition converges with execution
and systematicallytesting a frameworkthat studies the inci-
(Weick 1993a). Therefore, the more proximate the design
dence of improvisationin new product activities. Because and implementationof an activity in time, the more that ac-
improvisationis a type of innovative behaviorthat often in- tivity is improvisational.
volves fast learning(Eisenhardtand Tabrizi1995), we draw This definition is consistent with core conceptualiza-
on existing literatureon innovationand learningto make our tions of improvisationin several bodies of literature.For ex-
case for why certainenvironmentaland organizationalfac-
ample, improvisationis referredto as "thinkingin the midst
tors are importantto improvisation.However, we are also of action"in education (Irby 1992, p. 630), occurringwhen
careful to show that improvisationis distinct from innova- "actsof composing and performingare inseparable"in com-
tion and that these factors provide special insight into im- munication (Bastien and Hostager 1992, p. 95), "reading
provisation,given its extemporaneousnature. and reacting in parallel" in sports psychology (Bjurwill
In addition to examining the incidence of improvisa- 1993, p. 1383), "real-timecomposition"(Pressing 1984, p.
tion, we also seek to understandthe conditions in which im- 142; Pressing 1988) and "making decisions affecting the
this arti- composition of music during its performance"(Solomon
'Althoughwe use the termorganizational throughout 1986, p. 226) in music, and representing"no split between
cle, we do not meanto suggestthatthe focalactionis necessarily
level. In general,measurement
at the overallorganizational of or- design and production" in organizational studies (Weick
ganizational phenomena occurs within a strategicbusiness unit 1993a, p. 6). By focusing on the simultaneityof events, this
(Eisenhardt andTabrizi1995). researchalso follows in the traditionof organizationaltheo-
Improvisationin New ProductDevelopment/3

ries of temporalorder(Cohen, March,and Olsen 1972; Van improvise in different settings. Weick (1993a) describes in
de Ven 1986, 1993). detail the improvisationaland other actions of individual
Improvisationalso can be distinguishedfrom other, re- firefightersin a disastrousfiretrap,though he also explores
lated concepts in literature(Moorman and Miner 1998). team aspects of the situation.
Most important,we argue that improvisationis a special In addition to individual improvisation,observers have
case of intraorganizationalinnovation,which is defined as emphasized that collective improvisation also occurs
deviation from existing practices or knowledge (Rogers (Crossanand Sorrenti 1997; Preston 1991; Weick 1993a, b,
1983; Zaltman,Duncan, and Holbek 1973). All improvisa- c). One detailed report (Hutchins 1991) describes actions
tion, by definition, involves some degree of innovationbe- taken by the crew of a ship whose navigationalsystem had
cause improvisationinvolves the creationof action outside brokento make theirway into a harbor.To avoid danger,the
currentplans and routines.Thereare many otherkinds of in- crew memberscalled out estimates of coordinates,calculat-
novationbeyond improvisation,however.For example, if an ed subpartsof the data needed to make navigationalchoic-
organizationinnovates a new way to distributea productby es, and communicated partial information to one another
analyzing customer needs, gatheringfacts, and planning a repeatedly.Although a transcriptof their interactionsindi-
new channel, the organizationhas innovatedbut not impro- cates that no one crew member understood the complete
vised. In addition,thoughall improvisationhas some degree system they had improvisedor exactly why they were suc-
of innovation,the degree of innovativenesscan vary enor- ceeding, the crew developed a set of routinesthat workedto
mously. For example, some improvisationsare very innova- get the ship into the harbor.
tive and deviate far from existing routines, such as in the We follow this approachby focusing on organizational
case of "freejazz" (Berliner 1994) or of NASA teams im- improvisation,which includes improvisationby groups, de-
provisingto rescueApollo XIII by makingradicaluse of ob- partments,or whole organizations.Many observersassume
jects outside prior routinesor structures(Lovell and Kluger that such entities generateand execute plans. Therefore,we
1995). Other improvisationsonly involve minor deviations attemptto study them as carryingout these activities either
from currentroutines, such as when musicians add embell- in sequence or nearly simultaneously,and thus improvising
ishments to existing melodies (Bailey 1980) or productde- or not. Although the nature of collective or organizational
velopment teams add features to existing products(Miner, features remains contentious (Argyris and Schon 1978;
Moorman,and Bassoff 1997). Walsh 1995), we follow other work describing organiza-
Improvisationis also distinct from other constructsim- tional features, such as culture (Deshpande, Farley, and
portantto strategicfirm behavior,such as adaptation,learn- Webster1993; Deshpandeand Webster1989), organization-
ing, and opportunism.First, improvisationis distinct from al informationroutines(Jaworskiand Kohli 1993; Kohli and
adaptation, which involves the adjustmentof a system to Jaworski1990; Moorman1995), and memory(Cohen 1991;
externalconditions (Campbell 1989). We arguethat adapta- Huber 1991; Moormanand Miner 1997; Walsh 1995; Walsh
tion does not necessarily invoke improvisationalaction by and Ungson 1991). To highlight the natureof organization-
organizations.Instead,adaptationcan be achieved in a vari- al improvisation,Appendix A provides examples of firms
ety of ways, including both preplanningand improvisation. varying in the degree of improvisationin their new product
Second, if learning is a process that requiresthe discovery, activities.
retention, and exploitation of stored knowledge, including Priorresearchsuggests that interactionsamong persons
informationor behavioralroutines(Huber 1991; Levitt and who are improvisingfrequentlyproducescollective improvi-
March 1988), then not all learningis improvisation.For ex- sation. In an improvisationaltheatricalgroup, for example,
ample, learningmight involve a carefullypreplannedexper- one actormight make a comment,to which a second will re-
iment whose results are recordedand interpretedand that spond with an associationbetween the commentand another
requires no improvisationwhatsoever.Third, when organi- topic, and then a thirdactormight link these issues to a third,
zations are described as opportunistic, they are likely to inclusive topic (Crossanand Sorrenti1997; Mangham 1986;
seize attractive,unexpected developments or opportunities Spolin 1963). The theatricalgroup did not plan the scene in
proactively (Aaker 1988; Miner 1987; Mintzberg and advance,and the patternthatarises is not a simple sum of in-
McHugh 1985; Quinn 1980, 1986).2As with the other dis- dependentimprovisationalactions but a collective system of
criminationvariables,opportunismcan be achieved through interactionthat createsand enacts the scene simultaneously.
means other than improvisation,such as developing a plan. The idea that a system of interactioncan produce col-
In addition, improvisation sometimes arises in disastrous lective improvisation also is supported by Hutchins's
situations and involves overcoming obstacles more than it (1991) case study of the crew improvising a solution to
does taking advantageof unexpectedopportunities. their ship's failed navigationsystem and by Barley's (1986)
observationthat hospital technicians and radiologistsjoint-
Organizational Improvisation
ly improvisedroutines in response to new technology. In a
Many observers have described improvisationby individu- powerful set of case studies displaying improvisation,
als. For example, people have describedways in which in- Dougherty (1992) describes new product development
dividual actors, athletes, therapists,musicians, or teachers teams interactingin ways that do not follow established or-
ganizationalroutines and notes, "Successful developers vi-
2Thisviewof opportunismas benign(Miner1987,p. 334)omits olated ... routines and created a new social order for their
the definitionof opportunismas "self-seekingwith guile"that collaborativeefforts. They developed mutuallyadaptive in-
Williamson(1975)proposes. teractions in which knowledge of the work was developed

as the work unfolded"(p. 192). In one example, Dougher- tional styles in productdevelopmentall focus on settings in
ty (1992) cites the case of SALECO's development team, which unexpectedstimuli create the need for organizational
which broke routines by using products assembled from action but also weaken the effectiveness of prior planning.
off-the-shelf parts purchased externally rather than those The central premise of this line of thinking is that improvi-
manufacturedin-house. This team also opted to introducea sation might have special value in these circumstances
softwareproductwith some bugs in it instead of holding to (Miner 1987). We formalize the logic of this stream of
the routineof perfect quality control, because they realized thought in the next three hypotheses, which are shown in
thatusers caredmore aboutthe numberof applicationsthan Figure 1.
they did about perfect operation. The impactof environmentalturbulence.When the envi-
These examples and others suggest that collective im- ronmentin which an organizationoperatesexperiences a lot
provisationoften builds on and incorporatesindividualim- of change, the organizationhas several choices. It can ignore
provisation.However, individualimprovisationalone is not externaldemandsor shocks that suggest the need to change
sufficient for collective improvisation.Instead,the joint ac-
plans and continue with previously plannedactivities; it can
tivities of individualpeople createa collective system of im-
attemptto speed up its planningand execution cycles so that
provisational action. In addition, there are occasions in they remain distinct but happen more quickly (Eisenhardt
which a person's behavior, planned or improvisational, and Tabrizi 1995); or it can move towardan improvisation-
sparkscollective activities that are improvisationalin nature al approachthat merges planningand execution processes.
(Burgelman1983; Hutt, Reingen, and Ronchetto 1988). For In some cases, fast-changingenvironmentscan destroy
example, in the case of plannedor deliberateindividualbe- the value of existing competencies (Tushmanand Anderson
haviorcreatingorganizationalimprovisation,Eisenhardtand
1986). In such circumstances, organizations might find it
Tabrizi(1995) find that leaders'deliberatebehaviorsplayed
necessary to improvise or compose new behaviors while
an importantrole in speeding the developmentof highly it-
executing them. As Weick (1979, p. 102) states, "If there
erative and experientialnew productdevelopment(see also exists a truly novel situation, one in which there is no anal-
Miner 1987; Quinn 1986). Other times, collective improvi-
ogous experience in the past, then the only thing the person
sation results from individual behavior that is itself highly can do is act." In other words, strategy implementationac-
extemporaneous(Mintzbergand McHugh 1985). In either tually can be "madeup as firms go along" (Weick 1993c, p.
case, the individualimprovisationmust move to the organi- 2). Consistent with this view, organizationalscholars have
zationallevel for collective improvisationto occur.In short,
arguedthat the increasedpace of competition might require
theremust be an element of collective design and execution.
organizationsto develop an improvisationalcompetency to
prosper (Brown and Eisenhardt 1995; Eisenhardt and
Tabrizi 1995; Mintzberg and McHugh 1985). The basic
Conceptual Framework logic here is thatexogenous shocks or demands come along
Having established the natureof improvisationand its po- more rapidly than an organizationcan anticipate,and orga-
tentialimportanceto contemporaryorganizations,we devel-
nizations often respond to such situations by improvising
op arguments about the incidence and effectiveness of ratherthan not responding.We hypothesize that
improvisationin organizations.
Hi: The greaterthe level of environmentalturbulence,the
Factors Influencing the Incidence of greaterthe incidenceof improvisation
organizational in
Improvisation new productactions.
Theory and prior researchsuggest many factors that might The impact of real-time organizational information
enhance the chances that improvisationwill occur in orga- flows. The logic of responsiveness implies that awareness
nizationalactivity. First, improvisationmight occur because of external or internal surprises can trigger organizational
of a lack of organizationaldiscipline, so thatan organization improvisation. Therefore, the more an organization main-
makes up new plans as it goes along simply because it lacks tains access to informationflows, the more likely it is to be-
the rigor to follow prior plans (Cooper and Kleinschmidt come aware of either external shocks or unexpected
1986; Etzioni 1964). Second, an organizationdeliberately internalsurprises.
might encouragespontaneousactivities thatare inconsistent Some literaturepoints to a particulartype of information
with priorplans or activities, suggesting that it has "learned flow-that which flows in real-time interaction among
to improvise" (Burgelman 1983; Hutt, Reingen, and group members-as an importantstimulus to group impro-
Ronchetto 1988; March 1976; Moormanand Miner 1998). visation. Bastien and Hostager (1988), for example, docu-
Third, improvisationmight occur within what we call the ment the nonverbalcues band membersgive one anotherin
logic of responsiveness. This stream of thinking suggests jazz improvisations.Likewise, Spolin (1963) points to the
that organizationssometimes face unexpectedjolts or sur- criticality of real-time cues flowing among improvisational
prises that make priorplans irrelevantor incomplete in im- theaterplayers as their scenes unfold. In addition, real-time
portantways. Such jolts often are coupled with a context in informationflows between the actors and the audience not
which it is difficult to refrainfrom takingaction or complete only inform but also stimulate specific improvisationalac-
a new planningcycle before taking action. Weick's (1993a) tivities. For example, a troupe might extend an improvisa-
work on improvisationby firefighters,Preston's(1991) ex- tional skit, spurredby real-time audience reactions.
ample of improvised decision making during a strike, and We define real-time informationflows as those that oc-
Eisenhardtand Tabrizi's (1995) findings about improvisa- cur during or immediately prior to an action (Eisenhardt
Improvisationin New ProductDevelopment/ 5

FIGURE 1
Factors Influencing the Incidence and Effectiveness of Organizational Improvisation
MainEffects
- - - -- ModeratingEffects
1989). These flows contrast to informationprocesses that H2:The greaterthe level of real-timeorganizational
informa-
occur well in advance of an action or that are used after an tion flows, the greaterthe incidenceof organizational
action to evaluate its impact. Real-time informationflows improvisation in newproductactions.
are likely to occur in face-to-face interactionsand electron- The impactof organizationalmemory.We have suggest-
ic communications,in which there are few time delays and ed that real-time informationabout internalor external sur-
great opportunitiesfor feedback (Eisenhardt1989; Sproull prises might enhance the chances that improvisation will
and Keisler 1991). Therefore,team meetings in which deci- occur. In contrast,prior work suggests that stored informa-
sions and interpretationsare made and behaviorsare carried tion, in the formof organizationalmemory,will reducethese
out are often sources of real-time informationflows (Dick- chances. As with organizational improvisation, there is
son 1997). some disagreement regarding whether organizations store
Using our logic of responsiveness,we predictthatheav- informationin memory as people do. However, there is a
ier real-timeinformationflows will createmore possibilities growing sense across disciplines that organizations have
for organizationsto be exposed to unexpected information frames of reference,routines, and structuresthat reflect the
that invites improvisedaction. We make this predictionfor presenceof storedknowledge (for a review of this literature,
three reasons. First, real-time informationis, by definition, see Cohen 1991; Cohen and Bacdayan 1994; Cohen and
timely. Therefore,unlike informationthatmay get to a deci- Levinthal 1990; Walsh 1995; Walshand Ungson 1991; Win-
sion maker too late for action to be taken, real-time infor- ter 1987). We adopt that perspectivein this articlebut focus
mation is inherently more actionable. Second, because it on the level of knowledge contained in an organization's
occurs during or immediately prior to an action, real-time memory, which we previously have defined as collectively
informationhas an urgency to it that is likely to evoke im- held beliefs, behaviors, or physical artifacts(Moormanand
mediate responses, which probably cannot be planned. Miner 1997, p. 93). Therefore,a high level of organization-
Third, real-time informationflows are more novel because al memory would be presentwhen a projector action phase
they evolve in a more randommannerthannon-real-timein- representsfamiliar territory,a new product requiresonly a
formation.For example, real-timeinformationflows during modest change in an old project, the technological or cus-
a meeting would be more likely to involve unexpected in- tomer basis for the new product is part of the firm's long-
formation than those emanating from a series of written standingrepertoire,there are well-establishedteam routines
memos. These qualities provide greaterpotentialfor impro- because the durationof the team members' service is high,
visational activity.We predictthat or a particularaction phase (e.g., prototypedevelopment) is
6 1 Journalof Marketing,July 1998

an establishedfirm-level competency (Moormanand Miner Preston 1991; Weick 1993a, c, 1996). For example, when
1997). improvisation works well, it can produce aesthetically
By definition,organizationalmemoryrepresentslearned pleasing outcomes in a theatricalor musical setting (Hatch
ways of thinking and behaving. As such, memory often is 1997). Likewise, it can provide instrumentalvalue for orga-
activated automaticallyin certain situations. For example, nizations or groups by solving problems or capitalizing on
firms tend to develop line extensions of existing products uniqueopportunities(Weick 1993b, 1996).
ratherthancreatecompletely new ones (Andrewsand Smith Here, we investigate several types of outcomes that fo-
1996). Firms also tend to use well-developed routines and cus on organizationaleffectiveness, which we define as the
processes for developing and introducing new products degree to which an action achieves instrumentaloutcomes
(Day 1994; Leonard-Barton1992; March 1991; Moorman of value for a firm (Walkerand Ruekert 1987). We focus on
1995; Moormanand Miner 1997) and thereforelearn fewer two types of effectiveness outcomes: productand process.
new routines(Sinkula 1994). For productoutcomes, we investigate design effective-
Preston (1991, p. 89) discusses the negative effect that ness, defined as the degree to which new product features
priormemory is likely to have on the incidence of improvi- are high quality and high performance,and market effec-
sation, noting:"Inthe case of these familiarsituations... the tiveness, defined as the degree to which the new product
scope for improvisation is more constrained."Likewise, meets the demands of target customers. An organizational
jazz musicians have commentedon the paradoxof needing example of these outcomes might be a productdevelopment
to learngreat artists'works to improvise well but then find- team's improvisationof a new casing unit for a previously
ing themselves trappedby this learning (Weick 1993c). As unprotectedpart on a commercial research instrument.Ef-
Berliner(1994, p. 206) notes, "Inone of the greatestironies fective improvisationof the unit would requirethat the ma-
associated with improvisation,as soon as artists complete terials and size of the casing match each other and reflect a
the rigorouspracticerequiredto place a vocabularypattern quality level (design effectiveness), and that the unit fit the
into theirlargerstore, they must guardagainst its habituated customer's needs so that the product works in the settings
and uninspireduse." for which it was designed (marketeffectiveness).
The tendency for existing knowledge to restrict the Improvisationnot only influences the effectiveness of a
rangeof options is a common challenge for innovationof all new product,but, we reasoned, it also could affect the ef-
types but is an especially strong impedimentto improvisa- fectiveness of the new product development process. Two
tional action. In improvisation,the time between composing indicatorsof term process effectiveness seemed particularly
and executing is small and/or nonexistent.We suggest that important:teamfunctioning,which refersto the team's com-
the pressureof fast action enhances the possibility that an mitmentlevel to the project,and team learning, the level of
organization will rely on existing routines, regardless of knowledge the team gains in performinga new productac-
whethera learnedresponse is warranted.Therefore,we hy- tion. Finally, the effectiveness of new productdevelopment
pothesize that processes is also, to some degree, revealed in the cost effi-
ciency (financialinvestmentlevel) and time efficiency (time
memory,thelowertheinci-
H3:Thegreatertheorganizational investmentlevel).3 The next three sections relate improvisa-
improvisationin new product
dence of organizational
tion to these productand process outcomes.
actions.
Themoderatingimpactof environmentalturbulence.We
Factors Influencing the Effectiveness of predict that the rate of environmentalturbulencewill mod-
Improvisation erate the improvisation-effectivenessrelationship,because
In the priorsection, we highlightedthreefactorsthatwe pre- it shifts the advantages and disadvantages of formal plan-
dict will influence the incidence of improvisation.The val- ning versus improvisation.If, for example, an organization's
ue or effectiveness of that improvisation,once engaged in environmentis stable and continuous, planning in advance
by a firm, is anothermatter.There has been a tendency to of action offers that organizationmany possible advantages
think of improvisationas either helpful or hurtfulto organi- (e.g., Armstrong 1982; Miller and Cardinal 1994). The or-
zations, as we noted previously.We addressthese equivocal ganizationcan take the time to do complete planning,know-
perspectives by identifying selected factors that moderate ing that the assumptions and facts guiding the plans
the impact of improvisationand determinewhether impro- probably will still hold at the end of the planning cycle.
visation benefits or hurts organizations.We begin our dis- Thus, the organizationcan harvest the tremendouscoordi-
cussion of these factors by identifying the new product nation and control benefits of good planning, including
development outcomes we believe might be potentially as- avoiding inconsistentand wasteful action, coordinatingac-
sociated with improvisation.We then considerhow the same tivities of multiple actors who may not communicateeasily
importantinformationalfactors that influence the incidence
of improvisationalso might moderate its impact on these
outcomes. This dual role creates several potentialtrade-offs
for organizationsto manage. 3Although it is moretypicalto separateefficiencyandeffective-
ness,productdevelopment efficiencycaninfluenceoverallorgani-
Focal new product development outcomes. There are zationaleffectiveness.Therefore,to maintaina common,unifying
several outcomes typically associated with improvisation focuson newproducteffectiveness,we placeefficiencyunderthe
(Brown and Eisenhardt1995; Eisenhardtand Tabrizi 1995; rubricof effectivenessin ourmix of dependentvariables.

with one another,and assuringthatmost actions are focused importantto the success of the new productoutcomes. We
on a single goal. Improvisationoffers no particularadvan- predict that
tage in this setting. Therefore,in the presence of low envi- H5:The greaterthe level of organizational real-timeinforma-
ronmental change, improvisation will be more disruptive tion flows, the greaterthe likelihoodthat improvisation
than helpful. will generateeffective(a) productsand (b) processesin
In contrast,rapidenvironmentalturbulenceincreasesthe new productdevelopment.
odds that improvisationwill provide value. Extensive for- The moderating impact of organizational memory.Al-
mal plans in such conditions can have negative conse- though H3 predicts that organizationalmemory will reduce
quences because they consume time and resources and the incidence of improvisation,there is considerable prior
providecounterproductiveguides to action when the context research supporting the idea that a strong organizational
changes faster than the planning cycle (Eisenhardt and memory will enhance the effectiveness of improvisational
Tabrizi 1995). These circumstances,therefore,actually im- action. First, much improvisationappearsto arise from the
prove the chances that improvisationwill be effective. We recombination of previously successful subroutines of
propose the following: knowledge and action (Borko and Livingston 1989; Levi-
H4:The greaterthe level of environmental turbulence,the Strauss 1967; Nonaka 1990). For example, a firm with well-
greaterthe likelihoodthat improvisationwill generate developed marketingresearchcompetencies and significant
effective(a) productsand (b) processesin new product consumer insight could recombineexisting knowledge and
development. skills to improvise new strategiesto respondto unanticipat-
The moderating impact of real-time organizational in- ed changes in consumer behavior. In support of this idea,
formationflows. Real-timeinformationflows also can facil- Weick's (1993a) analysis of the Mann Gulch disaster em-
itate a positive effect for improvisation by playing a phasizes that the ability of key members of the fire team to
powerful coordinatingrole. One of the crucial functions of use their existing skills in a novel way was crucial to effec-
plans is to coordinate the action of multiple actors (Gal- tive improvisation. Music observers note that musicians
braith 1973). In the absence of plans, coordinationmust oc- with strong preexistingrepertoiresof melodies, chords, and
cur through other mechanisms. Immediate information rhythmsand familiaritywith otherplayers producethe most
aboutthe context in which the action is occurringandthe ac- powerful improvisations(Berliner 1994). Weick's (1993b,
tions of other participants enables such coordination p. 353) characterizationof organizationdesign as involving
(Bastien and Hostager 1988, 1992; Menzel 1981). In the- improvisationreflects the importanceof memory.He states,
atricalimprovisation,for example, actorscontinuallyattend "If we think of designers as people who improvise, then the
to and process instant informationon audience reaction to materialsthey have availableto work with are the residue of
guide their subsequentactions (Spolin 1963). This feedback their past experience and the past experience of people in
replaces the coordinatingfunction of a plan because the ac- their design group, the meanings attachedto this past expe-
tors respond to the same audience cues (Huber and Mc- rience, observationalskills, and their own willingness to re-
Daniel 1986). ly on imaginativerecombinationof these materials."
In a product development context, Robins (1991) de- The now-famousaccountof Honda's introductionof the
scribes a thriving company that introducesnew productsat U.S. Supercubin the United States provides a final example
least once a month but has no formal strategicplanning.He of the importance of organizational memory. Honda's
claims that coordinationwas achieved because teams have plannedintroductionof large motorcyclesexperiencedtech-
"an insatiable appetite for market intelligence," and each nical difficulties because nontraditionalmotorcycle cus-
team member"continuouslygathersand disseminatesanec- tomers tried to buy the smaller Supercubsbeing ridden by
dotal data from the marketplace"(p. 336). Imai, Nonaka, Honda's representatives.The Honda team respondedto this
and Takeuchi(1985, p. 358) also describe productdevelop- demand by improvising a new strategy to sell Supercubs
ment projects markedby a high degree of experimentation throughsports stores, in contrastto following their original
as effective when team membersare encouraged"to extract plan to focus on large motorcycles. Clearly, the effective-
as much information from the marketplace and ... to bounce ness of this improvisationdependedon the Hondateam hav-
[ideas] off other members." Following this research, we ing a rich repertoire of marketing, sales, financial
suggest that real-time informationflows can not only bring management, and technical routines that could be recom-
the "news"thatpromptsimprovisation(H2),but also replace bined into an internallyconsistent strategy that linked suc-
the coordinatingrole of a plan when actors improviseto the cessfully to the changing environment (Mintzberg 1996;
same incoming information. Mintzberget al. 1996; Pascale 1996).
In addition to providing coordination,real-time infor- Although this argumentapplies to many forms of inno-
mation flows enable actors to learn the consequences of vation, we believe it carries special strengthwith respect to
theiractions as they improvise.This immediateinformation, improvisation. In planned innovation, organizations can
in turn, enhances the chance that improvisationwill be ef- gather in advance the tools needed to implement change.
fective because it creates learning about relevant ongoing They can acquire physical resources, such as machines, as
events (Gioia 1988; Granovetter 1973, 1985). Consistent well as advice and ideas from sources outside the organiza-
with this line of reasoning, Eisenhardtand Tabrizi (1995) tion. The extemporaneousnatureof improvisation,howev-
find that repeatediterationsin the productdevelopmentcy- er, dictates thatthereis little or no space between conceiving
cle-which, they argue, provide real-time feedback-are of and executing an innovation.Thus, the improvisation-ef-
July 1998
8 / Journalof Marketing,

fectiveness relationshipwill be even more dependenton ex- riety of variables, including activities that occurredprior to
isting organizationalmemory.We predictthe following: or at the same time as the event and outcomes associated
with the action event.
H6:The greaterthe organizationalmemory,the greaterthe
likelihoodthat improvisationwill generateeffective(a) Determiningthe scope and natureof these action events
productsand(b) processesin new productdevelopment. was, therefore,an importantpartof the researcheffort. New
product development projects are composed of a series of
Method ongoing activities that are, in some sense, seamless. There-
fore, to understandand evaluate improvisation,we had to
Setting both appreciatethis largerunfoldingof activities and decide
Data were collected from two midsized firms: FastTrack,a how to divide it into a series of events that could be evalu-
developer of electronic instruments,and SeeFoods, a manu- ated andjudged independently.
facturerof food products.These companiesoffer several ad- Action events are defined as discrete activities under-
vantages for testing our hypotheses. First, both companies taken by a new product development unit. Following from
are well establishedand have formalizedstructuresbut vary our conceptualization, we restricted our sample to organi-
in size ($2.4 million and $2.6 billion in annual sales, re- zational-level actions, defined as those undertakenby the
spectively). In addition, both companies have formalized new product team. Therefore, actions taken outside of the
product development processes. Each has detailed steps group or without the group's approval were not consid-
throughwhich the developmentprocessmust go and various ered organizational for our purposes. An action event
hurdles to meet before its productsmove from step to step. could involve a team engaging in any of the following ac-
For example, SeeFoods has developed its productplanning tivities: making changes to a new product, calling a sup-
process to the degree that it considers the process a distinct plier to change the size of a part, making a decision to use
competency and treats it as a trade secret. FastTracksuc- a new distributor,releasing test procedures, finding prod-
cessfully has achieved IS09000 certificationof its product uct problems, preparing documents for regulation filing,
developmentprocesses, which indicatessome degree of for- doing a store walk, creating concept boards, making tar-
malization. These propertiesmake the firms good settings
geting decisions, generating brand names, deciding to de-
for examining improvisation.The firms also provide rich
lay a project, engaging in a focus group briefing, or
settings for testing our hypotheses because each company participatingin a focus group or a creativity session. One
takes productdevelopment seriously and links it to overall additional condition was placed on the action identifica-
organizational success. They therefore engage in various tion process: The action had to have some possibility
product development activities, which provided us with (even remote) of influencing new product outcomes. This
many opportunitiesto observe the developmentprocess. condition did not bias our sample toward actions that were
The two firms do contraston two dimensions, therefore
likely to influence outcomes. Instead, it eliminated minor
providing some variationin our study conditions. First, one activities of the team, such as decisions about when and
is a consumer packaged-goods firm, and the other a tech- where to meet and other conversation that was related to
nology-oriented industrial firm. Second, in one firm we social activities among team members and not the prod-
studied, the projectwas in the concept and prototypedevel- uct.
opment stages, whereas in the other firm, the projectwas in
the marketdevelopment and productintroductionstages of To ensure that our identification of the actions was
the new productdevelopment process. Both aspects of dif- complete and unbiased, we used several safeguards. First,
ferentiation improve the generalizabilityof our results. In four transcriptswere selected randomlyfrom each site, and
both firms, employees understood we were conducting a action events were coded independentlyby two investiga-
tors familiar with the site. This approachyielded high in-
study of productdevelopment,but no mention was made of
improvisationto avoid demandeffects about its incidence or terjudge reliability (92% agreement). Second, product
impact. In each firm, we focused on one development pro- development team meetings evolved as primary data
sources from which we derived action events because they
ject, which was selected because it was representativeof
other projectsin each firm.The teams workingon these pro- representeda consistent vehicle that broughtall the project
members together and because the meetings were the pri-
jects both were cross-functional,with ten active membersat
SeeFoods and seven at FastTrack. mary means by which members exchanged information
about actions that influenced the project.To reduce the pos-
Data Collection Procedures sible bias in team meetings, project teams were asked on
three differentoccasions to list all the things they had done
Investigatorsattended,recorded,and transcribedthe meet-
on behalf of a project during the preceding week. These
ings of the productdevelopmentteams duringa nine-month
lists were comparedwith meeting transcripts,and coverage
period. Meetings generally were held once a week. Howev-
er, holidays, vacations, and schedule conflicts meant some was adequate(90% coverage).
meetings were missed, which resulted in approximately Each of the team meetings we observedcould potential-
25-30 meetings attended in each of the firms. To generate ly producedozens of actions, making theirevaluationby re-
fine-grainedbut systematicdata at the level of specific team spondents difficult and burdensome.Therefore, to reduce
actions, we identifiedaction events from these meetings and the set of actions that would constitute our sample, we ran-
asked key informantsto evaluatethem in termsof a wide va- domly selected 2 organizationalactions from those coded in

a given meeting.4This resultedin the selection of 107 action questionnaires:50 action assessments immediately follow-
events for our sample. When an action event was selected ing the action and 50 short-termimpact forms four weeks
for informantevaluation, we set in motion a series of data after the action. Therefore, they completed two question-
collection events. To begin, informantscompletedan action naires each week for the study. Given this workload, we
assessment form, within a week of the action occurring,that tried to make each questionnaireno more than a page long.
asked them to rate the action and variousenvironmentaland This meant sacrificing some depth on individualmeasures.
organizationalactivities that occurred prior to and during We focused our efforts on developing multi-item measures
the time the action was unfolding. This included an infor- for those constructsthat were considereda prioridifficult to
mant rating of organizationalimprovisation,memory, and measure,such as improvisation.
real-time informationflows and an assessment of the level Because the composition of formative measures is dri-
of environmental turbulence associated with the action ven by conceptual criteria-which is coverage of the con-
event. After approximatelyfour weeks, the same informants struct domain-and not by predictions of correlation
evaluated the short-term impact of each action event in between items in that space, formative measures were not
terms of productand process outcomes. The key informants subject to reliabilityor factor analytic approaches(Bagozzi
in completing both forms were the team leaders because 1994). The remaining reflective measures were examined
they were in attendance at every meeting, were aware of for unidimensionalityand reliability.In Table 2, we present
team actions, and had a broadview of the project. an overview of the psychometrics associated with each
measure.
Measures
Appendix B contains a complete listing of all the mea-
In this section, we reportthe propertiesof our measures,ap- sures used in this study. The organizationalimprovisation
proximatelyhalf of which were multi-itemand half single- level of each action event was evaluated on the action
item measures. Multi-item measures were a mix of assessment form and was measured on three semantic
formativeand reflective indicators.This mix was adoptedto differentialseven-point scales with the following anchors:
safeguard against the hazards of key informant burnout. (1) figured out action as we went along/action followed
Each of our key informantscompleted approximately100 a strict plan as it was taken, (2) improvised in carryingout
this action/strictly followed our plan in carrying out this
action, and (3) ad-libbed action/not an ad-libbed action.
40ur initialgoal was to samplesystematically highandlow im- The mean improvisationlevel is M = 4.252 (s.d. = 1.985)
provisationactionsto ensurethatoursamplewouldcontainvari- and the coefficient alpha exceeded acceptable standards
ance. However,afterattempting to do this for severalweeks,we
abandoned the approachfor two reasons.First,fromourobserva- (a = .79).
tions of team meetingsand interviewswith key informants,it Because of the centralityof improvisationto our work,
seemedthatmuchof whatoccurredduringmeetingswas,at least two additionalsafeguardswere taken to measure it. First, it
in part,animprovisation. Therefore,variancewouldbe easierthan was ratedby two investigatorsinvolved in the site, and there
we hadexpectedto capture.Second,we determined it was more was 70% agreementregardingwhether improvisationwas
appropriate for our informants to assessthe degreeto whichthey low (1), moderate (4), or high (7). The mean investigator
improvisedthanforus to makesuchjudgmentswithouta clearun-
derstanding, a priori,of howmuchor how littleplanninghadpre- rating(M = 4.014, s.d. = 1.539) also comparedwell with the
ceded an action. This realizationshifted our design, and we mean informantrating. Second, because we claim that im-
allowedinformants to ratethedegreeof improvisation in anaction provisation is distinct from innovation, we examined the
followingits enactment. discriminantvalidity of the level of improvisationand the
TABLE 2
Measurement Information
Standard
Mean Deviation (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
(1) Organizationalimprovisation 4.25 1.98 .83

(2) Environmental turbulence 2.12 1.22 .18 -
(3) Organizationalreal-time
informationflows 5.86 1.85 .03 -.14 -b
(4) Organizationalmemory 4.36 1.78 -.60 -.07 .04 .84
(5) Design effectiveness 5.03 1.06 -.08 -.23 .30 .07 a
(6) Marketeffectiveness 4.87 .81 -.22 -.12 .30 .29 .65 .70
(7) Cost efficiency 4.05 .76 -.08 -.20 .06 .08 .19 .11 b
(8) Time efficiency 4.08 .89 .01 -.14 .20 .03 .05 .03 .35 .89
(9) Team functioning 4.61 .80 -.34 -.12 .20 .17 .43 .60 .11 .16 .86
(10) Team learning 5.12 .93 -.30 -.24 .29 .25 .53 .64 .17 .13 .60 .73
aFormative scale,thereforealphais notreported.
bSingle-itemmeasure,thereforeno alphaorcorrelation is reported.
Notes:Thecoefficientalphaforeach measureis on thediagonalandthe intercorrelations
amongthe measuresareon the off-diagonal.
Corre-
p < .05.
lationcoefficientsgreaterthan.20 aresignificant,
101 Journalof Marketing,July 1998

level of innovationof the focal action. Informantsrated in- or to the fixed coefficient (AX2(I)= 52.00), which indicat-
novation and improvisationon the same form. Innovative- ed discriminant validity between organizational memory
ness was measured by asking key informantsto rate the and improvisation.
level of innovationon a two-item semanticdifferentialscale
with (1) innovativeaction/ordinaryaction and (2) novel ac- Dependent variables. These variables were assessed by
our informantsfour weeks after the action event occurred.
tion/standardaction as the anchors.The two innovativeness
Informantswere given a one-page survey with the action
items correlatedwell (p = .70).
event described in detail at the top of the page. In addition,
The test of discriminantvalidity between innovationand
because of the time lag, a portionof the transcriptrelevant
improvisationrequiredconstrainingand freeing the phi co- to the action generallywas attachedto the page to jar the in-
efficient between the two measures using LISREL 8. The
formant'smemoryof the event. Informantsdid not have ac-
model with the free coefficient was found to be superiorto
cess to their originalevaluationsof the event or their ratings
the fixed coefficient (AX2i)= 5.31), exceeding the standard
of the informationalactions occurringaroundit. All depen-
necessary to show discriminant validity (AX2(l)= 3.84). dent variableswere evaluatedon a seven-point scale, where
Therefore, innovation and improvisation are empirically 7 was a positive and I was a negative effect of the action on
distinct.5
the particularoutcome. Informantswere asked to rate how,
The three explanatoryvariableswere measuredon sev- "on balance,"the action event has or is likely to have influ-
en-pointLikertscales in which the anchorsdependedon the enced each dependent variable. They appearedto have no
variable.The degree of environmentalturbulenceoccurring
around the action event was measured using a three-item, problems in making such assessments.
formative measure, where I was "none"and 7 was "a lot" Following our conceptualization,two product-effective-
ness dependentmeasureswere used in this research.Design
for the following items: When the action was taken, how
effectiveness was measuredwith a two-item formativescale
would you ratethe level of change [definedas any deviation that assessed the impact of the action on the performance
from the statusquo] within (1) your team, (2) your firm, and and design of the product.The two items have an acceptable
(3) externalsources [customers,suppliers,distributors].The correlation (p = .48). Influenced by Griffin and Page's
mean level of externalchange was 2.136 (s.d. = 1.243). Be-
(1993, 1996) work, we measuredmarketeffectiveness with
cause the measure is a formative indicator,ratherthan re- a three-itemscale that describedthe impact of the action on
flective, we do not reporta reliabilitycoefficient. the sales, customeracceptance,and success of the new prod-
Informantsalso evaluatedthe level of real-timeorgani- uct. The items have an alpha of .70.
zational informationflows that occurredusing a single-item Four process effectiveness dependent variables also
measure that asked for a ratingof the level of face-to-face, were evaluated. Cost efficiency was measured by a single-
telephone, or e-mail information transferredamong team item measure of the estimated cost structure of the new
members just before the focal action (M = 4.078, s.d. =
product(Griffinand Page 1993, 1996). Time efficiency was
1.058), using the same anchors. measuredby a three-itemmeasurethatasked respondentsto
Organizationalmemory level was measured by asking rate the (1) length of the productdevelopment process, (2)
informantsto evaluatethe memorylevel regardingan action
speed of the productdevelopment process, and (3) project
using a four-item scale adapted from our previous (1997) timeliness. These items are reliable (a = .89).
study. Informants were asked to rate the extent of their Team functioningrefersto the impactof the action event
agreement with the following items: For this action, my on the degree to which the team works well together.This
team has (1) well-defined procedures, (2) a standardap- was evaluated using a three-itemmeasure that asked infor-
proach,(3) a great deal of knowledge, and (4) strong skills. mants to assess the impact of the action on (1) team com-
The measure had a mean of 4.392 (s.d. = 1.784) and ade- mitment level, (2) team functioning, and (3) team
quate reliability(a = .79). enthusiasm.These items are reliable(a = .86). Finally, team
Of all the independentvariables, organizationalimpro-
learning was measured by asking informantsto assess the
visation and memory are the only measures that are multi-
impactof the action on (1) the way the team thinks aboutthe
item and reflective in nature.Therefore, we examined the
project, (2) the team's certaintylevel, (3) the team's under-
discriminantvalidity of organizationalmemory and impro-
standing level, and (4) how much the team learned. These
visation. As previously, the test of discriminantvalidity in- four items are reliable (a = .73).
volved constraining and freeing the phi coefficient Three of the process effectiveness outcomes (time effi-
between the two measures using LISREL 8. The model
ciency, team functioning,and team learning)are multi-item
with the free coefficient again was found to be far superi- and reflective in nature.Therefore, measure development
required examining their discriminantvalidity. As we did
previously, we established a base model that did not reflect
5Weideallywouldhaveperformed discriminantvaliditychecks the correlationbetween measures. Then we examined how
betweenall the variablesfromwhichwe previouslyconceptually model fit changed when we constrainedthe phi coefficient
distinguishedimprovisation (includingadaptation,learning,and between different pairs of the three measures to equal one.
opportunism).However,becauseof thesevereconstraints imposed Results indicate that the model with the free phi coefficient
on ourinformants,we choseto focusourempiricaleffortson in-
novation.We believe it is closest conceptuallyto improvisation was a better fit in all three cases, which indicateddiscrimi-
and,therefore,perhapsthecompetingvariablemostcrucialto our nant validity in time efficiency and team learning (AX2(I)=
perspective. 16.53), time efficiency and team functioning (AX2(I)=

28.17), and team learning and team functioning (AX2() = fy. The central tendency then was toward improvisation.
8.78). However, with a standarddeviation of 1.98, there was also
quite a bit of variance in improvised behavior. In Table 3,
Model-Testing Approaches PartA, we depict the frequency distributionof improvisa-
We used three distinctmodel testing approachesto examine tion in our sample.
the proposed hypotheses. First, we used simple descriptive
statistics to examine the existence of improvisationin the Factors Influencing the Incidence of
new productactions we sampled. Second, we used a multi- Improvisation
variate linear regression model to examine the impact of Considering the factors that could influence the incidence
environmentalturbulence,organizationalreal-timeinforma- of improvisation, we tested the first three hypotheses in a
tion flows, and organizationalmemory on the incidence of single multivariate regression model. The results, which
improvisationin new productactions. Third, we performed appearin Table 3, Part B, suggest that the overall model is
a split group analysis (Arnold 1982; Cohen and Cohen significant (Adjusted R2 = .39, F(388)= 20.97, p < .001).
1983) to examine the impactof the threemoderators-envi- Results also indicate that environmental turbulence is a
ronmentalturbulence,organizationalreal-time information marginally significant, positive predictor of the level of
flows, and organizationalmemory-on the organizational improvisation in new product actions (b = .252, t = 1.88,
improvisation-effectivenessrelationships.This approachin- p < .10, two-tailed test), in support of H1. Organizational
volved creatinghigh and low levels of each moderatorvari- real-time information flows do not have a significant im-
able by performinga median split. We then examined the pact on the incidence of organizationalimprovisation (b =
relationship between improvisation and the various out- .100, t = 1.14, p > .10), thus failing to support H2. Finally,
comes in the high and low moderatorvariable conditions organizationalmemory has a negative effect on the level
and compared the regression results from these two condi- of improvisation (b = -.687, t = -7.520, p < .001), in sup-
tions using a t-test, to determine if the regression coeffi- port of H3.
cients were different across the two moderatorconditions
(Pedhazur1982). If the t-test of differencesin the beta coef- Factors Influencing the Effectiveness of
ficients was significant, we had found evidence of modera- Improvisation
tion and inspected the direction of moderation.We chose
The split group analyses results appear in Table 3, Part C.
this approach over moderatorregression analysis (MRA)
We note there that high levels of environmentalturbulence
because MRA demandsthat all main and interactioneffects
have a positive influence on improvisation'simpact on de-
associated with the proposedmoderatinginfluences be en-
tered into the model (Pedhazur1982).6 In this study, MRA sign effectiveness (t(86)= 2.83, p < .05). We find that when
environmentalturbulenceis low, improvisationhas a nega-
would have involved seven predictors(i.e., improvisation,
tive effect on design effectiveness, but in the presence of
memory, real-time information,environmentalturbulence,
high environmentalturbulence,improvisationimprovesde-
improvisation x memory, improvisationx real-time infor-
mation, and improvisationx environmentalturbulence).Us- sign effectiveness. Environmentalturbulencedoes not have,
however, a statistically significant moderating impact on
ing MRA with the numberof variablesand the sample size market effectiveness (t(86) = 1.21, p > .10). These results
of this study likely would have resulted in underpowered
tests of the hypotheses. thereby support H4a with respect to technical design and
qualityoutcomes but not in terms of the product'seffective-
ness using marketindicators.
Results Environmentalturbulencehas equally mixed effects on
The Incidence of Improvisation the improvisation-process outcome relationships. Turbu-
lence improves the extent to which the team reports it
Simple descriptive statistics suggest thatorganizationalim- learned (t(86)= 3.42, p < .05) and functioned smoothly
provisation occurredin our sample of new productactions. (t(86)= 1.98, p < .05) while taking improvisationalactions,
The mean level of improvisation,as rated by informants, in support of H4b. However, the improvisation-cost effi-
was 4.252 on a seven-point Likert scale (s.d. = 1.985),
ciency relationship becomes weaker and more negative
where 7 representsgreaterimprovisation.The scale exhib- when turbulence is high, and the improvisation-time effi-
ited considerable range,runningfrom I to 7 with a mode of
ciency relationship is not influenced at all, thus failing to
5 and a median of 4.667. The distribution is fairly even
supportH4b.These results suggest importanttrade-offs for
across all levels of improvisationbut is skewed slightly to- the use of improvisation in conditions of environmental
ward higher levels of improvisation.For example, if we de- turbulence.
fined as "primarilyimprovisational"those actions that were
Organizationalreal-time informationflows have a more
rated higher than five, 47.5% of the actions would qualify; uniform positive influence on the extent to which impro-
if we used a cutoff of higher than six, 24.1% would quali- vised new productactions influence design (t(94)= 6.21, p <
.05) and market (t(94)= 4.76, p < .05) effectiveness, in sup-
port of H5a. However, real-time informationflows do not
MRA
level,a univariate have the same positive influence on process outcomes, thus
6Despitedifferencesat themultivariate
test is identicalin structureto the split-groupanalysisused here failing to supportH5b.Improvised new productactions do
(forrelatedproofs,see Arnold1982;CohenandCohen1983). not have a greaterimpact on cost efficiency (t(94)= .66, p >
July 1998

TABLE 3
Tests of Hypothesized Relationships
A.The Incidence of Organizational Improvisation

Mean: 4.25
StandardDeviation: 1.98
Mode: 5.00
Median: 4.67
Frequency Percentage
1.00 14 13.1%
1.01-1.99 4 3.8
2.00-2.99 7 6.5
3.00-3.99 18 16.8
4.00-4.99 13 12.1
5.00-5.99 25 23.4
6.00-6.99 9 8.2
7.00 17 15.9
107 100 %
B. Factors Influencing the Incidence of Organizational Improvisation

Overall model, F(3,88)= 20.97, p < .001, Adjusted R2 = .39
Independent Variables
Environmental Real-Time Information Organizational
Turbulence Level Flow Level Memory Level
Dependent Variable b t-value b t-value b t-value

Improvisation .252 (1.88**) .100 (1.14) -.687 (-7.52*)
C. Factors Influencing the Effectiveness of Organizational Improvisation
ModeratingVariables
Environmental Real-Time Information Organizational
Turbulence Level Flow Level Memory Level
(n = 86) (n = 95) (n = 94)
Dependent Variables Lowa High t-value Low High t-value Low High t-value
Product effectiveness
Design effectiveness -.17 .17 2.83* -.57 -.01 6.21* -.13 .08 1.65**
Marketeffectiveness -.27 -.16 1.21 -.54 -.18 4.76* -.13 -.06 .64
Process effectiveness
Cost efficiency .02 -.22 -2.91* -.14 .09 .66 -.18 .09 2.83*
Time efficiency .07 .09 .16 .09 .02 -1.46 -.18 .17 3.15*
Teamfunctioning -.36 -.20 1.98* -.40 -.39 .10 -.41 -.23 1.98*
Team learning -.35 .00 3.42* -.59 -.28 3.81* -.32 -.11 1.98*
aNumbers inthe lowandhighcolumnsrepresentthestandardized on each dependentvari-
betacoefficient(b)forthe impactof improvisation
ableunderlowandhighmoderating conditions.
*p< .05.
**p<.10.
.10), time efficiency (t(94)= -1.46, p > .10), or team func- ious process outcomes, including cost efficiency (t(94)=
tioning (t(94)= .10, p > .10) when real-time information 2.83, p < .05), time efficiency (t(94)= 1.98, p < .05), team
=
flows are high ratherthanlow (see Table3, PartC). Only the functioning (t(94)= 3.15, p < .05), and team learning (t(94)
impact of improvisationon team learning improves when 1.98, p < .05). These results supportH6b.Likewise, organi-
real-timeinformationflows are high (t(94)= 3.81, p < .05). zational memory marginally improves the extent to which
Finally, H6 predicts that high levels of organizational improvised new product actions result in design effective-
memory will increase the likelihood that improvisationwill ness (t(94)= 1.65, p < .10). However, memory does not im-
generate effective products and processes in new product prove the likelihood that improvised new product actions
development.The results indicate that organizationalmem- will result in marketeffectiveness (t(94)= .64, p > .10), thus
ory uniformlyimprovesthe impactof improvisationon var- providing mixed supportfor H6a.
Improvisationin New ProductDevelopment113

Discussion and Implications enues for such a framework.For example, the role of cul-
Our conceptual work builds on prior interdisciplinaryre- ture, structure,boundaryspanners, and productchampions
as catalysts for and facilitators of improvisationwould be
searchand suggests that improvisationcan play a role in the
new productdevelopmentprocess.An investigationof mar- appropriatefactors. Additional researchalso could involve
examining the role of individualimprovisationin organiza-
keting literatureprovidesfew positive empiricalaccountsof tional improvisation,a factor we did not investigate.
extemporaneousaction in managerialaction, suggesting in-
stead thatmore fruitfulaction is plannedand then executed. Third, though this researchinvolved examining the im-
In an attemptto addressthis gap in the literature,we have pact of improvisationon short-termnew producteffective-
documentedthe incidence of improvisationand the factors ness, it also would be valuable to examine the impact of
that influence that incidence. Drawing on equivocalities in improvisationon long-term organizationaloutcomes. Our
researchalso was limited to new productdevelopment ac-
literature,we further suggest that improvisationis under-
stood best as perhapshaving both positive and negative out- tions. However,many othermarketingcontexts are also rel-
comes for firms. This mixed assessment draws us to try to evant contexts for the study of improvisation.Advertising
understandthe conditions in which improvisationmight be and personal selling stand out as areas in which we would
deployed effectively by organizations.We propose and test expect improvisationto occur at even higher levels. Addi-
several such conditions. tional research could examine the incidence of improvisa-
In this section, we addressthe theoreticaland practical tion and attemptto demonstratethe generalizabilityof our
potentialof our view of improvisationin several ways. First, findings across multiple contexts.
we discuss the limitationsof our work. Second, we review Finally, furtherresearchcould addressthe possibility of
the patternof our results in more detail and with an eye to- common methods varianceinfluencing our results, because
ward understandingthe conditions in which the moderators the same informantsrated aspects of both actions and out-
comes. However, because observers independently identi-
(I) change improvisation'seffect from negative to positive
or (2) reduce the negative impact of improvisation.Third, fied the action, the ratings of actions and outcomes were
we discuss several trade-offsin managingthe incidence and accomplished at significantly different times, and infor-
effectiveness of improvisationin organizationsand high- mants had no recordof priorratings when performingtheir
outcome ratings, the chance of informant preconceptions
light the implicationsof these trade-offs.
producingthe resultshere is reduced.Furtherresearchcould
Limitations use multiple measures of actions and outcomes to ensure
Our researchhas several limitations.First,despite the longi- even furtherthe lack of common methodsor informantbias.
tudinalapproachwe adopted,the use of two firms limits the
Pattern of Improvisation Results
generalizabilityof our results.Similarto Hutt,Reingen, and
Ronchetto(1988), we initially sought to controlfor many of Considerable research on organizations suggests that for-
the firm factors that might influence either the rate of im- malized organizationswith well-developed product devel-
provisationor its impactby limiting ourselves to two firms. opment procedures are relatively unlikely to engage in
Furthermore,the challenges of longitudinalaccess to orga- improvisation(Scott 1987). Contraryto that research, our
nizations, especially to proprietaryactivities such as new first basic finding is that,even in two well-establishedorga-
product development processes, made our method choice nizations with formal structures,roles, and procedures,im-
more reasonable.This is particularlytrue given that we se- provisationoccurs with substantialregularityin the product
lected actions from among those thatoccurredduringweek- developmentprocess.
ly product development meetings, all of which were Prior research also has tended to highlight either the
attended,recorded, and transcribed.Other approachesthat dangers of improvisationor its potential for helping firms
may be more externallyvalid might have createdotherprob- adapt. Our pattern of results supports a more contingent
lems in generating an unbiased sample of actions. There- view of improvisation.For example, we provide some sup-
fore, our approachoffers solid, internallyvalid evidence of portfor the traditionalconcern regardingthe risks of impro-
improvisationthat furtherresearchmight examine in more visation because, in five of the specific outcome variables
firms using a less sensitive methodology. we observed, our moderatingvariableshad a positive effect
Second, our hypotheses focus on three informationfac- but workedby reducingthe negativeeffect of improvisation.
tors that have been discussed in priorliteratureon improvi- Thus, though the moderatingconditions enhancedthe value
sation and that deserve empirical attention.Although still of improvisation,as we had expected, the conditions were
limited, these factors include the effect of differentinforma- not strong enough to make the net effect of improvisation
tion sources (internal and external) and types (stocks and positive.
flows) on improvisation. Further research could involve In other cases, however, the moderatingconditions re-
considering a more comprehensivestudy of improvisation versed the negative impact of improvisation.For example,
that extends the connections between informationand im- in conditions with low organizationalmemory, improvisa-
provisationpresentedhere. This approachcould develop a tion had a negative effect on design effectiveness, cost effi-
more general framework of the antecedents and conse- ciency, and time efficiency. However, in the presence of
quences of improvisationin new productdevelopment, us- high organizationalmemory, improvisationhad a positive
ing relevant industry, firm, product, environmental, and effect on these outcomes. These findings supportthe gener-
individual team factors. Many of the factors examined by al argumentthatemergentprocesses might have value in un-
Hutt, Reingen, and Ronchetto (1988) would be fruitful av- certain or ambiguous conditions (Burgelman 1983; Miner
July 1998
14/ Journalof Marketing,

1987). They also support the more recent arguments of zone of real-time information flow in which sufficient
scholars who claim that improvisationrepresentsan impor- amounts of real-time informationare needed to promote a
tant competency thatcan producevalue for organizationsin positive relationshipbetween improvisationand productef-
certain conditions (Eisenhardtand Tabrizi 1995; Moorman fectiveness. However, flows cannot be so high as to create
and Miner 1998; Weick 1987). negative improvisation-processeffectiveness relationships,
which, over time, might undermine product effectiveness
Trade-Offs in Improvisation levels.
A common theme that runs throughour conceptual frame- Finally, as with the other informationalmoderators,the
work and resultsis thatimprovisation,similarto most strate- influx of information about environmentalchange brings
with it certain trade-offs. In particular,our results suggest
gic actions, involves trade-offs and potential synergies for
that high levels of informationaboutenvironmentalchanges
marketingorganizations.This theme is evidenced in several
ways. during improvisationresult in increased productdesign ef-
Recall that our findings show that memory reduces the fectiveness. Across all of the informational moderators,
likelihood of improvisation,but it also increases the effec- productdesign effectiveness has the most to gain from high-
tiveness of improvisationwhen improvisationdoes occur. ly improvisationalactions when levels of informationare
Therefore, the same organizationalfeature that makes im- high. However, consistent with the other results, firms have
to accept that the influx of the high levels of environmental
provisationeffective is likely to reducethe chances of its oc-
currence. Too powerful a memory, then, can remove change informationmight have correspondingrisks, partic-
improvisation from the organization'srepertoire,whereas ularly higher costs. Such trade-offsalso appearin other re-
too little memory can renderthe improvisationthat occurs search on new products(Griffin and Page 1993; Moorman
ineffective. This result suggests that there is a thresholdof 1995).
In summary,these results suggest that improvisationis a
memory level at which improvisationis a valuable organi-
zational activity, with levels below or above this threshold strategyof emergentlearning (Mintzberg1996) that can be
reducing the chances of such an impact. This trade-off im- employed as a substitutefor planning (Weick 1987). How-
plies that organizationsmust minimize the fixating aspects ever, our results clearly suggest that improvisation is not
of memory when improvisationis needed and evoke memo- necessarily a free good, nor is it one that translatesinto ef-
fective outcomes in all conditions. On the contrary,our re-
ry as improvisationis unfolding if it is to be effective. This
delicate balance of restrainingand infusing memory at cer- sults consistently emphasize that improvisation must be
tain times requires a greater understandingof memory, its directedexplicitly, its trade-offsand tensions acknowledged
forms, and the degree to which these forms restrictextem- and managed,and the conditions in which it is effective un-
derstoodand nurturedby organizations.
poraneousactions in organizations.For example, in a forth-
coming article (1998), we distinguishbetween the effects of
declarative organizationalmemory (facts and theories) on Future Research Issues
the novelty of organizationaloutcomes versus the effects of There are many issues that our initial inquiryinto improvi-
proceduralorganizationalmemory (skills and routines) on sation did not consider.We discuss several here as a way of
the timeliness of organizationaloutcomes. establishing a strategyfor additionalresearchon this topic.
Another trade-off associated with memory that is evi- Regardingimprovisationgenerally,we recommendthat fur-
dent in our results is that organizationalmemory facilitates ther research consider whether improvisationis driven by
the impact of improvisationon all new productand process firm mismanagement,environmentalchange, or the deci-
outcomes, except those associated with externalmarketef- sion to use improvisationpurposivelyas partof firm strate-
fectiveness. Although this is consistent with research that gy. In the domain of product development, we encourage
suggests an internal firm focus should reduce market suc- investigation of improvisation'soccurrence and impact in
cess (Day and Nedungadi 1994; Deshpande, Farley, and different (1) project phases, (2) productcategories, and (3)
Webster1993), none of our memorymeasuresinvolve skills industries. Whether a product development project repre-
and knowledge regardinghow a new product activity fits sents an incrementalor a radical change from a prior prod-
with customer needs and wants. Therefore, the value of uct is an especially importantcontingent worthy of further
memoryappearsto be linked tightly to its measuredcontent. research.The risks for improvisationintuitively seem high-
Real-time informationflows moving throughorganiza- er in radical product development projects because of the
tions presenta differentset of trade-offsfor firms. Real-time probablelack of relevant organizationalmemory to inform
flows were found to increasethe extent to which improvisa- the productdevelopmentprocess. At the same time, impro-
tion produces effective new productsbut to reduce the pos- visation may be more likely to occur in radical productde-
itive effect of improvisation on process outcomes velopment projectsthat lack memory.Teasing out these and
(excluding learning).Improvisationswere less cost and time other possibilities is an importantnext step.
efficient, and groups engaging in them appearedto function Another topic we did not address is the nature of the
less effectively when the level of real-time information group or individualfactors that spawn improvisationduring
transferwas high. Despite the inefficiencies associated with the new product development process. The research tradi-
real-time information flows, improvisation continued to tion on productchampioningincludes projectsthat are initi-
promote design and market effectiveness when real-time ated outside the formal new productdevelopment process
flows were high. Therefore, similar to our prior discussion (Burgelman 1983; Hutt, Reingen, and Ronchetto 1988). It
of memory, our results appear to recommend a restricted would be interestingto consider what motivates these infor-

mal improvisationefforts. What form does improvisation flux of informationabout environmental turbulence, so as
take when the focus is on a majorstrategicdecision that in- to promote effective new product development outcomes
volves multiple stakeholdersacross business units?The fo- when improvisation occurs. The management of such re-
cus of the presentstudy is restrictedto improvisationin the sources requires attention to the trade-offs and synergies
narrowconfines of a structureddevelopmentprocess. between improvisation,organizationalfactors, and the en-
Our focus was the impact of improvisationalactivities vironmentalcontext.
on the projectin which they occurred.However, it is possi-
ble that improvisations have long-term impacts as well.
Team memberssometimes observedthat a particularimpro- Appendix A
visational action not only worked in the currentproject but An Example of Organizational Improvisation
also could be used in other settings or futureprojects. Our To make concrete the distinction between improvisational
quantitativeresults supportsuch an impactand suggest that and nonimprovisationalactivities, consider the contrastbe-
improvisationpositively affected team learningoutcomes in tween two productdevelopment processes in a single orga-
high information conditions. Additional research could nization. For ProductA, the organizationfollows its typical
fruitfullyinvestigatewhetherimprovisationserves as a sys- productdevelopmentprocedure.The marketingdepartment
tematic form of unplanned experimentation in organiza- analyzes marketpotential, pricing questions, and details of
tions. If this function is confirmed, the potential customer demand for a new instrumentthat is based on an
"second-order"impact creates an additional factor in the
emerging technology.The engineering departmentanalyzes
calculus of improvisation'svalue to organizations.Each im- technical problems with prior products, the feasibility of
provisationmight have, on average,a low expected value as meeting quality specifications at certain price levels, the
a possible new routinefor the organization.But on rareoc-
availability of key components at particularprices, and the
casions, an improvisational act (or "local experiment") time needed to produce the new product. On the basis of
might representa real improvementover priorpracticesand these analyses, the senior officers approve a plan, budget,
thus be a very useful experiment (Miner, Moorman, and and timeline for this product'sdevelopment, manufacturing,
Bassoff 1997). and launch. A productteam is appointed to implementde-
Finally, our qualitative observations lead us to suspect tailed design and prototyping activities. After being
that, in additionto building a baseline model of factors that checked, final manufacturingspecifications and procedures
can move improvisationfrom a hindranceto a potentialad- are approvedand triggerimplementationof the manufactur-
vantage,contemporaryresearchersshould entertainthe pos- ing process and productintroduction.
sibility that the boundary conditions for organizational ProductB follows a differentpath.A customerof one of
improvisationmightbe changing.Corporateintranets,com- the firm's currentscientific testing instrumentscomplains to
puter-aideddesign, and manufacturingand point-of-saleda- members of the original design-and-supportteam that he
ta can change the temporal links between actions in ways
needs an instrumentto assess certain features of selected
that previously were not possible. In many ways, this
opaque liquids, instead of those of the clear liquids for
change enhances the potential for accurateplanning;how- which the instrumentwas designed. Team members meet
ever, it also might enhancethe possibility of fusing planning and think of a recent scientific advance that may make it
and acting. Therefore,it appearsthatimprovisation'sbound-
possible to investigate necessary materials. Using time be-
ary conditionsare changingeven as we begin to examine the tween otherprojectsto which they officially were assigned,
phenomenonin a systematic way. two team members pull together a new product in three
months that uses parts from old products and a few new
Conclusion parts they had ordered.They build the new machine them-
This articleexamines the incidence and effectiveness of im- selves as they progress.At the end of the processes, they sell
the custom-designedmachine to the customer.
provisation during the new product development process.
In ProductA, planning formally preceded implementa-
Hypotheses were developed examining the impact of vari-
ous types and sources of informationon the level and effec- tion in both design and manufacturingactivities. Details of
tiveness of improvisation. We find that improvisation is both product features (price, performance specifications,
prevalent and occurs when organizationalmemory is low components,and potentialsources for them) and the product
but environmentalturbulence is high. Our results support development process (team members, responsibilities,
traditional concerns that improvisation can reduce new checkpoints, and intermediatetarget dates) were specified
product effectiveness but also indicate that informational before they were implemented.Manufacturingplans, speci-
factors emanating from the environmentand organization fied in advance, detailed aspects of production outcomes
can reduce these negative effects or even create a positive and procedures.Clearly, ProductB representeda more im-
effect of improvisationon new productoutcomes. provisational activity. A broad goal, but little product or
These results suggest that there are conditions in which manufacturingplanning,precededdevelopment. Final tech-
improvisation might be not only what organizations do nical performance levels, exact components, size, shape,
practice but also what they should practice to flourish. We and actualassembly all unfoldedas the team progressed.Al-
suggest that these conditions involve the careful deploy- though small improvisations to solve unexpected design
ment and management of other organizational resources, problemsarose duringdevelopmentof ProductA, a dramat-
such as memory, real-time informationflows, and the in- ically greaterproportionof both the product and manufac-

turing design occurred directly during their implementation Product Effectiveness Outcomes
for Product B, indicating improvisation. Seven-point Likert scale, where I is a negative effect
and 7 is a positive effect.
This action has or is likely to have the following effect on:
Appendix B a. Design effectiveness. New Measure
Measures
*Productdesign
Organizational Improvisation New Measure
*Productperformance
Seven-point semantic differential scale.
Rate the action: b. Market Effectiveness. Driven, in part,
*Figuredout action as we went along/Action followed a strict Griffin and
y Page (1993,1996)
plan as it was taken. *Productsales
*Improvisedin carryingout this action/Strictlyfollowed our *Customeracceptance
plan in carryingout this action.
*Generalsuccess of the product
*Ad-libbedaction/Not an ad-libbedaction.
Process Effectiveness Outcomes
Environmental Turbulence New Measure
Seven-point Likert scale, where 1 is a negative effect
Seven-point Likert scale, where 1 is none and 7 is a lot. and 7 is a positive effect.
Whenthe action was taken,how would you rate the lev-
This action has or is likely to have the following effect on:
el of change in the following areas? (change is defined as
a. Cost efficiency. Driven, in part,
any deviationfrom the status quo):
*Withinyour team. by Griffin and Page (1993,1996)
*Withinyour firm. *Productcosts
*Withinexternal sources (customers,suppliers,distributors). b. Time efficiency. Moorman (1995)
Organizational Real-time Flows New Measure *Lengthof projectdevelopmentprocess
Seven-point Likert scale, where I is none and 7 is a lot. *Speedof productdevelopmentprocess
Of information receivedfrom this source, how much was *Projecttimeliness
face-to-face, phone, or e-mail:
*Teammembers. c. Team functioning. New Measure
*Teamcommitmentlevel
Organizational Memory Adapted from *Teamfunctioning
Moorman and Miner (1997)
?Teamenthusiasm
Seven-point Likert scale, where 1 is disagree and 7 is
agree. d. Team learning. New Measure
For this action, my team has: *The way we think about the project
*well-definedprocedures. *Ourcertaintylevel
*a standardapproach. ?Ourunderstandinglevel
?a great deal of knowledge. *How much we have learned
?strongskills.
REFERENCES
Aaker, David A. (1988), Strategic Market Management, 2d ed. Bailey, Derek (1980), Improvisation:Its Nature and Practice in
New York:John Wiley & Sons. Music. Ashbourne, Derbyshire, Great Britain: Moorlan Pub-
Anderson, Paul F (1982), "Marketing,StrategicPlanning,and the lishing Co. Ltd.
Theory of the Firm,"Journalof Marketing,46 (Spring), 15-26. Barley, Stephen R. (1986), "Technologyas an Occasion for Struc-
Andrews, Jonlee and Daniel C. Smith (1996), "In Search of the turing:Evidence from Observationsof CT Scannersand the So-
cial Orderof Radiology Departments,"AdministrativeScience
Marketing Imagination: Factors Affecting the Creativity of
MarketingProgramsfor MatureProducts,"Journal of Market- Quarterly,31 (March),78-108.
Bastien, David T and Todd J. Hostager(1988), "Jazzas a Process
ing Research, 33 (May), 174-87. of OrganizationalInnovation,"CommunicationResearch, 15
Argyris, Chris and Donald Schon, (1978), Organizational
(5), 582-602.
Learning: A Theory of Action Perspective. Reading, MA: and (1992), "Cooperationas CommunicativeAc-
Addison-Wesley.
complishment:A Symbolic InteractionAnalysis of an Impro-
Armstrong, J. Scott (1982), "The Value of Formal Planning for vised Jazz Concert," CommunicationStudies, 43 (Summer),
StrategicDecisions: Review of EmpiricalResearch,"Strategic 92-104.
ManagementJournal, 3 (July/September),197-211. Berliner,Paul E (1994), Thinkingin Jazz: The InfiniteArt of Im-
Arnold, Hugh H. (1982), "ModeratorVariables:A Clarificationof provisation. Chicago: Universityof Chicago Press.
Conceptual,Analytic, and PsychometricIssues,"Organization- Bjurwill, Christer(1993), "Readand React: The Football Formu-
al Behavior and HumanPerformance,29 (April), 43-175. la," Perceptualand Motor Skills, 76 (June), 1383-86.
Bagozzi, RichardP (1994), "StructuralEquationModels in Mar- Borko, Hilda and Carol Livingston (1989), "Cognitionand Impro-
keting Research: Basic Principles,"in Principles of Marketing visation: Differences in MathematicsInstructionby Expert and
Research, RichardP. Bagozzi, ed. Cambridge,MA: Blackwell, Novice Teachers,"AmericanEducationalResearchJournal, 26
317-84. (Winter),473-98.

Boyd, HarperW., Jr. and Orville C. WalkerJr. (1990), Marketing tiveness in Japanese Firms: A QuadradAnalysis," Journal of
Management:A StrategicApproach. Homewood, IL: Richard Marketing,52 (January),23-36.
D. Irwin Inc. and FrederickE. WebsterJr. (1989), "OrganizationalCul-
,and Jean-Claude Larreche (1998), Marketing tureand Marketing:Defining the ResearchAgenda,"Journalof
Management:A StrategicApproachwith a Global Orientation, Marketing,53 (January),3-15.
3d ed. Boston: Irwin/McGraw-Hill. Dickson, Peter R. (1997), Marketing Management, 2d ed. New
Brown, Shona L. and Kathleen M. Eisenhardt(1995), "Product York:The Dryden Press.
Development:Past Research,PresentFindings, and FutureDi- Dougherty, Deborah (1992), "InterpretiveBarriersto Successful
rections," Academy of Management Review, 20 (April), Product Innovation in Large Firms," OrganizationScience, 3
343-78. (May), 179-202.
Burgelman,RobertA. (1983), "A Process Model of InternalCor- Egge, Eric (1986), "MotivatingBuyer Actions," American Sales-
porate Venturingin the Diversified Major Firm,"Administra- man, 31 (August), 24-27.
tive Science Quarterly,28 (June), 223-44. Eisenhardt,KathleenM. (1989), "MakingFast StrategicDecisions
(1994), "FadingMemories:A Process Theory of Strategic in High-Velocity Environments," Academy of Management
Business Exit in Dynamic Environments,"AdministrativeSci- Journal, 32 (September),543-76.
ence Quarterly,39 (March),24-56. and Behnam N. Tabrizi (1995), "Accelerating Adaptive
Campbell, David (1989), "An Introductionto Nonlinear Dynam- Processes: Product Innovation in the Global Computer In-
ics," in Lectures in the Sciences of Complexity: XIII-XXII, dustry," Administrative Science Quarterly, 40 (March),
Daniel L. Stein, ed. Redwood, CA: Addison-Wesley. 84-110.
Cohen, Jacob and PatriciaCohen (1983), AppliedMultipleRegres- Etzioni, Amitai (1964), Modem Organizations.Englewood Cliffs,
sion/Correlation Analysis for Behavioral Sciences. Hillsdale, NJ: PrenticeHall.
NJ: LawrenceErlbaumAssociates. Galbraith,Jay (1973), Designing Complex Organizations.Read-
Cohen, Michael D. (1991), "IndividualLearning and Organiza- ing, MA: Addison-Wesley.
tional Routine:EmergingConnections,"OrganizationScience, Gioia, Theodore (1988), The ImperfectArt. New York: Oxford
2 (February),135-39. UniversityPress.
and Paul Bacdayan(1994), "OrganizationalRoutines Are Granovetter,Mark(1973), 'The Strengthof WeakTies,"American
Stored as Procedural Memory: Evidence from a Laboratory Journal of Sociology, 78 (May), 1360-80.
Study,"OrganizationScience, 4 (November), 554-68. (1985), "EconomicAction and Social Structure:The Prob-
, James G. March, and Johan Olson (1972), "A Garbage lem of Embeddedness,"American Journal of Sociology, 91
Can Model of OrganizationalChoice," AdministrativeScience (November),481-510.
Quarterly, 17 (March), 1-25. Griffin,Abbie (1997), 'The Effect of Projectand Process Charac-
Cohen, Wesley M. and Daniel A. Levinthal (1990), "Absorptive teristics on Product Development Cycle Time," Journal of
Capacity:A New Perspectiveon Learningand Innovation,"Ad- MarketingResearch, 34 (February),24-35.
ministrativeScience Quarterly,35 (March), 128-52. and Albert L. Page (1993), "An Interim Report on Mea-
Cooper, RobertG. and Elko J. Kleinschmidt(1986), "An Investi- suring Product Development Success and Failure," Journal
gation into the New ProductProcess: Steps, Deficiencies, and of Product Innovation Management, 10 (September),
Impact,"Journal of ProductInnovationManagement,3 (June), 291-308.
71-85. and (1996), "PDMA Success MeasurementPro-
and (1987), "New Products:What SeparatesWin- ject: Recommended Measures for Product Development Suc-
ners from Losers,"Journalof ProductInnovationManagement, cess and Failure,"Journal of ProductInnovationManagement,
4 (3), 169-84. 13 (November),478-96.
Cosse, Thomas J. and John E. Swan (1983), "StrategicMarketing Hatch,MaryJo (1997), "JazzingUp the Theory of Organizational
Planning by Product Managers-Room for Improvement," Improvisation,"Advances in Strategic Management, Vol. 14,
Journal of Marketing,47 (Summer),92-102. James P. Walsh and Ann Huff, eds. Greenwich, CT: JAI Press,
Cravens,David W. andCharlesW. LambJr.(1993), StrategicMar- 181-91.
keting ManagementCases, 4th ed. Homewood, IL: RichardD. Holbrook, Morris B. (1995), Consumer Research: Introspective
Irwin Inc. Essays on the Studyof Consumption.ThousandOaks, CA: Sage
Crossan, Mary and Marc Sorrenti(1997), "Making Sense of Im- Publications.
provisation,"in Advances in Strategic Management,Vol. 14, Huber, George P. (1991), "OrganizationalLearning: The Con-
James P. Walsh and Ann Huff, eds. Greenwich, CT: JAI Press, tributing Processes and the Literatures,"Organizational Sci-
155-80. ence, 2 (February),88-115.
Cyert, Richard M. and James G. March ([1963] 1992), A Behav- and ReubenR. McDaniel (1986), "Decision-MakingPara-
ioral Theoryof the Firm, 2d ed. Oxford: Blackwell. digm of Organizational Design," Management Science, 32
Dalrymple, Douglas J. and LeonardJ. Parsons (1990), Marketing (May), 572-89.
Management:Strategyand Cases, 5th ed. New York:John Wi- Hutchins,Edwin (1991), "OrganizingWorkby Adaptation,"Orga-
ley & Sons. nization Science, 2 (February),14-39.
Day, George (1990), Market-DrivenStrategy.New York:The Free Hutt, Michael D., Peter H. Reingen, and John R. Ronchetto Jr.
Press. (1988), "Tracing Emergent Processes in Marketing Strategy
(1994), "The Capabilities of Market-DrivenOrganiza- Formation,"Journal of Marketing,52 (January),4-19.
tions,"Journal of Marketing,58 (October),37-52. Imai, Ken-ichi, Ikujiro Nonaka, and HirotakaTakeuchi (1985),
and PrakashNedungadi(1994), "ManagerialRepresenta- "Managing the New Product Development Process: How
tions of Competitive Advantage,"Journal of Marketing, 58 Japanese Companies Learn and Unlearn,"in The Uneasy Al-
(April), 31-44. liance, K.B. Clark,R.H. Hayes, and C. Lorenz,eds. Cambridge,
and Robin Wensley (1983), "MarketingTheory with a MA: HarvardUniversity Press, 337-81.
StrategicOrientation,"Journal of Marketing,47 (Fall), 79-89. Irby,David M. (1992), "How Attending Physicians Make Instruc-
Deshpande, Rohit, John U. Farley, and Frederick E. Webster Jr. tional Decisions when ConductingTeachingRounds,"Academ-
(1993), "CorporateCulture,CustomerOrientation,and Innova- ic Medicine, 67 (10), 630-38.
July 1998

Ittner,Christopherand David F Larcker(1997), "ProductDevel- Analysis," Journal of Product Innovation Management, 11
opment Cycle Time and OrganizationalPerformance,"Journal (November), 397-417.
of MarketingResearch, 34 (February),13-33. Moorman,Christine (1995), "OrganizationalMarket Information
Jain, SubhashC. (1997), MarketingPlanning and Strategy,5th ed. Processes: CulturalAntecedents and New ProductOutcomes,"
Cincinnati:South-WesternCollege Publishing. Journal of MarketingResearch, 32 (August), 318-35.
Jaworski,BernardJ. and Ajay K. Kohli (1993), "MarketOrienta- and Anne S. Miner (1995), "Walkingthe Tightrope:Im-
tion: Antecedents and Consequences,"Journal of Marketing, provisation and Information Use in New Product Develop-
57 (July), 53-71. ment," Marketing Science Institute Report No. 95-101.
Kerin,Roger A. and RobertA. Peterson(1995), Strategic Market- Cambridge,MA: MarketingScience Institute.
ing Problems: Cases and Comments,7th ed. Englewood Cliffs, and (1997), 'The Impactof OrganizationalMem-
NJ: Prentice Hall. ory on New Product Performanceand Creativity,"Journal of
Kohli, Ajay K., and BernardJ. Jaworski(1990), "MarketOrienta- MarketingResearch,34 (February),91-107.
tion: The Construct, Research Propositions, and Managerial and (1998), "OrganizationalImprovisation and
Implications,"Journal of Marketing,54 (April), 1-18. Organizational Memory,"Academy of Management Review,
Kotler,Philip (1994), MarketingManagement:Analysis, Planning, forthcoming.
Implementation,and Control, 8th ed. Englewood Cliffs, NJ: Nonaka,Ikujiro(1990), "Redundant,OverlappingOrganization:A
Prentice Hall. JapaneseApproachto Managingthe InnovationProcess," Cal-
Lehmann,Donald R. and Russell S. Winer (1994), ProductMan- ifornia ManagementReview, 32 (Spring), 27-38.
agement. Boston: RichardD. Irwin Inc. Olson, Eric M., Orville C. Walker,and RobertW. Ruekert(1995),
Leonard-Barton,Dorothy (1992), "Core Capabilities and Core "Organizing for Effective New Product Development: The
Rigidities: A Paradox in Managing New Product Develop- ModeratingRole of ProductInnovativeness,"Journal of Mar-
ment,"Strategic ManagementJournal, 13 (Summer), 111-25. keting, 59 (January),48-62.
Levi-Strauss,Claude (1967), The Savage Mind. Chicago: Univer- Pascale, RichardT. (1984), "The Honda Effect," excerpted from
sity of Chicago Press. "Perspectives on Strategy: The Real Story Behind Honda's
Levitt, Barbara and James G. March (1988), "Organizational Success," California Management Review, 26 (Spring),
Learning,"Annual Review of Sociology, 14, 319-40. 47-72.
Lovell, Jim and Jeffrey Kluger (1995), Apollo XIII. New York:Si- (1996), "Reflections on Honda,"California Management
mon and SchusterInc. Review, 38 (Summer), 112-17.
Mangham,lain L. (1986), Power and Performancein Organiza- Pedhazur,ElazarJ. (1982), MultipleRegression in Behavioral Re-
tions: An Exploration of Executive Process. Oxford: Basil search: Explanationand Prediction.New York:Holt, Rinehart,
Blackwell. and Winston Inc.
March,JamesG. (1976), "TheTechnologyof Foolishness,"in Am- Peter, J. Paul and James H. Donnelly Jr. (1998), MarketingMan-
biguity and Choice in Organizations,James G. Marchand Jo- agement: Knowledge and Skills, 5th ed. Boston: Irwin/Mc-
han P. Olsen, eds. Norway: Universitetsfarlgt,69-81. Graw-Hill.
(1991), "Explorationand Exploitation in Organizational Pfeffer, Jeffrey (1982), Organizationsand Organization Theory.
Learning,"OrganizationScience, 2 (February),71-78. New York:Putnam.
and HerbertA. Simon (1958), Organizations.New York: Pressing, Jeff (1984), "Cognitive Processes in Improvisation,"in
John Wiley & Sons. Cognitive Processes in the Perception of Art, W.R. Crozier
Menzel, M. (1981), "Interpersonaland Unplanned Communica- and A.J. Chapman,eds. Amsterdam:North-Holland, 345-63.
tions: Indispensableor Obsolete?" in Biomedical Innovation, (1988), "Improvisation:Methods and Models," in Gener-
E.B. Robertset al., eds. Cambridge,MA: MIT Press, 155-63. ative Processes in Music: The Psychology of Performance,Im-
Miller, Chet C. and LauraB. Cardinal(1994), "StrategicPlanning provisation, and Composition, John A. Sloboda, ed. Oxford:
and FirmPerformance:A Synthesisof More thanTwo Decades Oxford UniversityPress, 129-78.
of Research," Academy of Management Journal, 37 (6), Preston, Alistair (1991), "Improvising Order," in Organization
1649-65. Analysis and Development,I.L. Mangham,ed. New York:John
Miner,Anne S. (1987), "IdiosyncraticJobs in FormalizedOrgani- Wiley & Sons.
zations," Administrative Science Quarterly, 32 (September), Quelch, John A., RobertJ. Dolan, and Thomas J. Kosnik (1993),
327-51. Marketing Management: Text and Cases. Homewood, IL:
(1990), "StructuralEvolutionThroughIdiosyncraticJobs: RichardD. Irwin Inc.
The Potential for UnplannedLearning,"OrganizationScience, Quinn,James B. (1980), Strategiesfor Change: Logical Incremen-
1 (2), 195-210. talism. Homewood, IL: RichardD. Irwin Inc.
, Christine Moorman, and Paula Bassoff (1997), "Organi- (1986), "Innovation and Corporate Strategy: Managed
zation Improvisationin New Product Development," Market Chaos,"in Technologyin the ModernCorporation:A Strategic
Science InstituteReportNo. 97-110. Cambridge,MA: Market- Perspective, Mel Horwich, ed. New York: Pergamon Press,
167-83.
ing Science Institute.
Mintzberg,Henry(1994), The Rise and Fall of StrategicPlanning. Robins, Fred (1991), "MarketingPlanning in the Large Family
New York:The Free Press. Business," Journal of Marketing Management, 7 (October),
(1996), "Learning1, Planning0," CaliforniaManagement 325-41.
Review, 38 (Summer),92-93. Rogers, Everett (1983), The Diffusion of Innovations. New York:
andAlexandraMcHugh(1985), "StrategicFormationin an The Free Press.
Adhocracy," Administrative Science Quarterly, 30 (June), Scott, Richard W. (1987), Organizations:Rational, Natural, and
160-97. Open Systems,2d ed. Englewood Cliffs, NJ: Prentice Hall.
, RichardT. Pascale, MichaelGoold, and RichardP. Rumelt Sinha, Deepak K. (1990), "TheContributionof FormalPlanningto
(1996), "The 'Honda Effect' Revisited," California Manage- Decisions," Strategic ManagementJournal, 11 (6), 479-92.
ment Review, 38 (Summer),78-117. Sinkula, James M. (1994), "Market InformationProcessing and
Montoya-Weiss,Mitzi M. and Roger Calantone(1994), "Determi- OrganizationalLearning,"Journal of Marketing,58 (January),
nants of New Product Performance:A Review and Meta- 35-45.
Improvisationin New ProductDevelopment119

Solomon, Larry (1986), "ImprovisationII," Perspectives of New Weick, Karl E. (1979), The Social Psychology of Organizing, 2d
Music, 24 (2), 224-35. ed. Reading, MA: Addison-Wesley.
Song, Z. Michael and MarkE. Parry(1997), "The Determinantsof (1987), "Substitutes for Strategy," in The Competitive
Japanese New Product Successes," Journal of MarketingRe- Challenge: Strategiesfor Industrial Innovation and Renewal,
search, 34 (February),64-76. David J. Teece, ed. New York:Harperand Row, 221-33.
Spolin, Viola (1963), Improvisationfor the Theater:A Handbook (1993a), "The Collapse of Sensemakingin Organizations:
of Teaching and Directing Techniques.Evanston, IL: North- The Mann Gulch Disaster,"AdministrativeScience Quarterly,
western UniversityPress. 38 (December), 628-52.
Sproull, Lee and Sara Keisler (1991), Connections:New Waysof (1993b), "OrganizationalRedesign as Improvisation,"in
Workingin the NetworkedOrganization.Cambridge,MA: MIT Organizational Change and Redesign, George P. Huber and
Press. William H. Glick, eds. Cary, NC: Oxford University Press,
Sutton, Howard(1990), MarketingPlanning. New York:The Con- 346-79.
ference Board. (1993c), "Managingas Improvisation:Lessons from the
Tushman, Michael and Phillip Anderson (1986), "Technological Worldof Jazz," working paper,University of Michigan Gradu-
Discontinuities and OrganizationalEnvironments,"Adminis- ate School of Business Administration.
trative Science Quarterly,31 (September),439-65. (1996), "DropYourTools: An Allegory for Organizational
Van de Ven, Andrew H. (1986), "CentralProblemsin the Manage- Studies,"AdministrativeScience Quarterly,41 (June), 301-13.
ment of Innovation," Management Science, 32 (May), Williamson, Oliver E. (1975), Marketsand Hierarchies: Analysis
590-607. and AntitrustImplications.New York:The Free Press.
(1993), "Managingthe Process of OrganizationalInnova- Wind, Jerryand Vijay Mahajan(1997), "Issues and Opportunities
tion," in OrganizationalChange and Redesign, George P. Hu- in New Product Development:An Introductionto the Special
ber and William H. Glick, eds. Cary, NC: Oxford University Issue,"Journal of MarketingResearch, 34 (February),1-11.
Press, 269-94. and Thomas S. Robertson (1983), "MarketingStrategic:
Walker,Orville C., Jr., and Robert W. Ruekert (1987), "Market- New Directions for Theory and Research,"Journal of Market-
ing's Role in the Implementationof Business Strategies,"Jour- ing, 47 (Spring), 12-25.
nal of Marketing,51 (July), 15-33. Winter,Sidney G. (1987), "Knowledgeand Competence as Strate-
Walsh, James P. (1995), "Managerialand OrganizationalCogni- gic Assets," in The CompetitiveChallenge: Strategiesfor In-
tion: Notes from a Trip Down Memory Lane," Organization dustrial Innovation and Renewal, David J. Teece, ed. New
Science, 6 (May-June),280-321. York:Harperand Row, 159-85.
and Gerardo Rivera Ungson (1991), "Organizational Zaltman,Gerald, RobertDuncan, and Jonny Holbek (1973), Inno-
Memory," Academy of Management Review, 16 (January), vations and Organizations.New York:John Wiley & Sons.
57-91.
July 1998


Moorman 1998 B

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Moorman 1998 B

Uploaded by

Copyright:

Available Formats

The Convergence of Planning and Execution: Improvisation in New Product Development

Author(s): Christine Moorman and Anne S. Miner

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

The Convergenceof Planning and

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

2 / Journalof Marketing,July 1998

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/3

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

4 / Journalof Marketing,July 1998

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/ 5

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

6 1 Journalof Marketing,July 1998

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/ 7

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/ 9

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

(1) Organizationalimprovisation 4.25 1.98 .83

101 Journalof Marketing,July 1998

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/11

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

A.The Incidence of Organizational Improvisation

B. Factors Influencing the Incidence of Organizational Improvisation

Dependent Variable b t-value b t-value b t-value

C. Factors Influencing the Effectiveness of Organizational Improvisation

Improvisationin New ProductDevelopment113

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/15

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

16 / Journalof Marketing,July 1998

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment/17

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

Improvisationin New ProductDevelopment119

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

This content downloaded from 152.3.153.148 on Mon, 23 Sep 2013 16:38:04 PM

You might also like