Research Policy 35 (2006) 545–561

Innovation in project-based firms: The context dependencyof success factors

Floortje Blindenbach-Driessen ∗, Jan van den EndeRSM Erasmus University, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands

Received 17 July 2004; received in revised form 17 January 2006; accepted 16 February 2006Available online 18 April 2006

Abstract

Innovation management literature typically concerns functionally organized firms. In this paper we investigate innovation man-agement in a different type of firm, the project-based firm. Project-based firms, such as engineering and construction companies,consultancies and system integrators, are service firms that solely execute projects for clients. We focus on new service developmentprojects in these firms. Based on an in-depth study of six projects in four different firms, we develop hypotheses on differencesbetween success factors for development projects in project-based firms and in functionally organized firms. Some of the successfactors for functionally organized firms, as described in the literature, appear to be more important in project-based firms, othersseem redundant. Our findings suggest that the specific structure and capabilities of project-based firms provide an explanation forthese differences.© 2006 Elsevier B.V. All rights reserved.

K

1

oooif

bov

f

(

0

eywords: Project-based firms; Innovation management; New service development projects

. Introduction

Most literature on innovation management implicitlyr explicitly concerns manufacturing and other technol-gy firms (Montoya-Weiss and Calantone, 1994). Manyf these firms are functionally organized. The questions whether the findings from this literature are also validor firms with a different type of organizational structure.

In this paper we study innovation processes in project-ased firms. Project-based firms have a project-basedrganization (Hobday, 2000) and use projects to pro-ide unique services to their clients. These services

∗ Corresponding author. Tel.: +31 10 408 1719;ax: +31 10 408 9014.

E-mail addresses: fblindenbach@rsm.nlF. Blindenbach-Driessen), jende@rsm.nl (J. van den Ende).

can be combinations of custom-designed products andrelated services. Examples are engineering and con-struction companies, consultancies and system integra-tors. Project-based firms contribute considerably to thegross national product of most Western economies (Gannand Salter, 2000; Knight Wendling Consulting, 2000).Innovation in these firms involves developing new orimproved services for current or prospective customers,or developing new technologies that can be used to solveclients’ problems better than existing technologies.

The organization of project-based firms is clearlydistinct from functionally organized firms, such as mostmanufacturing firms (Hobday, 2000). Starting pointof this research is the assumption that organizationalcontext, or more specifically the structure and capa-bilities required for the execution of projects by orderof clients, affects the management of developmentprojects. As a project manager of an innovation project

048-7333/$ – see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.respol.2006.02.005

546 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

in a project-based firm explained to us: “New productdevelopment seems to be all about project managementand we know how to manage projects”. In our opinionalso managers of project-based firms can learn from theinnovation management literature, if the findings of thisliterature are re-investigated for this type of firm.

1.1. Project-based firms

Project-based firms are firms that are set up aroundprojects (Gann and Salter, 2000), and that produce com-plex services for their clients (De Brentani and Ragot,1996; Prencipe et al., 2003). The services usually consistof the integration of products or systems, such as IT sys-tems, delivered in a business-to-business environment.Hobday (2000) distinguished between project-based andproject-led organizations. A project-based organizationis an organization in which the functional organiza-tion has become completely obsolete, without formalfunctional coordination of activities. Such an organiza-tion is entirely dedicated to one or more projects. In aproject-led organization the needs of projects outweighthe functional influence on decision-making. A project-led firm still has some characteristics of a functional firm,since there is some coordination of functionally equiv-alent activities. Our definition of project-based firmsencompasses both the project-led and the project-basedorganization. Although accountancy firms and lawyer’soffices also carry out projects, we exclude these firmsin this paper, since their projects often concern special-

definition, design, construction, implementation, to test-ing and operation, although not necessarily every projectincludes all phases.

On top of business projects, project-based firms alsoperform development projects. Development projectsare projects aimed at innovation, and take place sepa-rately from business projects. In development projectsnew services are developed for a range of customers withthe objective to commercialize these services. Thesedevelopment projects are the central subject of thispaper.

Few authors have studied innovation activities inproject-based firms. Gann and Salter (2000), who werethe first to address this subject, noted that innovationactivities in project-based firms are typically notexecuted in separate R&D departments, but performedwithin or closely related to business projects. Executionof development projects with the same resources asused for business activities is typical for service firms(Sundbo, 1997; Sundbo and Gallouj, 2000). Keeganand Turner (2002) have pointed out the reluctanceof managers to develop innovations within businessprojects. Moreover, they point out that for innovationsdeveloped within business projects, the application oftraditional linear project management practices havea negative impact on the success of these innovativeactivities.

We will demonstrate that this latter problem alsoapplies to development projects that take place sepa-rately from business projects. Gann and Salter (2000)

ized services of limited size and not the integration ofproducts and systems.

Business projects are those projects of project-basedfirms that are executed by order of a specific exter-nal client. These project offer unique solutions to eachclient (Hobday, 2000). Clients typically initiate a busi-ness project, define the specifications, provide finan-cial resources, and at the end benefit from the deliv-erables (Turner, 1999). Business projects are managedautonomously (Hobday, 2000), primarily bounded by thecontractual agreement with a client (Turner, 1999). Fre-quently, a multi-disciplinary team is used to execute abusiness project and more disciplines will be involvedif a project is more complex. Expertise and reputationof personnel is important for the success of businessprojects, although many project-based firms counterbal-ance fluctuations in demand with temporary personnel.Business projects differ across firms with respect to dura-tion, content, and project size. For instance, the content ofa construction project bears little resemblance with thatof an IT project. Despite such differences, the phasesof such projects show many similarities, from concept

suggest that project-based firms could better make amore strict distinction between their generic businessprocesses, including research and development, andbusiness projects. At the same time they note that separat-ing these activities will likely hamper knowledge transferbetween R&D and business projects.

In this paper we focus on development projects ofproject-based firms. Development activities that are partof business projects are excluded, since these activitiesare usually customer specific and not intended for fur-ther commercialization. Moreover, the embeddednessof development activities within business projects hin-ders the study of the management of these activitiesand reduces the comparability with development projectsin other firms and industries. Focusing on developmentprojects of project-based firms makes it possible to com-pare new service development in project-based firmswith the existing literature on new service and new prod-uct development. This approach also enables the inves-tigation of the impact of the structure and capabilities ofproject-based firms on the management of developmentprojects.

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 547

1.2. New service development

Since most project-based firms are service firms, thefindings of the literature on new service development(Cooper and Edgett, 1999; De Brentani, 1989; Martinand Horne, 1993; Sundbo, 1997) apply to project-basedfirms. Many authors on new service development arguethat differences in product and service firms’ outputaffect the success factors for innovation. They mentionin particular the intangibility and perishability of ser-vices, and the required user participation in their delivery(De Brentani, 2001; Gallouj and Weinstein, 1997). Fac-tors that are considered more important for new servicedevelopment projects than for new product develop-ment projects are: having a distinct and reliable con-cept (De Brentani and Ragot, 1996), synergy with thefirm’s current business (De Brentani and Ragot, 1996;Martin and Horne, 1993), senior management support(Cooper, 2001; De Brentani, 2001; Griffin, 1997), andcustomer involvement (Bowen and Ford, 2002; de Jonget al., 2003; Martin and Horne, 1993). On the otherhand a structured approach would be less important (DeBrentani, 2001; Griffin, 1997; Martin and Horne, 1993).However, in their meta study, Henard and Szymanski(2001) found only a few differences between successfactors for new service and new project development:a higher importance of market synergy, a lower impor-tance of a structured approach and of cross-functionalteams. The higher importance of market synergy can beexplained with reference to the specific character of ser-vdavfiatoa

sstwccptticw

to manufacturing firms (Woodward, 1980). Secondly, inproject-based firms business projects are more impor-tant than the functional organization (Hobday, 2000).Often project-based firms are even organized per areaof expertise, combining sales, research and productionin one department dedicated to this specific area, mak-ing functional departments entirely redundant (Hobday,2000). Thirdly, project-based organizations have uniquecapabilities with respect to project-management, inter-nal collaboration and collaboration with customers andsuppliers. We expect that the these differences have moreimpact on the management of development projects, thanthe impact due to different idiosyncratic features in whatis developed, e.g. services versus products.

We choose an exploratory approach because of thelimited availability of publications on innovation man-agement in project-based firms. Based on qualitativecase-study research we develop hypotheses on successfactors for new service development projects in project-based firms, compared to success factors for new productand new service development projects in other firms.We define a success factor as a factor that has a signif-icant positive impact on the success of a developmentproject and thus contributes to explaining the differencebetween success and failure of development projects,within a specific group of firms. Our definition of suc-cess factors does not imply that these factors are easilyimitable; compliance with a factor can still be difficult.The reason to use success factors is that the innova-tion management literature has provided an extensive

ices, since the intangibility of a service makes it moreifficult to explain the advantages of a new service indvance. As a consequence the acceptance of a new ser-ice depends on the already established reputation of therm in that specific area (De Brentani, 1989; De Brentanind Ragot, 1996). There is, however, no obvious rela-ion between the specific character of services and thether two differences, lower importance of a structuredpproach and cross-functional teams.

Because of the limited results of the literature on newervice development, in this paper we investigate newervice development projects from a different perspec-ive. We do not focus on the effect of the difference inhat is produced, but on the effect of the organizational

ontext. New service development in project-based firmsould be quite different from what is described in the newroduct and service development literature, because ofhe different organizational structure and capabilities ofhese firms, compared to other service or manufactur-ng firms. Firstly, project-based firms deliver unique andomplex services that ask for an organizational structureith less hierarchy and a higher span of control compared

set of such factors, which facilitates the comparisonbetween project-based firms and current literature. Sec-ondly, these factors refer to a broad range of aspects ofdevelopment projects, and therefore form a good start-ing point for an exploratory study (Eisenhardt, 1989).Finally, despite the exploratory character of this study,we do not want to confine ourselves to a descriptiveapproach only. Success factors facilitate the investiga-tion of the effects of specific behaviours and conditionson performance within project-based firms, and thus alsothe development of hypotheses regarding these effects(Brown and Eisenhardt, 1995; Ernst, 2002; Montoya-Weiss and Calantone, 1994). Moreover, we address theinfluence of the specific structure and capabilities ofproject-based firms on these success factors.

2. Methodology

2.1. Sample

The four companies, in which we studied the devel-opment projects, concerned an engineering firm, an

548 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

engineering-consultancy firm, a construction company,and a financial consultancy firm. As suggested byEisenhardt (1989, p. 537), variety in industries and sizewithin our sample enabled us to control environmen-tal variation, while the focus on project-based firmsconstrained variation due to organizational differencesamongst these firms. We therefore selected differenttypes of project-based firms. The size of these firms var-ied from approximately 80 employees for the financialconsultant to more than a 1000 for both the engineeringand the engineering-consultancy firm. Also ownershipdiffered. One consultancy firm had a partnership struc-ture; the other consultancy firm was a subsidiary of afirm listed on Euronext. The construction firm was familyowned and two foundations owned the engineering firm.

We also constrained variation by investigating thedevelopment projects of these project-based firms andexcluding innovation efforts on business projects. Eachof the four firms had development projects that were exe-cuted separately from their business activities, and noneof the firms had a separate R&D department. Two ofthe four firms had development committees to guide thefirm’s new service development efforts. The construc-tion firm was renowned for its innovativeness. Withinthe consultancy firm development projects were well

institutionalised. The sample had a slight bias towardsproject-based firms with a prospector strategy (Miles andSnow, 1978), thus, towards firms that try to be ahead ofthe competition by being the first with new products, newtechnologies, or with entering new markets. The reasonwas that firms with a tradition of innovation appeared tobe more interested to participate in this research. Noneof the firms operated in truly dynamic markets with fre-quent changes in technology. One business unit of theengineering firm was involved in the relatively dynamicmarket of soil cleaning, a market that in the Netherlandsstarted in the 1980s and has been growing since.

To achieve theoretical replication (Yin, 1994), weasked each firm to select a successful and a less suc-cessful or failed project. The construction firm had onlyone project available for our study. One project in oneof the other firms had to be skipped, as it concernedthe improvement of an internal process rather than thedevelopment of a new service. The six remaining projectswere all recently completed or about to be finalized (seeTable 1). The sample included incremental as well asradical innovations (Garcia and Calantone, 2002): watercontest was a new to the world service concept and tar-geted at a new market. Architecture was a new to theworld concept targeted at the firm’s existing market. Soil

Table 1Case descriptions

Case

al consung conctested dcute subbased oevelopcontest. The pal. Howhe engiernment, which

ding moldings aribute tohich im

a gearnt thatrather s

nding aients hae internate-of-tt madehold ha

Outsourcing A dedicated development team of a financiand combined it into one integral outsourciresult was a successful new service. It wasefficiently and effectively organize and exeadaptations had to be made to the concept,

Water Contest A governmental prize contest was used to dengineering firm. Participation in the prizeleading to a lot of free publicity for the firma business success, because it was too radicconcept were considered very valuable by t

Soil Cleaning An engineering firm developed, with a govbut technology development is still ongoing

Building Modules A construction company tried to apply builcombination with the pre-design of two buiclients of the business projects did not conthave been put on hold (for other reasons), wprocess was put on hold as well.

e-Business An engineering consultancy firm developedjuniors, which convinced senior managemetheir logistic knowledge. Management wasjuniors the task to prove their concept by fijuniors developed the gearbox. Although clclient. This was partly due to collapse of th

Architecture A financial consultant wanted to create a sttheir clients. After a first attempt the projecdeveloped. This second attempt was put on

ltant gathered knowledge of three outsourcing projects of this firmept. Experienced but mostly junior staff executed this project. Theirectly on a business project. The new concept made it possible tosequent outsourcing trajectories. Only minor refinements andn the experiences of the first business project.the concept of a radical new solution for water management by anresulted in a nomination amongst the best five in the first round,rize itself was not won, and the developed concept as a whole was notever, the knowledge gained, the obtained publicity, and parts of the

neering firm.grant, a new soil cleaning technology. Pilot projects were successful,delays commercialization.dules in their construction process. They developed the concept ins part of business projects, in order to have concrete examples. Thethese development efforts. Coincidentally both business projectsplied that the implementation of building models in the construction

box for the e-Business market. The initiative was taken by a group ofthe firm should enter the e-Business advice market on the basis ofceptical but agreed to fund the development project, and it gave thefirst client. Without further guidance from senior management, thed been consulted on the concept, the project team failed to find a firstet hype in the meanwhile.he-art business architecture for the insurance and banking business ofa restart, but there was still little consensus about what needed to belf-way, because another development project got priority.

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 549

Table 2Overview of the success factors from literature

Success factors mentioned in the literature Sources

Planning of workPlanning and effective execution 2,5Contingent approach 1,4

Senior management involvement 1,2,3,4,5

TeamCross-functional teams 1,2,4Expertise 1,4Heavyweight project leader 1,2Product champion 2,4External communication 1,2

Involvement of outside partiesCustomer involvement 1,2Supplier involvement 1

Activities undertakenPre-development 2,3,5Market research and testing 2,3,5Launch 4,5

Used meta-reviews: 1 = Brown and Eisenhardt (1997); 2 = Ernst(2002); 3 = Montoya-Weiss and Calantone (1994); 4 = Panne (2003);5 = Henard and Szymanski (2001).

Cleaning was an addition to existing lines, with new tothe world technology. e-Business was new to the firm,Building Modules an addition to existing services andOutsourcing was an improvement of an existing service.

For each project we interviewed at least three persons:the project leader, his or her supervisor from senior man-agement, and one or more project members. In total weconducted 21 interviews, each about 1.5 h, which weretaped and subsequently transcribed (NVivo program,Richards, 1999). The interviews were semi-structured,ensuring that at least the items listed in Table 2 wereaddressed. The respondents also got ample opportunityto express their views on the most important success fac-tors in their project. During the interviews with seniormanagement we also queried the more general charac-teristics of the firm, the structure of the firm and thecapabilities needed to execute business projects. In addi-tion, we studied project documentation, such as projectplans, minutes of meetings, progress and final reports, incompany archives.

2.2. Success factors

We derived a set of success factors from five meta-studies on new product and new service development(Brown and Eisenhardt, 1995; Ernst, 2002; Henard andSzymanski, 2001; Montoya-Weiss and Calantone, 1994;Panne, 2003), see Table 2. The meta-reviews were not

conclusive about planning of work. Some recommenda rather structured approach of planning and effectiveexecution (Ernst, 2002; Henard and Szymanski, 2001).Others advocate a more contingent approach—planningand effective execution for incremental projects and amore organic approach for radical development projects(Brown and Eisenhardt, 1995; Panne, 2003). We haveincluded both in Table 2. A detailed description of eachfactor is given in the results section. We limited ourselvesto process factors, because of our focus on the organi-zation and management of development projects. Othertypes of success factors, such as product features, marketcharacteristics or price, were therefore excluded.

We defined project success by means of a multiple-criteria scale, including the following indicators(Adapted from Griffin and Page, 1996):

- Project performance: on time, within budget, quality.- Market performances: use of service by clients, possi-

bly as part of other services, impact on reputation ofthe firm.

- Learning effects for future innovation activities.

We included reputation as it appeared to be a rela-tively important goal for the development projects withinthe firms we studied (see also De Brentani, 1989). Forinstance, a new technology was introduced not to gainmarket share, but to demonstrate the firm’s state-of-the-art knowledge. We had to exclude some performancecriteria from the list of Griffin and Page (1996), such

as the ROI rates, increase of market share, profit mar-gins and turnover, since the project-based firms in oursample had not tracked these data for the developmentprojects. In line with the literature, which indicates thatrating success by multiple sources increases validity(Hoegl and Gemuenden, 2001; Yin, 1994), we askedthree respondents – a team member, the project man-ager and the supervisor from senior management – torate the project’s performance.

2.3. Procedure

As a starting point we investigated for each projecthow the project had scored on our performance indi-cators, and whether the project had complied with thefactors listed in Table 2. If a respondent indicated thata project did not comply we inquired after the rea-sons. Since most respondents worked or had workedon business projects, they could provide us with goodinsights in each firm’s structure and capabilities, partic-ularly as present in business projects. It was difficult forthe respondents to assess whether the firm’s characteris-

550 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

Fig. 1. The influence of the project-based context on success factorsfor development projects.

tics, such as the organizational structure or capabilities,affected the degree of compliance with a factor, becausemost respondents had no reference point other than theirown firm or other project-based firms they worked for.

Next we assessed the impact of each factor on thesuccess of each development project. This assessmentwas based on the discussions we had with the respon-dents about whether or not the factors as described inTable 2 contributed significantly to success, or uponabsence were perceived as omission negatively impact-ing project’s success. The respondents could clearlyindicate whether or not a factor had been relevantfor the success of their respective project. During thiswhole procedure we kept receptive of other potentialsources of differences or similarities, in line with Eisen-hardt’s suggestion for exploratory case study research(Eisenhardt, 1989). Alternative explanations of our find-ings are addressed in Section 4.

The hypotheses regarding success factors for devel-opment projects in project-based firms are derived in twoways:

The degree of compliance refers to relationship (1)in Fig. 1. In the event of a relatively low compliancewith a specific factor, potentially due to the structure orcapabilities of the project-based firms, we hypothesizethat project-based firms can improve the performance ofdevelopment projects by adhering to this factor. In casethe structure or capabilities of the project-based firmsseemed to lead to a high degree of compliance with a spe-cific factor, we hypothesize that such an omnipresent fac-

factors, we hypothesize that the project-based contextaffects the importance of this success factor as indicatedby relationship (2) in Fig. 1.

The need to comply with a specific success factorrefers to relationship (3) in Fig. 1. Most success factorsin the innovation management literature are aimed atovercoming specific problems in a development project(for instance related to internal or external collabora-tion). In the event of reduced problem(s) targeted by afactor in project-based firms, we expect that the needto comply with that factor is limited and that the suc-cess factor is thus less important. On the contrary, inthe case that the structure or capabilities of the project-based firms seemed to increase the problem, or led tothe introduction of new problems, we hypothesize thatthe respective factor is more important for developmentprojects in project-based firms compared to other firms.The need to comply is thus also expected to affect theimportance of a success factor.

3. Results

Table 3 summarizes the perceived performance ofthe six development projects. With respect to marketperformance, only outsourcing was truly successful,underscoring the problems project-based firms havein exploiting new services. Five of the six projectswere considered successful with respect to the projectperformance criteria ‘on time’ and ‘within budget’.Only one project was said to have delivered superior

tor will not differentiates between successful and faileddevelopment projects, nor do we expect that additionaladherence to such a factor contributes significantly toperformance of development projects within this groupof firms.

When our observations suggested that the firm contextinfluenced the degree of compliance with the success

quality; the quality delivered by the other projectswas perceived as mediocre or even insufficient. Activepromotion and positive reputation effects were evidentin three of the projects. Two projects failed to capturelearning effects. For the other projects, presentationswere held to introduce the concept to colleagues. In oneof these cases (e-Business) this knowledge transfer wasineffective, because the concept eluded the audience.In each case, there appeared to be a high degree ofconsensus amongst the respondents about the project’sperformance. Overall we classified two of the projectsas successful, two as mediocre, and two as failed.

In what follows, we discuss our results by factor. Wegive a short description of each factor based on innova-tion management literature and describe how the projectsin our sample complied with the success factor and towhich extent compliance with the factor contributed orpotentially could have contributed to the success of thedevelopment projects. Next we discuss how the con-text of the project-based firms influenced the compliancewith each success factor and to what extent the contextincreased or reduced a factor’s contribution to success.

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 551

Table 3Perceived performance assessment of the projects

Outsourcing Water Contest Soil Cleaning Building Modules e-Business Architecture

Project performanceTime ++ ++ ± − + –Budget ++ +/− + + + +/−Quality ++ +/− +/− + − –

Market performanceUse by clients ++ +/− − − – –Reputation ++ ++ ++ −/+ − –

Learning effectsExchanged knowledge with colleagues ++ ++ ++ +/− − –Captured knowledge for future reference ++ + ++ +/− – –

Overall performance ++ + +/− −/+ − –

Finally, we formulate hypotheses regarding the rela-tive importance of the success factor for developmentprojects of project-based firms, compared to develop-ment projects in functionally organized firms.

3.1. Planning of work

Cooper (2001) stresses the importance of planningand effective execution of development projects. Severalauthors add that overlapping phases and iteration posi-tively affect performance, particularly in case of projectswith a high uncertainty, such as radical developmentprojects (Brown and Eisenhardt, 1997; De Meyer et al.,2002; Eisenhardt and Tabrizi, 1995).

In all our cases we found a straightforward projectplanning approach in development projects. The plan-ning procedures were similar to those used in busi-ness projects, in which on-time delivery in accordancewith the contractual agreements is important. The firmsapplied the same methods to their development projects;projects were planned sequentially and executed strictlyaccording to these plans. As a consequence, the projectleaders considered changes or delays in developmentprojects undesirable; they even perceived changes anddelays as failure. They did not apply iterative planningmethods, not even in the radical projects. The expe-riences from the more radical cases suggest that theuncertainty in these projects made the straightforwardplanning ineffective. An illustrative example was theproject Architecture, where the project leader forced atitsr

reducing the team’s productivity dramatically. It clearlyprecipitated the team’s downfall.

The project-based firms applied straightforward plan-ning methods, because of their experience on busi-ness projects with this method. This core competenceseemed to have become a core rigidity (Leonard-Barton,1992), as project-based firms have difficulty to adaptto a contingent planning approach (Keegan and Turner,2002), also for their development projects. We suspectthat adherence to a contingent planning approach inaccordance with uncertainty could have significantlyimproved performance of the development projects westudied. We therefore hypothesize that a contingent plan-ning approach is a more important success factor fordevelopment projects for project-based firms comparedto functionally organized firms.

3.2. Senior management

3.2.1. Project selectionEffective selection procedures, exemplified in the

so-called funnel, contribute to the success of develop-ment projects, since weak projects are cancelled andmore resources remain for the other ones (Cooper, 2001;Cooper et al., 1998; Wheelwright and Clark, 1992).

No explicit selection criteria existed for project selec-tion in the four firms. The responsible executives, eithersenior managers or development committees, intuitivelymade approval decisions. Moreover, acceptance crite-ria appeared to depend on availability of employees. For

ime boxing approach upon his team members, whilet was a radical project with many uncertainties. Theeam failed to comply since they did not know what theyhould, and could deliver within each time box. Thisesulted in conflicts and frustrations within the team,

example, e-Business was rejected at first, but when a fewconsultants ‘sat on the bench’ it got approved after all.

Senior management assigned a budget to eachapproved project. However, the resources necessary werenot always made available. The development projects

552 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

competed for the same resources as the business projects,while development projects rarely got priority over busi-ness projects. This resulted in frequent changes in thecomposition of the development project-teams; some-times even the whole core team had changed beforethe project finished. Clearly such changes had a neg-ative impact on project performance. In spite of theresource constrains the project teams still delivered ontime. Team members often worked (unpaid) overtimeto meet their deadlines. Most project members did notmind the overtime, because they took pleasure in thedevelopment projects. This implied that a smaller num-ber of registered hours were spent than expected and thatmost projects reported within time and budget. However,the lack of resources and the strict keeping to dead-lines resulted in a decreased quality of the output. Thetrade-off between time and quality is mentioned in theliterature (Cooper, 2001; Wheelwright and Clark, 1992),but the scale rarely tips to time above quality. This pref-erence is especially awkward, since the firms we studiedoperated in rather stable markets where speed is pre-sumed to be less important (Cooper, 2001; Eisenhardtand Tabrizi, 1995; Eppinger, 2001; Wheelwright andClark, 1992)

In the firms with a development committee respon-sible for the execution of development projects, assign-ment of resources and budget was more in balance. Inaddition, we observed that in one firm more explicitproject selection increased the visibility of developmentprojects. In this firm, employees were freed from busi-

Clark, 1992). Such support should consist not only oftangible support as addressed above, but also of intan-gible support (Ernst, 2002). Senior management has toexert ‘subtle control’, by providing a vision on the tar-geted outcome of the project (Brown and Eisenhardt,1995), but should not disturb the project on a daily basisor strangle the project by too much control (Bonner etal., 2002; Gersick, 1994).

Although senior management or the developmentcommittee initially approved plans for each develop-ment project, ‘subtle control’ and a vision of seniormanagement on the contents of the projects appeared tobe absent in all projects. Detailed process or end productrequirements and a vision on the product as a wholewere missing in the project plans. Only occasionally andmainly informally senior management was informedabout progress. As one of the respondents said “I wouldappreciate it when they (senior managers) would inquireabout the project now and then, as recognition of thework we do, but apparently they have a lot of trust in us”.At the same time, senior managers complained that theyhad little grasp on the projects and felt unsure about theoutcome. Apparently senior management did not realizethat their involvement, beyond conflict resolution, wouldbe appreciated. Some senior managers were afraid thatinvolvement in the contents of a development projectwould be interpreted as mistrust. As a consequenceof the low degree of management involvement, theproject teams could rather easily lower the quality ofthe deliverables, in order to cope with the resource

ness projects during the period they were scheduledon a development project, resulting in fewer changeswithin the development project teams. On the contrary,the firms without a development committee executedany development project that seemed reasonable, as longas surplus resources were available. Especially in thesefirms the many dispersed initiatives were not alwaystaken seriously. In most project-based firms employ-ees will be shared between development projects andthe always more urgent business projects (Sundbo andGallouj, 2000). It is likely that explicit project selec-tion will increase the status of development projects andas such will make it easier to free and retain resourcesfor development projects. We hypothesize therefore thatin project-based firms explicit selection of developmentprojects is a more important success factor than in func-tionally organized firms.

3.2.2. SupportSupport from senior management is a crucial success

factor for development projects (Brown and Eisenhardt,1995; De Brentani and Ragot, 1996; Wheelwright and

limitations. Two projects in which external partiesplayed an important role, Soil Cleaning and WaterContest, clearly differed in this respect. These projectshad fixed quality objectives, and these objectives weresupervised and used as yardsticks by the externalparties.

The reluctance of senior management to get moreinvolved in projects seemed to originate from businessprojects, in which they only interfered in case of troubles.Business projects operate completely autonomously insome project-based firms (Hobday, 2000). In all fourfirms, senior management was also used to delegatemuch authority to business project managers. This auton-omy worked well for business projects, which are usuallybound by contracts with clients, but the developmentprojects without external control wandered off with-out additional guidance. We therefore hypothesize thatwithin project-based firms senior management support,i.e. the provision of a clear vision on the outcome of theproject (Brown and Eisenhardt, 1995), is more impor-tant for the performance of development projects than infunctionally organized firms.

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 553

3.3. Team structure

3.3.1. Cross-functional teamsThe innovation management literature emphasizes

the importance of cross-functional teams and effectivecollaboration within those teams as success factors(Brown and Eisenhardt, 1995; Cooper, 2001; Cooperand Edgett, 1999; De Brentani and Ragot, 1996; Griffin,1997; Lovelace et al., 2001; Song and Montoya-Weiss,2001).

The core of each project team with three to four peoplewas rather small in our sample. No functional depart-ments (R&D, marketing, operations, sales, etc.) existedand hence the concept of cross-functional teams is irrel-evant in the context of project-based firms. Within thisspecific structure we viewed multi-disciplinarity as theequivalent of cross-functionality. The team members ofthe projects we investigated were familiar with operat-ing in multi-disciplinary teams. Collaboration amongstproject members of various disciplines was in generaljudged as very good, also within the under-performingprojects.

The project organization of the firms we studiedenhanced the collaboration between different disci-plines. Employees were familiar with the requirementsand needs of the other disciplines and were used tocollaborating. Because the level of compliance withmulti-disciplinary collaboration was high in the project-based firms we studied, multi-disciplinary collaborationappeared neither a necessity, nor a differentiating fac-tpucl(oblo

3

tE1mia1

c

quate sourcing of expertise was an important conditionfor success. For the project Architecture, for instance,the architecture experts were only advisers to the projectteam, not members. From the sideline these experts stilltried to leave their mark on the project, which confusedand frustrated the project team. On the other hand, theSoil Cleaning team had all expertise on board, whichcontributed to the effectiveness of this team. Expertswere typically the driving force behind many businessprojects and as such only partly available for develop-ment work. Availability of expertise seemed to add sig-nificantly to the effectiveness of the development teamswe observed, but was not a prerequisite for approvalof development projects. Since in project-based firmsemployees are typically selected on the basis of avail-ability instead of expertise (Pinto and Covin, 1989),we hypothesize that in these firms expertise is an moreimportant success factor for development projects thanin functionally organized firms.

3.3.3. Project leadersMany scholars plea for heavyweight project leaders

(Brown and Eisenhardt, 1995; Rothwell et al., 1974;Wheelwright and Clark, 1992). Wheelwright and Clark(1992, p. 195) have defined heavyweight project leadersas leaders that are capable to interpret the market, under-stand the multi-languages of different departments, dealwith engineering issues, communicate effectively insidethe team as well as outside, while guarding the concept,and resolve conflicts. They consider such heavyweight

or between success and failure of the developmentrojects. Each project benefited from the already presentnderstanding of team members’ backgrounds and dis-iplinary needs. We expect that multi-disciplinary col-aboration is omnipresent in most project-based firmsHobday, 2000) and consequently also exists on devel-pment project. We therefore hypothesize that in project-ased firms a multi-disciplinary development team is aess important success factor compared to functionallyrganized firms.

.3.2. ExpertiseThe availability of relevant expertise contributes

o the success of development projects (Brown andisenhardt, 1995; Cooper, 2001; Cooper and Edgett,999; De Brentani and Ragot, 1996; Griffin, 1997). Inany functionally organized firms a lot of attention

s therefore dedicated to the provision of knowledge-ble staff for development projects (Pinto and Covin,989).

In the cases we studied, team members were typicallyhosen on the basis of availability, not expertise. Ade-

project managers important to overcome departmentsilos and to pull a development project together. Theyconsider a lightweight project leader to be ineffectivefor more radical projects, because he or she is rathera messenger than a manager (Wheelwright and Clark,1992, p. 195).

A heavyweight project leader headed one project,Building Modules. Less experienced project leadersfrom middle and junior management led the otherprojects. Senior managers we interviewed did not com-plain about the ineffectiveness of project leaders, andno project leader said to have had difficulties withthe above described tasks. Only conflict resolution wasnot their task, but their superiors’. A handicap of theuse of lightweight project leaders is resource alloca-tion. As said before, most project leaders had difficul-ties to get and retain resources for their developmentprojects; not because of their limited weight, but predom-inantly because of low commitment of senior manage-ment (Turner, 1999). Hence, as we observed, the projectleaders could rather easily accomplish their tasks; heavy-weight leaders seemed to have been superfluous.

554 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

Project leaders of project-based firms are in the ‘mainline of communication and can exercise control to coor-dinate and integrate specialists and functions in creativenew ways, focusing on the needs of the projects. Becausethere are few internal lines of command and communi-cation to interfere with project objectives, the internalcoordination tasks become thus simpler and clearer’(Hobday, 2000, p. 890). This also seems to hold for thetasks of a leader of a development project. We hypoth-esize therefore that heavyweight project leaders are lessimportant as a success factor for development projectsin project-based firms.

3.3.4. Product championsThe presence of a product champion is important for

the promotion of an development project (Ancona andCaldwell, 1992; Brown and Eisenhardt, 1995; Rothwellet al., 1974). A product champion should preferably bethe initiator of a new product and stay involved in aproject throughout its life-cycle (Panne, 2003). The lit-erature leaves open whether a project leader or someonefrom senior management better fulfill this role (Ernst,2002).

In four of our projects the project leader acted asthe product champion. In one project, Architecture, theproduct champion was someone from senior manage-ment, which appeared unsuitable. Not fully aware ofall the issues going on within the development project,he pushed the project too much. For the other cases anenthusiastic, highly involved promoter of the concept

munication also helped to reduce resource constraints.The less successful projects lacked this kind of com-munication, because these project members had little toshare, hence, external team communication boosts thesuccess of already promising projects only.

Our findings confirm that external team communica-tion is important for development projects, leading tothe hypothesis that external team communication con-tributes to the success of development projects in project-based firms to an equal degree as in other firms.

3.4. Outside parties

3.4.1. Customer involvementCustomer involvement is essential for development

projects (Brown and Eisenhardt, 1995; Ernst, 2002; VonHippel, 1986), at least in incremental projects. In radicalprojects it is less important since customers cannot antic-ipate the problems and opportunities involved in such aproduct (Christensen, 1997). Customers were involvedin two of our projects. In Soil Cleaning, a customer pro-vided a testing site but had little role in the project itself.In Building Modules, a client’s building requirements ofa business project was used as a blueprint. In both cases,the clients were thus hardly involved in the develop-ment activities. Outsourcing deliberately did not involvetarget customers; the project leader was of the opinionthat the team had enough experience and insights in cus-tomer demands. The involvement of customers seemedto add little to the already known information on cus-

within and outside the firm contributed to the credibilityof each of the projects. Such a product champion wasable to gain commitment of each team member. In thisrespect we found no indications that the role of a prod-uct champion would be different in project-based firmsthan elsewhere. This leads to the hypothesis that a prod-uct champion contributes to the success of developmentprojects in project-based firms to an equal degree as infunctionally organized firms.

3.3.5. External team communicationAncona and Caldwell (1992) found that teams that

focus their external interaction both to persuade others ofthe importance of a team’s work and to coordinate, nego-tiate, and obtain feedback from outside groups, makethese teams move ahead quickly on budgets and sched-ule in the short term, and manage to produce the mostinnovative products over the course of the developmentprocess. This finding is confirmed by our more success-ful projects (Outsourcing, Water Contest, Soil Clean-ing, Building Modules), which were actively promotedwithin their respective firms. Such external team com-

tomer needs and demands, and as such did not seem tocontribute to the success of the development projects westudied.

Close collaboration with customers is typical forbusiness-to-business project-based firms (De Brentaniand Ragot, 1996; Gann and Salter, 2000). Firms thatknow their customers well have less need to investigatecustomer needs (Maidique and Zirger, 1985). We there-fore hypothesize, in line with the reasoning of Maidiqueand Zirger, that customer involvement contributes lessto the success of development projects in project-basedfirms compared to functionally organized firms.

3.4.2. Supplier involvementSuppliers should be involved as early as possible in

development projects to improve quality and to preventdelays (Clark et al., 1987; Iansiti and Clark, 1994). Intwo of the development projects, early collaboration withsuppliers, sharing technical know-how, was meaningful.Both these development projects were executed underthe control of the respective project-based firms; thus,the suppliers were not co-producers. The respondents

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 555

perceived collaboration with suppliers as a commonbusiness procedure, since their firms already collabo-rated with these suppliers on business projects. Therespondents indicated that supplier involvement madethe development projects more complex, for instancecomplicating negations about license agreements forbusiness projects-to-be-acquired in the future. This lat-ter problem seemed to be closer related to the inexpe-rience with licensing, than with difficulties related tosupplier collaboration. The other projects required nosupplier involvement, because the expertise of supplierswas already present, or could be developed in-house.

Project-based firms generally have extensive andclose relationships with suppliers (Gann and Salter,2000), and as we observed, the development projectsmade use of these existing supplier networks. Supplierobstructionism, which jeopardizes the contribution ofsupplier involvement to project performance (Primo andAmundson, 2002), did not seem to play a role.

From our observations it seems likely that in casesupplier collaboration is needed, suppliers are indeedbeing involved. The level of compliance to this fac-tor is thus high in project-based firms. We did observethat supplier involvement made the development projectmore complex, but according to our respondents, theadded complexity did not negatively affect project per-formance. We did not observe that supplier involvementwas a differentiating factor between success and failure,since suppliers were involved in all projects in which thiswas appropriate, and in the other projects we could notfiisf

3

3

ncas

wAcsatcu

investigation, but based on the many other developmentinitiatives within these firms that were not specificallyaddressed in this research; business cases could facili-tate early project closure of less promising projects.

The firms in our sample had little experience in mak-ing business cases. Business projects clients typicallyinvestigate the feasibility of a project themselves. In gen-eral, the creation and approval of business plans helps togain more realistic insight in the potential of an invention(Cooper, 2001), and also contributes to portfolio man-agement of development projects initiatives (Roussel etal., 1991). Since project-based firms likely lack capabil-ities in this area, we hypothesize that creating a businesscase, or executing a pre-development phase, is a moreimportant success factor for development projects inproject-based firms than in functionally organized firms.

3.5.2. Market researchMarket research is an important success factor for

development projects, at least for incremental projects(Brown and Eisenhardt, 1995; Ernst, 2002; Montoya-Weiss and Calantone, 1994). Market research does notcontribute to the success of more radical developmentprojects (De Brentani, 2001; Song and Montoya-Weiss,2001). Firms that know their customers well can relyon their gut feeling to determine the preferences of theircustomers (Maidique and Zirger, 1985).

The respondents had the idea that they clearly under-stood the preferences of their potential customers. Con-sequently none of the project teams performed market

nd indications that supplier involvement would havemproved project performance. Therefore, we hypothe-ize that supplier involvement is a less important successactor in project-based firms than in functional firms.

.5. Activities undertaken

.5.1. Pre-developmentIn the pre-development stage, commercial and tech-

ical feasibility are investigated (Ernst, 2002), whichan be used for project approval. Project selection waslready discussed above, here we focus on feasibilitytudies; the creation of a business case.

Business cases with financial and market forecastsere lacking in five of the six projects. For one project,rchitecture, the project leader made a financial fore-

ast, based on his experience only. Although the mostuccessful project, Outsourcing, had no business plan,nd the least successful project, Architecture, containedhe most elaborate business plan, we claim that a businessase contributes to project performance. Not in partic-lar related to the six projects that were subject of our

research. When there were doubts – as was the case inthe more radical projects water contest and e-Business– some of the potential customers were contacted. Forinstance the e-Business team consulted two potentialcustomers for the tool’s specifications, but these sameclients were not willing to buy the tool afterwards. Later,one of these clients came back with a request for a sim-ilar e-business solution. Unfortunately by that time theinvolved employees with the necessary e-business exper-tise had left the firm. For the more radical developmentprojects market research did not seem to have addedto project success, which corresponds with the findingsof De Brentani (2001) and Song and Montoya-Weiss(2001).

We got the impression from our respondents that cus-tomer’s needs were well understood, because of the goodcontacts senior managers, project leaders and team mem-bers had with potential customers. As said before, closecollaboration with customers is typical of project-basedfirms and in general will make market research infor-mation redundant (Maidique and Zirger, 1985). Mostof the new developed services never reached the mar-

556 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

ket, it was therefore difficult to assess whether or notcustomer needs were indeed fulfilled. In addition, asexplained above, most projects lowered quality targetsdue to time constrains. We think that the problem ofinsufficient quality was not due to a lack of understand-ing customer needs, but in not adhering to the qualityobjectives set in order to meet these customer needs. Wetherefore hypothesize that market research contributesless to the success of development projects in project-based firms, than in functionally organized firms.

3.5.3. TestingAnother way to investigate whether a new service ful-

fils customer needs, is to test it (Edvardsson and Haglund,1995; Thomke, 2003). Testing a service is more com-plex than testing products because it can be done onlyin collaboration with customers (Bowen and Ford, 2002;De Brentani and Ragot, 1996; Shostack, 1984; Thomke,2003). Five of our projects created testable services. Onlyone project, Soil Cleaning, actually included interme-diate testing in pilot projects, to refine and update thetechnology. Outsourcing was tested on the first businessproject and refined. In the project “water contest”, thecontest was not used as a test to upgrade the concept.After the nomination the concept was slightly adaptedto suit the wishes of the jury, but when the final conceptwas not awarded, it was not developed further to suit itspotential customers.

Since market research does not seem very useful inproject-based firms, testing may offer a better opportu-

in the area of soil cleaning and capable to handle all soilpollution problems with the most recent technologies.Team members presented at conventions, seminars andworkshops to promote the firm’s reputation in general,not the new service in particular.

Outsourcing promoted its new service through aworkshop with a limited number of potential clients.Shortly thereafter the firm acquired the targeted numberof contracts for new outsourcing projects. A respondentstated: “It is a market in which you need to make a lotof noise, to ensure that the name of your firm imme-diately pops up, when clients think about outsourcing”.The team noted that the firm had to avoid creating capac-ity problems: “These are not the type of jobs (businessprojects) of which we can run 5 or 6 in parallel, becauseespecially in the implementation phase you need a largecrew”. Hence, Outsourcing provides a good example ofhow a balanced launch campaign can be used to promotea new service without creating an over-demand.

Like business-to-business firms, also project-basedfirms seem to have to promote their new service moreactively (De Brentani and Ragot, 1996). Such launch-ing efforts have to be balanced to ensure that humanresources can keep up with the demand. If orders can-not be executed properly, a firm’s reputation can getdamaged, and for project-based firms reputation is veryimportant (Gann and Salter, 2000). We hypothesizetherefore that for development projects in project-basedfirms a (balanced) launch is a more important successfactor than for development projects in functionally orga-

nity to adapt a new service to customer needs. Just as inother service firms, the testing phase seems to be the mostappropriate moment to adapt a service to customer needs(Martin and Horne, 1993; Thomke, 2003). We hypoth-esize therefore that in project-based firms testing willcontribute more to the success of development projectsthan in functionally organized firms.

3.5.4. LaunchLaunch activities create awareness of a new prod-

uct or service amongst target customers (Panne, 2003).De Brentani and Ragot (1996) stress that it is importantthat professional services are promoted to customers.“It is not enough that new service products solve clientproblems; their benefits must be clearly understood andperceived as superior to competitive offerings”.

Some teams considered their task completed whenthey had developed a concept. Few of the projects plansincorporated strategies to enter the market, whereas someprojects were used to promote the firm’s reputation ingeneral. For example the Soil Cleaning project was usedto show that the engineering firm was the absolute expert

nized firms.

4. Discussion

We have investigated to which extent success fac-tors from the new product and new service develop-ment literature can be applied to development projectsin project-based firms. We excluded innovative activitieson business projects, the projects performed for a specificclient. We made an in-depth investigation of six cases ofdevelopment projects in four project-based firms. Thesecases suggest that success factors frequently mentionedwithin the literature also apply to some extent to project-based firms. However, there are remarkable differences,both positive and negative, in the degree that these firmsconform to these factors and in the importance of thesefactors (see Table 4 ). Some success factors seem tobe more important for project-based firms comparedto other firms: the application of contingent planningapproaches, explicit project selection, senior manage-ment support, the availability of sufficient experts, mak-ing business cases and testing and launching the new

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 557

Table 4Summary and hypothesized differences between project-based firms and functionally organized firms

Set of successfactors

Compliance withfactor of observeddevelopment projects

Observedcontribution tosuccess

Contextualinfluences

Hypothesized relativecontribution tosuccess

Planning of workPlanning and contingent approach No use of contingent

planning approachesStraightforward planningapproach hinderedespecially the moreradical developmentprojects

Capabilities instraightforward planningmade appliance of morecontingent planning toolsdifficult

More important

Senior management involvementProject selection Explicit project selection

only in firms withdevelopment committee

Selection increasedvisibility and status,subsequently moreresources were available

Competition for resourceswith more urgent businessprojects hinderedresource allocation

More important

Support Projects operatedautonomously withoutmuch senior managementsupport

Too much autonomymade that the projectswandered off, either in thewrong direction or towardinsufficient quality

Senior management wasused to give muchautonomy to projects.External clients usuallyguard the quality ofprojects. This led to a lowdegree of seniormanagement support fordevelopment projects

More important

TeamCross-functional teams Multi-disciplinary

collaboration was selfevident and equallypresent on all six projects

Multi-disciplinarycollaboration did notseem to differentiate thesuccessful from the lesssuccessful projects

Multi-disciplinary projectorganization enhancedcollaboration and createda high level of mutualunderstanding

Less important

Expertise Experts not alwayssufficiently available fordevelopment projects

Absence of expertsdecreased performance

Competition for expertswith more urgent businessprojects reduced theiravailability

More important

Heavyweight project leader Heavyweight projectleaders were rarely used

Light or medium weightproject leaders weresufficient to manage thedevelopment projects

Each organization wasfully geared towards theexecution of projects,which facilitated the taskof the project leaders

Less important

Product champion Enthusiastic and drivenindividuals championedthe projects and products

Product championspromoted their projectseffectively

None Equally important

External team communication Successful projects usedexternal communicationchannels to promote theirprojects within the firm

External promotionenhances the visibility ofproject (for successfulprojects only)

None Equally important

Involvement of outside partiesCustomer involvement Sufficient knowledge

present within team,clients were contacted forradical developmentprojects

Customer needs seemedwell understood. Clientswere not a good source ofinformation for the moreradical projects

Close collaboration withcustomers on businessprojects made that clientneeds were wellunderstood, leading to alower need to involvecustomers

Less important

558 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

Table 4 (Continued )

Set of successfactors

Compliance withfactor of observeddevelopment projects

Observedcontribution tosuccess

Contextualinfluences

Hypothesized relativecontribution tosuccess

Supplier involvement When needed supplierswere involved.Collaboration withsuppliers went smooth

Involvement of suppliersdid not seem todifferentiate thesuccessful from the lesssuccessful projects

Development projectsmade use of the existingrelationships withsuppliers

Less important

Activities undertakenPre-development Only for one project a

business case was madeBusiness cases will helpto create more realisticexpectations ofdevelopment projects andcan also be used forproject selection

The project-based firmshad little experience withmaking business cases

More important

Market research No market research wasdone

Customer needs seemedwell understood

Close collaboration withcustomers on businessprojects made that clientneeds were wellunderstood

Less important

Testing Some projects tested thenew services

Testing was perceiveduseful to adapt thedeveloped service to thecustomer needs,especially for the projectsin which customers hadnot been involved

None More important

Launch Project launch notintegral part ofdevelopment projects

Balanced launchedcampaign, in which thecapacity to deliver wastaken into account, madethe project Outsourcingsuccessful

The project-based firmswere used to wait forrequests from customersfor business projects.Organized launchactivities were anexception

More important

services. Other factors seem to be less important: theuse of cross-functional teams, heavyweight project man-agers, collaboration with customers and suppliers andperforming market research. The involvement of prod-uct champions and external team communication seemto be equally important for the success of new servicedevelopment projects compared to functionally orga-nized firms. Characteristics of project-based firms thatcould explain these differences are the structure of thesefirms consisting of multidisciplinary departments, lead-ing to capabilities in internal collaboration, the autonomyof the project managers relative to senior management,the guidance of business projects by clients, leading toa high priority of business projects over other activities,and the capabilities in external collaboration.

We used success factors as a reference. This approachallowed us to explore systematically the differencesbetween project-based firms and the literature on

functionally organized firms. Distinguishing betweenthe actual compliance on success factors and the needto comply took us away from a purely descriptiveapproach and facilitated developing hypotheses. More-over, this approach stimulated a vision on innovationmanagement that takes into account the wider firmcontext and the resulting requirements with respectto the organization and management of developmentprojects.

Our paper contributes to the literature on innova-tion management in project-based firms, by providinginsights in the management of development projects inproject-based firm. Furthermore, we investigated howthe organizational structure and capabilities of thesefirms impacted the management of the developmentprojects. We thereby challenge the claim of Eisenhardtand Martin (2000) that success factors for innovationmanagement are universal:

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 559

“Commonalities arise because there are more and lesseffective ways of dealing with the specific organi-zational, interpersonal and technical challenges thatmust be addressed. . . Just as there are better and worseways to hit a golf ball or ski a mogul field, there aremore and less effective ways to execute new productdevelopment projects”.

Instead we argue for a more contingent approach withrespect to the management and organization of new prod-uct and service development activities.

Our results support some of the findings of new ser-vice development literature. For instance, Griffin (1997)found that new service development projects havefewer phases than new product development projects,particularly in underperforming firms. The omission ofpre-development, market research, testing and launchactivities in some of our projects supports this finding.As we indicated in the introduction, we relate thedifferences between project-based firms and other firmsto firm characteristics, whereas authors on new servicedevelopment relate their findings to the characteristicsof what is produced. Our approach may provide newexplanations for findings from the new service devel-opment literature. For instance, in the introduction wementioned that Henard and Szymanski (2001) found alower need for multidisciplinary teams in new servicedevelopment activities, but that this lower need canhardly be attributed to the different character of servicescompared to products. We also found a lower need formb

4

iifi

opbidoafsfHh

integrated these activities within its business organiza-tion. We would therefore suggest keeping developmentprojects close to the business activities, but explicitlyapplying different managerial procedures, as indicatedin this paper.

5. Limitations and directions for future research

This study has several limitations. We investigatedonly six development projects in four firms, in whichwe explored the effects of capabilities on the organi-zation and the success of development projects, andcompared our findings with the innovation managementliterature. To confirm the generalizibility of our findingsfuture research is needed using a larger sample of devel-opment projects of both project-based and functionallyorganized firms. Moreover, such research can includefactors not addressed in this paper, such as the effectsof co-production of new services and collaboration withexternal parties such as research institutes.

Furthermore, we focused on the similarities betweenfirms. There were significant differences between thefour project-based firms in size, industry and ownership.We did not find effects of these differences in our studyof development projects. For instance, the differencesin ownership structure between the family owned firmand the subsidiary of a firm listed on Euronext were notreflected in differences in the attention of top manage-ment for innovation or on innovative performance. Thelarger firms had more resources available (one of the key

ultidisciplinary teams, but we relate this to the capa-ilities of project-based firms in internal collaboration.

.1. Managerial implications

For managers of project-based firms our findingsmply that the current innovation management literatures useful as a reference, but that the situation of their ownrm requires considerable adaptation of the findings.

In accordance with the findings of (Engwall, 2003),ur study shows that the management of developmentrojects was influenced by the experience of project-ased firms with business projects, which are performedn a complex but rather stable environment. It wasifficult for the project-based firms to execute devel-pment projects with high uncertainty, unclear targetsnd lower complexity. Separating development projectsrom the rest of the organization leads to undesiredide-effects (Gann and Salter, 2000), particularly in theorm of reduced exchange of information (Griffin andauser, 1996). This was exactly the reason why one firmad dismantled its R&D organization in the past, and

success factors), but they did not seem to perform betterin innovation.

On the other hand the variety in our sample was alsolimited. All firms operated in supplier-driven industries(Pavitt, 1984), did not produce extremely complex ser-vices (Hobday, 2000), they all had a prospector strategy(Miles and Snow, 1978), and operated in a rather stableenvironment. Future research involving more firms andmore development projects will be necessary to investi-gate the effects of firm differences. Particularly researchon project-based firms in more science-based sectors andon firms with different strategies may reveal other firminfluences on success factors for innovation.

In the paper we paid little attention to differencesbetween the six development projects regarding theirtechnological diversity, relative importance for the firm,or size. However, it appeared that these differencesdid not impact the way development projects weremanaged. The managers were always enthusiastic,relatively young employees, who had an enormousdrive to develop something new for the benefit of theirfirm. None of the projects were driven by customers,

560 F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561

nor co-produced with suppliers, albeit two projectswere guided by external parties, a subsidy provider anda contest committee. The involvement of such outsideparties made a difference, since these projects hada structure more in accordance with that of businessprojects. Further research is required to investigate theeffects of differences in project characteristics.

The interplay between innovation activities on busi-ness projects and the development projects is anotherarea for further research. As we noted in the introduc-tion, part of the innovation activities in project-basedfirms is performed within business projects. An interest-ing issue for further research is the interplay between theexperiences and outcomes of these activities and thoseof dedicated development projects.

A more methodological issue for future researchinvolves the specific performance measures for develop-ment projects of project-based firms, and the appropriaterespondents to measure these. The traditional successmeasures such as return on investments and marketshare (Griffin and Page, 1996) were not applicable toevaluate the success of development projects in project-based firms. For instance, our respondents considered itimpossible to calculate break-even points or return oninvestment of development projects, since they couldnot determine the contribution of development projectsto the acquisition of new contracts. On the other hand,contribution to the reputation of the firm appeared tobe an important performance measure, since reputationis of great importance for project-base firms (Gann and

References

Ancona, D.G., Caldwell, D.F., 1992. Bridging the boundary: externalactivity and performance in organizational teams. AdministrativeScience Quarterly 37 (4), 634–651.

Bonner, J.M., Ruekert, R.W., et al., 2002. Upper management con-trol of new product development projects and project performance.Journal of Product Innovation Management 19, 233–245.

Bowen, J., Ford, R.C., 2002. Managing service organizations: doeshaving a thing make a difference. Journal of Management 28 (3),447–469.

Brown, S.L., Eisenhardt, K.M., 1995. Product development: pastresearch, present findings, and future directions. Academy of Man-agement Review 20 (2), 343–378.

Brown, S.L., Eisenhardt, K.M., 1997. The art of continuous change:linking complexity theory and time-paced evolution in relentlesslyshifting organizations. Administrative Science Quarterly 42, 1–34.

Christensen, C.M., 1997. The Innovator’s Dilemma: When New Tech-nologies Cause Great Firms to Fail. Boston Business School Press,Boston.

Clark, K.B., Chew, W.B., et al., 1987. Product development in the worldauto industry. Brookings Papers on Economic Activity 3, 729.

Cooper, R.G., 2001. Winning at New Products, Accelerating the Pro-cess from Idea to Launch. Massachusetts, Perseus Publishing,Cambridge.

Cooper, R.G., Edgett, S.J., 1999. Product Development for the ServiceSector. Massachusetts, Perseus Books, Cambridge.

Cooper, R.G., Edgett, S.J., et al., 1998. Portfolio Management for NewProducts. Addisson-Wesley, Reading, Mass.

De Brentani, U., 1989. Success and failure in new industrial services.Journal of Product Innovation Management 6, 239–258.

De Brentani, U., 2001. Innovative versus incremental new businessservices: different keys for achieving success. Journal of ProductInnovation Management 18, 169–187.

De Brentani, U., Ragot, E., 1996. Developing new business-to-business

Salter, 2000; Hoch et al., 2000). The common way toassess reputation is by asking respondents for the per-ceived contribution of a new service to the reputationof the firm in that particular area (De Brentani, 1989).The reliability of this approach may be questionable,but better measurement methods for reputation effects ofindividual new service development projects still have tobe developed.

The exploration of these issues will enhance ourunderstanding of the specifics of innovation manage-ment in project-based firms, and the characteristics ofproject-based firms in general.

Acknowledgments

We gratefully acknowledge the useful comments andsuggestions made at the EIASM International ProductDevelopment Management Conference, Brussels, 2003,and at the R&D Management Conference, Sesimbra Por-tugal 2004. We are also grateful to Steef van de Veldeand two anonymous referees for their helpful commentsand insights.

professional services: what factors impact performance. IndustrialMarketing Management 25, 517–530.

de Jong, J.P.J., Bruins, A., et al., 2003. Innovation in service firmsexplored: what, how and why? Literature Review. Zoetermeer EIMBusiness & Policy Research, 73.

De Meyer, A., Loch, C.H., et al., 2002. Managing project uncertainty:from variation to chaos. MIT Sloan Management Review 43 (2),60–67.

Edvardsson, B., Haglund, L., 1995. Analysis, planning, improvisationand control in the development of new services. International Jour-nal of Service Industry Management 6 (2), 24–35.

Eisenhardt, K.M., 1989. Building theories from case study research.Academy of Management Review 14 (4), 532550.

Eisenhardt, K.M., Martin, J.A., 2000. Dynamic capabilities: what arethey? Strategic Management Journal 21, 1105–1121.

Eisenhardt, K.M., Tabrizi, B.N., 1995. Accelerating adaptive pro-cesses: product innovation in the global computer industry. Admin-istrative Science Quarterly 40, 84–110.

Engwall, M., 2003. No project is an island: linking projects to historyand context. Research Policy 32, 798–808.

Eppinger, S.D., 2001. Innovation at the speed of information. HarvardBusiness Review, 3–10.

Ernst, H., 2002. Success factors of new product development: a reviewof the empirical literature. International Journal of ManagementReviews 4 (1), 1–40.

Gallouj, F., Weinstein, O., 1997. Innovation in services. Research Pol-icy 26, 537–556.

F. Blindenbach-Driessen, J. van den Ende / Research Policy 35 (2006) 545–561 561

Gann, D.M., Salter, A.J., 2000. Innovation in project-based, service-enhanced firms: the construction of complex products and systems.Research Policy 29, 955–972.

Garcia, R., Calantone, R., 2002. A critical look at technological innova-tion typology and innovativeness terminology: a literature review.Journal of Product Innovation Management 19, 110–132.

Gersick, C.J.G., 1994. Pacing strategic change: the case of a new ven-ture. Academy of Management Journal 37 (1), 9–45.

Griffin, A., 1997. PDMA research on new product development prac-tices: updating trends and benchmarking best practices. Journal ofProduct Innovation Management 14, 429–458.

Griffin, A., Hauser, J.R., 1996. Integrating R&D and marketing: areview and analysis of the literature. Journal of Product Innova-tion Management 13, 191–215.

Griffin, A., Page, A.L., 1996. PDMA success measurement project:recommended measures for product development success and fail-ure. Journal of Product Innovation Management 13 (4), 478–496.

Henard, D.H., Szymanski, D.M., 2001. Why some new products aremore successful than others. Journal of Marketing Research 38 (3),362–375.

Hobday, M., 2000. The project-based organisation: an ideal form formanaging complex products and systems? Research Policy 29,871–893.

Hoch, D.J., Roeding, C.R., et al., 2000. Secrets of Software Success:Management Insights from 100 Software Firms around the World.Harvard Business School Press, Boston, MA.

Hoegl, M., Gemuenden, H.G., 2001. Teamwork quality and the suc-cess of innovative projects: a theoretical concept and empiricalevidence. Organization Science 12 (4), 435–449.

Iansiti, M., Clark, K.B., 1994. Integration and dynamic capability:evidence from product development in automobiles and mainframecomputers. Industrial and Corporate Change 3 (3), 557–605.

Keegan, A., Turner, J.R., 2002. The management of innovation inproject-based firms. Long Range Planning 35, 367–388.

Knight Wendling Consulting, 2000. Bedrijfstaktoets 2000, Ministerie

L

L

M

M

Miles, R.E., Snow, C.C., 1978. Organizational Strategy, Structure, andProcess. McGraw-Hill, New York.

Montoya-Weiss, M.M., Calantone, R., 1994. Determinants of newproduct performance: a review and meta-analysis. Journal of Prod-uct Innovation Management 11, 397–417.

Panne, G.v.d, 2003. Success and failure of innovation: a literaturereview. International Journal of Innovation Management 7 (3),309–338.

Pavitt, K., 1984. Sectoral patterns of technical change: towards a tax-onomy and a theory. Research Policy 13, 343–373.

Pinto, J.K., Covin, J.G., 1989. Critical factors in project implementa-tion: a comparison of construction and R&D projects. Technova-tion 9, 49–62.

Prencipe, A., Davies, A., et al. (Eds.), 2003. The Business of SystemsIntegration. Oxford University Press, Oxford.

Primo, M.A.M., Amundson, S.D., 2002. An exploratory study of theeffects of supplier relationships on new product development out-comes. Journal of Operations Management 20, 33–52.

Richards, L., 1999. Using Nvivo in Qualitative Research. QualitativeSolutions and Research Pty. Ltd., Bundoora, Vic.

Rothwell, R., Freeman, C., et al., 1974. Sappho updated. ResearchPolicy 3, 258–291.

Roussel, P.A., Saad, K.N., et al., 1991. Third Generation R&D: Man-aging the Link to Corporate Strategy. Harvard Business SchoolPress, Boston.

Shostack, G.L., 1984. Designing services that deliver. Harvard Busi-ness Review, 133–139.

Song, M., Montoya-Weiss, M.M., 2001. The effects of perceivedtechnological uncertainty on Japanese new product development.Academy of Management Journal 44 (1), 61–80.

Sundbo, J., 1997. Management of innovation in services. The ServiceIndustries Journal 17 (3), 432–455.

Sundbo, J., Gallouj, F., 2000. Innovation as a loosely coupled sys-tem in services. International Journal of Services Technology andManagement 1 (1), 15–35.

van economische zaken.eonard-Barton, D., 1992. Core capabilities and core rigidities: a para-

dox in managing new product development. Strategic ManagementJournal, 13.

ovelace, K., Shapiro, D.L., et al., 2001. Maximizing cross-functionalnew product teams’ innovativeness and constraint adherence: aconflict communications perspective. Academy of ManagementJournal 44 (4), 779–793.

aidique, M.A., Zirger, B.J., 1985. The new product learning cycle.Research Policy 14, 299–313.

artin, C.R., Horne, D.A., 1993. Service innovation: successful ver-sus unsuccessful firms. International Journal of Service IndustryManagement 4 (1), 49–65.

Thomke, S., 2003. R&D comes to services: bank of America’spathbreaking experiment. Harvard Business Review 81, 70–79.

Turner, J.R., 1999. The Handbook of Project-based Management.McGraw Hill, Berkshire.

Von Hippel, E., 1986. Lead users: a source of novel product concepts.Management Science 32 (7), 791–805.

Wheelwright, S.C., Clark, K.B., 1992. Revolutionizing Product Devel-opment. The Free Press, New York.

Woodward, J., 1980. Industrial Organization: Theory and Practice.Oxford University Press, Oxford.

Yin, R.K., 1994. Case Study Research Design and Methods. SAGEPublications, Inc., Thousand Oaks.