Designing and facilitating collaboration in R&D: A case study

Designing and facilitating collaboration in R&D:A case study§

Jerald Hage a, Gretchen Jordan b, Jonathon Mote a,*, Yuko Whitestone a

a University of Maryland, United Statesb Sandia National Laboratories, United States

1. Introduction

Although the impact of co-location in research and technology or product development (R&D) istypically assumed to be positive, there are scant few empirical studies that provide an in-depthexploration of this practice (Kahn and McDonough, 1997). While a number of studies exist that focuson the related issue of collaborative networks (for example see Danilovic and Winroth, 2005; Johansenet al., 2005; von Corswant and Tunalv, 2002), often overlooked are the organizational mechanismsthat allow for integrating the diversity of cross-functional teams, both within and across organizations(Holland et al., 2000; Susman and Majchrzak, 2003). Further, there is very little in the literature thatexplores the initial organizational decisions that formed the co-location effort as well as themanagement practices that sustain the ongoing unit. Finally, a great deal of the studies that do exist

J. Eng. Technol. Manage. 25 (2008) 256–268

A R T I C L E I N F O

Article history:

Available online 5 November 2008

JEL classification:

O32

Keywords:

R&D management

Innovation

Complexity

Co-location

Cross-functional teams

A B S T R A C T

This case study aims to highlight the strategic decisions and

managerial practices in the formation and operation of a co-located

research unit within a national laboratory. The empirical evidence is

based on interviews with members of the research unit as well as

responses from a research environment survey. The findings of the

case study suggest specific strategies that are conducive not only for

the co-location of research units but also for research management

in general. Principal among these are the need to balance increases

in diversity and complexity with mechanisms of integration and the

use of specific management practices and leadership qualities that

support these activities.

� 2008 Elsevier B.V. All rights reserved.

§ The opinions expressed are those of the authors, not the U.S. Department of Energy or Sandia National Laboratories.* Corresponding author at: Center for Innovation, Department of Sociology, University of Maryland, College Park, MD 20742,

United States. Tel.: +1 301 405 9746; fax: +1 301 314 6892.

E-mail address: [email protected] (J. Mote).

Contents lists available at ScienceDirect

J. Eng. Technol. Manage.

journal homepage: www.elsevier.com/locate/jengtecman

0923-4748/$ – see front matter � 2008 Elsevier B.V. All rights reserved.

doi:10.1016/j.jengtecman.2008.10.005

mailto:[email protected]

http://www.sciencedirect.com/science/journal/09234748

http://dx.doi.org/10.1016/j.jengtecman.2008.10.005

focus primarily on new product development or the integration of design or marketing withmanufacturing, rather than the co-location of a research unit with another part of the R&D continuum.

The motivation to co-locate is typically driven by the assumption that cross-functionalcommunication and teams lead to increased or accelerated innovation. In the organizationalliterature, the role of a complex division of labor, like that found in cross-functional teams, has beenidentified as a critical factor in facilitating organizational innovation. As Hage (1999) demonstrated ina comprehensive review of the organizational innovation literature, a complex division of labor is akey determinant for facilitating innovation, as it encompasses the organizational learning, problem-solving, and creativity capacities of an organization. But the practice of co-location highlights a centralissue in the management of R&D of how to strike a balance between increasing the complexity of laborto increase innovation, while at the same time ensuring adequate integration (Nooteboom, 1999,2000). As Leenders et al. (2003) discuss, integrated and active interactions among researchers plays akey role in promoting the cross-fertilization of ideas and creativity necessary for innovation. While theneed for integration is recognized as critical in the management of R&D (Allen, 1977; Leenders et al.,2003; West, 2004), the successful attainment of integration represents a challenge for managers(Nihtila, 1999; Sicotte and Langley, 2000; Holland et al., 2000). And while the use of co-location isbecoming more common, the limited evidence of the results of these attempts at cross-functionalintegration is often mixed (Kahn and McDonough, 1997).

Further, it should go without saying that a co-location effort, or any research effort, cannot besuccessful without strong leadership, particularly in the initial management decisions. As Von Zedwitz(2003) discusses, the formation of new research units entails a series of decisions that impact thedevelopment of these units over time, including the selection of an appropriate manager. Andleadership practices and styles, in general, have been demonstrated to have a significant impact onR&D performance (Oh et al., 1991; McDonough and Barczak, 1991; Green, 1995; Sicotte and Langley,2000; Stoker et al., 2001; Cordero et al., 2004). Yet, we would argue that there are few studies thatprovide practical insights on successful leadership styles or practices in the R&D literature.

The objective of this paper is to discuss the issues of balancing diversity and integration andleadership in R&D through an examination of a case study of the formation and co-location of adedicated basic research unit within a manufacturing department (S&T MD) in a large nationallaboratory (hereafter NATLAB) in the United States. The research unit was formed to focus on themanufacturing department’s single product, a component which requires extreme precision, exoticmaterials and highly advanced processes in its manufacture. It was anticipated that the integration ofbasic research in the production facility would result in fewer technical surprises on the productionline and quicker resolution of problems that do arise. Our case study encompasses not only the currentactivities of the S&T MD, but also the initial decisions and actions that led to the formation of the unit.In this manner, the case study provides a relatively more comprehensive investigation of a novelapplication of the use of co-location.

But the case study should also have special interest to R&D managers for several reasons beyondthe issues raised above. First, the case study focuses on a co-location effort at the level of basicresearch, while most studies on co-location focus on efforts in product development. Second, the casestudy involves the co-location of a basic research unit within a manufacturing unit, which, to ourknowledge, is relatively rare. Finally, the need for scientific and technological research units formanufacturing is becoming greater because frequently radical innovations utilize advanced processtechnologies and the challenge of these manufacturing units involves addressing technicalcomplexities in the product and the manufacturing process.

In the next section of the paper, we discuss the applied theory of radical innovation that underliessome of the assumptions in the analysis of the case study. After a brief discussion of the methodology,we discuss the efforts of the unit in developing complexity or diversity and how integration isperceived by the five scientists in this unit, using both the interviews and the results of a researchenvironment survey. To provide some basis of comparison, the survey results of the unit’s scientistsare compared to another experiment in co-location within the NATLAB, which we will refer to asCOLO, as well as the overall researcher perceptions for NATLAB. In the subsequent section, we discussthe management practices and leadership style of the unit’s manager that facilitated the co-locationeffort, including a long-term scientific vision, cognitive mentoring and providing emotional support to

J. Hage et al. / J. Eng. Technol. Manage. 25 (2008) 256–268 257

researchers. The study concludes with a discussion of the results and implications for further researchon R&D management.

2. An applied theory of radical innovation

One of the central issues in the management of R&D is ensuring a balance between increasing thecomplex division of labor needed to pursue radical innovation and maintaining integration amongthe research team (Hage, 1999). A complex division of labor is important because, as Nooteboom(1999, 2000) hypothesized, radical innovation is more likely the greater the cognitive (or knowledge)distance there is within the team. Paradoxically, however, communication also tends to decline withcognitive distance. Hence, the challenge is to strike a balance between complexity and integrationrelative to the goal of radical innovation. Perhaps even more critical is the need to maintainintegration as the diversity or complexity grows, rather than assuming the two remain static overtime.

Although there is a considerable amount of research on the relationship between complexity andinnovation in the industrial innovation literature, there has been a surprising absence of researchabout the problems of integration (Hage, 1999). Further, Nooteboom’s concept of ‘‘optimal cognitivedistance,’’ which posits a successful balance between diversity and integration, has undergone onlycursory empirical examination to date (Wuyts et al., 2005). In recent qualitative studies ofbiomedicine, however, Hollingsworth et al. (in press) found integration does pose a serious challenge,particularly with regard to the recruitment of new competencies, which creates internaldifferentiation that prevents integration across these diverse perspectives.

Internal differentiation and the challenge of balance are compounded when the need arises foradditional expertise during the course of a research project, a common occurrence as new researchissues emerge. For instance, in a study of some 20 research projects in the same NATLAB, we found thatmost of them added new skills during the course of their 3 years and in the larger projects or programsof research sometimes had as many as eight different specialties (Jordan et al., 2005).

One might think that the problem of maintaining integration is easily solved if the new skills arerecruited into the same unit but this increases the size of the unit, reduces its flexibility, and, mostcritically, leads to internal differentiation as the specialists create occupational or disciplinary groupswithin the research project or program. It is this tendency for like-minded individuals to gravitatetogether that starts to inhibit communication within cross-functional teams. To prevent the growth insize and the internal differentiation, another solution is to temporarily hire people to join the project.This has the advantage of preventing the growth in size and allowing for flexibility but then thereother issues of whether the temporary people are truly becoming integrated into the project and theirdegree of commitment. After a brief discussion of the data and methodology, we turn to aninvestigation of the formation of the S&T MD to explore these issues.

3. The methodology

This study uses a single-case study design to explore a unique example of co-location, one that wewould argue can make a contribution to the existing knowledge and theory of R&D management. Theprimary data used in this case study comes from interviews and survey responses of the staff membersof the S&T MD, which is located at a large national laboratory. The laboratory currently employsseveral thousand researchers in over two-dozen disciplinary centers and has a multi-billion dollarbudget. Such large laboratories are interesting settings to explore questions about R&D, but have beenlargely overlooked in the literature. The S&T MD is a relatively small unit in the laboratory, consistingof one manager, five scientists, five technologists, one contractor, and one support staff. Thedocumentation of the history of the formation of the S&T MD, as well as its experience with complexityand integration, were obtained using the following three sources of data:

(1) One face-to-face interview and two short qualitative interviews by phone with the unit managerabout management rationale and strategies for the formation of the unit.

J. Hage et al. / J. Eng. Technol. Manage. 25 (2008) 256–268258

(2) One-half hour interviews with all the scientists (in person and by phone) and with several of thetechnologists in the department about managerial practices relative to three themes.

(3) A survey designed specifically to elicit information about specific aspects of a researchenvironment was administered to all scientists and technologists in the department.

The first set of data provided insight into the decisions made in the formation of a new departmentand the implications for evolution across time. The second set of data provided the scientists’perspective on management decisions, as well helped to identify specific managerial practices thatconnect to particular scores on the third source of data, the research environment survey. Discussionof survey results in the s provided in this paper utilizes only the responses of the scientific staff of thedepartment. The one-half hour limit on the qualitative interviews of course means that only a fewissues about managerial practices could be explored. For this paper, we focused on the issues ofcomplexity and integration, internal growth and external recruitment, and research unit size andflexibility.

The survey utilized in this analysis has been administered and tested in a number of R&D settings(Jordan, 2005; Jordan et al., 2003; Jordan and Streit, 2003b), including other national laboratories. Thetable survey covers key attributes of organizational structure and management practices within theresearch environment which were identified and defined through an extensive literature review andinput from 15 focus groups that included bench scientists, engineers, and technologists, as well astheir managers, across various R&D tasks (Jordan et al., 2003). In total, 36 attributes in four areas wereidentified as most important to creating an environment that fosters excellent research, and thesurvey attempts to determine the health of a laboratory’s research environment by way of theresearchers’ perceptions. For the purposes of this case study, the survey offered a useful way ofmeasuring the perceptions of scientists in the S&T MD on various issues related to complexity ordiversity and integration. In addition, because the survey had recently been administered at theNATLAB, it provides comparative data to better understand the experience of the S&T MD.

4. Initial decisions: managerial practices involving recruitment

In this case study, the origin of the need for a separate department to conduct basic or fundamentalscience for improving the manufacturing processes resulted from the founding manager’s role in theanalysis group that oversaw the transfer of the manufacturing task to NATLAB. Gradually, he hadperceived that the basic science for manufacturing was underemphasized. In particular, much of thework concentrated on handling short-term solutions to problems in manufacturing rather thanaddressing the fundamental underlying or long-term issues, similar to that discussed by Cesaroni et al.(2004). Over time, the manager began to believe that not only a separate unit, but also a co-locatedunit within the manufacturing unit, would maintain a better balance between long-term and short-term problem solving. As one interviewee put it:

‘‘In the preceding analysis group there was not enough physics or chemistry. Furthermore, therewere not enough technologists. [The manager] Started building new techniques and proceduresthat went beyond standard quality control techniques.’’

Other individuals who had been at the NATLAB for some time commented that the analysis groupwas under-resourced. It is in this regard that S&T MD represents an interesting co-location effort. Ingeneral, the rationale behind co-location is to place different types of research activities in proximityto one another, to interact on a more regular and sustained basis, and to leverage expertise andcompetencies (Kahn and McDonough, 1997). While most of the co-location efforts that have beendiscussed in the R&D literature have involved integrating design or marketing units withinmanufacturing units, the S&T MD is different in that it involves the co-location of a basic science unitwithin a design/manufacturing department of the NATLAB.1 However, the rationale and perceivedbenefits of the S&T MD were consistent with those identified previously.

1 The manufacturing unit is one of several at the NATLAB, which focus on products that are one of a kind or produced in small

batches, where there is either little incentive or too much complexity for private firms.


An unexplored issue in the study of the formation of new research units is the implicit charter, thatis, the basic objectives of the new unit. The charter of the S&T MD was summarized in the followingmanner by one of the interviewees. To perform basic research that contributes to the manufacturingprocess the following:

� Improve the lifetime of the product;� Improve the quality control of the process;� Prevent any technical surprises.

Given these objectives, a series of initial decisions about recruitment of staff are necessary. Withoutbeing readily apparent, implied in the recruitment of staff are a series of decisions about thecomplexity or diversity of skills of the department and its connections to other departments withinlarge organizations such as NATLAB.2 The interviews with the manager allowed us to ask about theinitial decisions that were made when the department was formed. One of the more fascinating oneinvolves the question of the complexity of the research unit as measured by the number of differentdisciplines and kinds of technologists and beyond this the diversity of the research portfolio. Themanager made an unusual but conscious decision not to have all the complexity and diversity withinthe new department but instead to achieve this by funding projects outside the department and bycarefully having the new staff develop joint projects with members in other departments. He felt thathaving individuals outside the department involved would result in more complexity than simplyhiring double or triple the number of individuals. This also had the effect of reducing costs. But giventhis decision, an important challenge involved how to ensure commitment to the joint projects. In thisregard, he arranged to have 30–40% of their salary paid over a 3-year period to guarantee participationby people outside the department in the joint projects. This stability of funding is rare within thecultural context of this particular NATLAB and thus was a major kind of incentive.

Why is this a critical decision, one that should be emulated when other new research units areformed? A large literature in organizational sociology and a much smaller series of studies of researchorganizations indicates that as size increases, there is a tendency for internal differentiation to occur,that is, to create sub-groups that gradually create barriers to integration and grow into departments.In this instance, if more physicists or chemists or engineers had been hired within the unit, therewould be tendency for those of the same discipline to form sub-units, hurting integration across thesedisciplines. Furthermore, the human capital of all the researchers is increased more by having jointprojects across unit boundaries rather than the same number of individuals and projects within theunit because it provides more access to the diversity of projects – and thus the pool of knowledge,skills, and equipment – in the other units that are now attached to this unit (Mote, 2005). In otherwords, both internal and external scientific communication is maximized at the same time that costsare reduced.

Another initial decision in the formation of the department is the character of the people that arerecruited. First and foremost, the type of leader plays an important role (Von Zedwitz, 2003). In thisregard, the founding manager was internal and had a comprehensive understanding of theorganizational culture and significant technical credibility. Further, the manager had previously beenpart of an advanced analysis group, which gave him a considerable amount of tacit knowledge aboutthe kinds of problems and also the pros and cons of specific individuals within the analysis group.Interesting enough, he again strived for balance by recruiting some from this group and mixing themwith individuals who had just been hired into this national laboratory.

Obviously, given the multi-disciplinary needs of the department, specifically physics andchemistry as well as technologists who were skilled in building new kinds of test equipment, it wasimportant to hire individuals with the correct skill sets. But the manager specifically decided to hire adiversity of individuals classified according to their social styles (e.g. analytical vs. entrepreneurial).

2 One of the intellectual errors in the management literature is to treat all organizations, whether very large or very small, as

the same. Large organizations have divisions which are further disaggregated into departments. The problems of inter-

departmental relationships, especially for research, are somewhat akin to joint research ventures between separate

organizations.


But besides the diversity, he also searched for individuals that exhibited a certain amount of flexibility.In other words, he also sought to provide the basis for the integration of this diverse set of social styles.This kind of thinking about the selection of individuals is quite unusual and indicates the manager’skeen awareness of the need to have both diversity and integration within his department although hemay not have thought of these concepts as such. He was keenly aware that the competitiveness ofindividuals would hinder integration; hence, he identified those who would be more willing tocollaborate. Again, the decision to increase complexity by developing collaborative research projectsacross department boundaries also became a selection criterion for the individuals in the department.Those who liked this were more likely to want to be in the department. In addition, some of theinterviewees also stressed that the manager was continuously looking for smart people.

Another important dimension of diversity is a diverse research portfolio. To better determine thediversity of the research portfolio, we analyzed the responses on the research environment survey tothe question about a researcher’s portfolio. In the survey results, the scientists had a mean value of 4.5(of 5.0) on the attribute of an integrated and relevant research portfolio, which was considerablyhigher than their counterparts in either COLO that we studied or all of NATLAB, with correspondingmeans which are 3.9 and 3.7, respectively. How did the manager achieve this perception? First, it wasthe mission of the new department to integrate fundamental understanding and applications tomanufacturing. Also, while he encouraged each person to develop several projects in consultationwith people internal and external to the department, he guided them in the process, ensuring thatthere would be more integration across projects and, furthermore, that the long-term vision would befacilitated. These survey results suggest that indeed he was successful in cultivating an integratedportfolio of projects. If diversity of research is an important goal, the survey and interviews supportthat the manager’s guidance fostered a respectable distribution in the researchers’ portfolios ofresearch projects, even in the face of having only a small number of scientists in the department.

As discussed earlier, this diversity in the research portfolio is desirable if the aims of the researchare radical advances. The S&T MD is responsible for both short-term problem solving and searching fora long-term solution to a basic problem with the exotic material, which would be a revolutionarybreakthrough if achieved. As we have already suggested these are different ways of thinking andrequire different skill sets as well as disparate kinds of research projects.

5. Managerial practices relative to complexity and integration

Now we turn to two recurring themes in the literature on innovation, the importance of complexityand integration. Integrating a diverse set of individuals with quite different projects is more difficultthan it may seem. And as one might expect, in this department there were some conflicts that werestill being resolved. Besides conflicts over expectations, a major source of difficulty wascommunication between people trained in quite disparate disciplines and sub-disciplines. Thiswas especially true for tacit communication, that is, the unspoken assumptions involved in our modelsof thinking.

5.1. Individual dimensions of complexity: creativity and risk taking

One respondent stated that when she started working in the department, the manager gave hertime to think and explore various potential projects with a variety of people across the organizationbefore selecting her projects. She found that the manager was realistic about the amount of time ittook to develop ideas. At the same time, this did not mean that individuals were not pushed. Onerespondent mentioned that the manager did this behind the scenes. Another respondent stated thatthe manager pushed people to think on their own. One managerial practice that the manager himselfmentioned was his encouragement that each individual applied for highly competitive internal fundsfor research. He was proud of the initial success that his people had in this difficult competition.

Creativity and risk taking were encouraged because the individuals were asked to work on bothbasic and applied science, that is splitting their time between these two kinds of research activitiescreating dual goals in their work comparable to the ideas stressed in Pasteur’s Quadrant (Stokes,1997). One respondent said that the manager ‘‘allowed us to do creative thinking almost all the time.’’


In the research environment survey, there are two groups of indicators that reflect various aspectsof creating an environment in which creativity and risk-taking, and thus innovation, can flourish. Thefirst set of items focuses on power and economic resources (Table 1) that is whether or not power andmoney are provided to pursue new ideas. Specifically on item 21 (the authority to make decisions), thescientists report a mean of 4.8, and on item 3 (resources and freedom to pursue ideas) a mean of 4.4,both of which are higher than the NATLAB overall and in COLO, the other co-location experiment.

One important area where the scientists took initiative is the starting of conferences for developinga knowledge community relative to the set of problems that concern this department. Two scientists,one in the S&T MD and one in the Design Department, initiated a now biannual conference. Theseconferences allowed for collective learning and also helped to foster a community of practice whereresearchers could share knowledge as questions arose.

But in the second group of indicators, which deal more with the process of innovation, items 1 (a senseof challenge and enthusiasm), 2 (time to think and explore), and 4 (commitment to critical thinking), themeans were 3.8, 3.6, and 3.6, respectively; two of the means (1 and 4) are lower than in COLO and NATLAB(see Table 1). The explanation of senior mangers for a lower score on item 1, sense of challenge, was thatthe department had purposefully limited the scope of research in the department in order to impact themanufacturing process within 3 years and several in the department would have preferred focus onother areas and approaches to the problem. However, two of these items, 2 and 4, also asked respondentsto indicate their preference for how much time they would like to have allocated to these attributes. Inboth instances, they reported that they wanted more time, on average 12% more time for thinking andexploring and 20% more time for critical thinking. It is interesting that the S&T MD wanted much more ofthis even though in the qualitative interviews they reported that they had been given a great deal. Thismight reflect the problem of developing a whole new program of research in basic science andtechnology relative to manufacturing, as well as the department having only 13 months of existence atthe time the survey was completed.

5.2. Integration within the department and between departments

One of the key managerial practices for building integration between departments is the invitationto those involved in research projects that are located outside the department to make presentationsto the entire staff of the department. Most of the department staff found that these meetings wereintellectually exciting. The meetings became a critical managerial practice for fostering integrationbetween the various projects within the S&T MD, as the common knowledge pool of everyone wasraised during the presentations and the questions and answers that were produced by them. Onerespondent who had worked in NATLAB stated that there was much more cross-fertilization in thisnew department than in the previous two departments that he had worked in.

Another managerial practice is that the manager encouraged each individual who had some conflictwith either another staff member within the department or one outside the department, to come andtalk to him about it, including what might be potential solutions. In other words, rather than the managerproviding the solution, he tried to guide them to recognize what they themselves could do.

In addition to the integration with other departments achieved through joint research projects, themanager worked to develop close relationships with the design department as well. According to onerespondent, he did this by avoiding competition and broke down the barriers between the S&T MD and

Table 1Measures of creativity and risk-taking.

Attribute S&T MD COLO NATLAB

Mean True time (%) Mean True time (%) Mean True time (%)

Authority to make decisions 4.8 86 3.9 68 4.0 70

Resources/freedom to pursue new ideas 4.4 78 3.3 57 3.2 54

Sense of challenge & enthusiasm 3.8 66 4.3 76 4.0 70

Time to think & explore 3.6 62 3.1 52 2.7 43

Commitment to critical thinking 3.6 62 3.9 68 4.0 70

Comparisons made using only responses from scientists (MTS): S&T MD (n = 5), COLO 2004 (n = 25), and NATLAB 2003 (n = 522).


the Design Department by asking the right questions and talking to them about their work. Because ofthis effort, the S&T MD had some say in the selection of a basic researcher who was hired into theDesign Department. In turn, this lead to the development of the co-location of some researchequipment that benefited both the S&T MD and the Design Department. As another technique forbuilding integration with the design group, the manager ‘‘pulled strings so that a member of the groupwould share an office with a member of the [S&T MD]’’. This has created a considerable amount ofconsulting about the next generation of design.

Despite these managerial practices, the three items that relate the most directly to integration inthe research environment survey do not indicate that the S&T MD has high integration. In Table 2,these items are 6 (cross-fertilization of ideas), 5 (teamwork and collaboration), and 8 (good internalproject communication); the means are 3.6, 3.4, and 3.4, respectively. Two of these means are lower inthe S&T MD than in COLO. Again, on two of these items, 6 and 8, when asked how much more would bepreferred, the respondents indicated that they would like to have a great deal more. Specifically, 16%indicated a desire for more time for cross-fertilization of new ideas and a sizable 28% more time forinternal project communication.

While bothCOLOand NATLABalsodesiredmore timeforbothofthese activities, thepreferredamountwas not as much as in the S&T MD. One respondent explained these scores by saying that it is a diversegroup that ‘‘varies between A and Z’’ and that the problems of competition and different social styles stillremain. Put another way, the effort to integrate a diverse portfolio places an additional strain on both thecross-fertilization of ideas and internal project communication. Again, the newness of the departmentmight also provide further explanation on the larger preferred amount of time for these items.

The relatively lower scores are consistent with the reports that we received in the informalinterviews about various conflicts that still had not been resolved. But they may also reflect the factthat teamwork and good internal communication cannot be developed immediately in a newdepartment. Indeed, we might characterize the lower scores as the ‘‘growing pains’’ of having 13people learning a new way to work with each other and with external projects and a manufacturingdesign department. At the same time, more than one respondent indicated that the managerfrequently gives advice on how to handle differences with other staff. Furthermore, these problemshave to be placed in an historical context. One respondent who knew the situation both before andafter the formation of the S&T MD stated that disagreements in the analysis group and conflict withthe design and production groups were present prior to the formation of the new department.

6. Multi-dimensions of leadership

While research on leadership styles in R&D remain rare, a handful of studies provide useful guidesfor identifying key aspects of effective leadership (Hollingsworth, 2004; Hollingsworth et al., in press).These studies have identified that effective leaders often have three essential qualities beyond theusual managerial skills, such as those involved in securing appropriate funding and the maintenanceof schedules:

(1) long-term scientific vision;(2) cognitive mentoring of researchers;(3) emotional support of researchers.

These qualities were explored in our interviews and we address them in turn below.

Table 2Measures of integration.



Cross-fertilization of ideas 3.6 62 3.9 68 3.2 55

Teamwork & collaboration 3.4 58 3.9 68 3.9 69

Good internal project communication 3.4 58 3.4 59 3.5 60

Comparisons made using only responses from scientists (MTS): S&T MD (n = 5), COLO 2004 (n = 25), and NATLAB 2003 (n = 522).


6.1. Vision

During the informal interviews, each of the individuals commented on the long-term vision thatthe S&T MD manager had for the research department. One commented that the S&T MD manager sawdown the road and tried to get partnerships so that the department and its vision could extend into thefuture.

The research environment survey measures the amount of research vision in several differentways. For item 33 (clear research vision), the mean is 4.4, and on item 36 (invests in futureopportunities), the mean is also 4.4. On item 32 (good identification of new opportunities), the mean is4.2, and on item 27 (senior management champions foundational research), the mean is 4.0. On all ofthese items, the mean values are considerably higher than they are in COLO and NATLAB in general(see Table 3). The means respectively for COLO are 3.9, 3.8, 3.6, and 3.4 and for NATLAB, in the sameorder, they are 3.9, 3.4, 3.2, and 2.9.

However, despite these quite high scores, when asked if they wanted even more time spent on clearresearch vision and strategies, the S&T MD indicated that they did, although not as large as theincreases preferred for COLO and in NATLAB.

6.2. Cognitive mentoring

We consider cognitive mentoring akin to the concept of ‘‘perceived organization support’’, that is,the general perception by employees of being valued and cared about by an organization (Eisenbergeret al., 1986). In general, it has been demonstrated that higher perceptions of organizational support arepositively associated with high individual performance, organizational commitment, and innovation(Eisengerger et al., 1990). However, we would suggest that our notion of cognitive mentoring isqualitatively different from the formal, vertically downward types of communication that has oftenbeen emphasized (Oh et al., 1991). Rather, more informal, expressive or supportive types ofcommunication can also be a key to leadership effectiveness, and this seems true of the S&T MD. Asone respondent commented:

‘‘[the S&T MD manager] talks to us and he is not nay saying. In fact, he is an intellectualpromoter. He picks up on what you say and then provides input.’’

This perception was pervasive through the S&T MD, as many respondents indicated that thedepartment was an exciting place to be because of the manager’s emphasis on openness and learning.In this vein, one respondent stated that the S&T MD manager approaches discussions as a learning tool.Another respondent indicated that he was finally beginning to understand the whole scope of theproblem that needed to be solved because he was now involved and interacting with researchersfocused on other aspects of the research problem, and this was actively encouraged by the manager.This comment is consistent with previous findings that employees who perceive a higher level oforganizational support are more likely to contribute acquired new knowledge and skills and to theirprojects (Wayne et al., 1997).

More practically, these two aspects, mentoring and support, are demonstrated in the S&T MDmanager’s encouragement of individuals to take courses and even work for additional degrees.

Table 3Measures of vision.



Clear research vision & strategies 4.4 78 3.9 68 3.9 69

Invests in future capabilities 4.4 78 3.8 66 3.4 59

Good identification of new opportunities 4.2 74 3.6 62 3.2 53

Mgmt champions foundational research 4.0 70 3.4 58 2.9 48

Comparisons made using only responses from scientists (MTS): S&T MD (n = 5), COLO 2004 (n = 25), and NATLAB 2003

(n = 522).


Indeed, it was noted that the S&T MD manager did not have a problem with individuals missingwork to attend classes. Allowing people to increase their knowledge is not only a way of providingsupport, but it is also a method for demonstrating that they are valued. It is interesting to note thatthis is an important corporate strategy of IBM and other firms to foster loyalty to the firm(Michlitsch, 2000).

Again, there are several attributes on the research environment survey that tap into the amount ofperceived support. On item 20 (management adds value to work), the mean was 5.0 (a perfect score),which was considerably higher than in COLO and in NATLAB as a whole. Also scoring relatively high isitem 15 (good professional development), which had a mean of 4.6. The corresponding means forCOLO are 3.9 and 4.1 and for NATLAB as a whole they are 3.7 and 3.6 (see Table 4). When asked if thereshould be more time devoted to improved professional development none of the respondents in thethree entities felt that much more time should be allocated. Thus, despite the differences in means,there is not a felt need.

6.3. Emotional support

Although typically overlooked in management studies and training, emotional support can play animportant role in management. Emotional support refers to information that assures individualsbeliefs that they are cared for and connected to others (Cobb, 1976; Ganster et al., 1986), and has beenfound to positively influence organizational outcomes such as organizational commitment andturnover (Abraham, 1999) and job satisfaction (Schaubroeck et al., 1989). On this topic, a fewcomments from the interviews are particularly interesting to relate. For instance, one respondent feltit was important to note that the S&T MD manager showed emotion in his understanding of peopleand their personalities. And in the same vein, one respondent suggested that the S&T MD manager was‘‘good at figuring out where people are coming from’’ or what might be called an empathic quality.Another observed that when there was success, the S&T MD manager showed the progress to everyoneso that there was public recognition of the entire department’s accomplishments. Although these areonly anecdotal observations, we would argue that they underscore an important area for additionalstudies in R&D leadership.

6.4. Other managerial practices

In addition to the three dimensions mentioned above, two other dimensions that aretraditionally considered as more ‘‘managerial’’ are also important: (1) the ability to garnerresources and (2) being able to work well with others. As one respondent who had considerableexperience in the NATLAB indicated, the S&T MD manager knew a lot of people and, perhaps morefundamentally, how to engage them in research planning. On the research environment survey, themean of the scientists’ responses for stable funding (item 11) is 4.6, considerably higher than ineither COLO or NATLAB as a whole for 2003. This is likely influenced by the guarantee of multi-yearfunding. On the item relating to good internal allocation of research funds (item 26) the mean is 4.0.For COLO, the means are 3.3 and 3.5, respectively, and the corresponding means for NATLAB are 3.1and 3.0 (see Table 5). This could be because many of the NG S&T staff was new to the process ofcompeting for internal project funding and this process is mostly outside the influence of their ownmanager.

Table 4Measures of perceived support.



Management adds value to work 5.0 90 3.9 68 3.7 65

Good professional development 4.6 82 4.1 73 3.6 61

Comparisons made using only scientists’ responses (MTS): S&T MD (n = 5), COLO (n = 25), and NATLAB (n = 522). Rank indicates

relative ordering among 36 attributes based on percentage time true within each category.


7. Discussion and conclusion

This paper was aimed at exploring in greater detail the formation, managerial practices andleadership styles of a unique co-location effort in R&D. The interviews and survey data allowed us tohighlight how the S&T MD and its manager had put in place processes and practices designed toprovide a balance between diversity and integration. As is clear from the research environmentsurvey, the problems of integration had not yet been solved at least by the measures of our interviewsand also the research environment survey. The literature suggests integration is not solved quickly. AsDanilovic and Winroth (2005) discuss, managers of cross-functional teams typically need to allowsome time to harmonize the organizational structure, workflow, and social and cognitive-leveldistances that people from different functions and specialties bring in with them. It should be notedthat this unit had been in existence for only 13 months and many of the people recruited into the unit,as well as the participating members outside the department, were still in the process of getting toknow each other.

What lessons does this case study offer R&D managers? With regard to the objective of maintaininga balance between complexity and integration so as to maximize innovation and the benefits of co-location, we would make the following observations. One especially interesting managerial practicewas the selection of people who scored high for flexibility. In NATLAB, courses were widely given on atypology of social styles and the manager used this as a way of identifying the appropriate individuals.Flexibility was a much needed attribute, as the scientists who were recruited had different socialstyles, disciplinary backgrounds, and levels of experience, and in their levels of experience both withinand outside NATLAB. Hence, the flexibility of each individual was essential in the integration of adiverse group of people.

Because some projects were funded outside the department, it was important to quickly integrateand gain the commitment of these individuals to the new unit as well as their own. This was done byproviding guaranteed funding at about one-third time for 3 years, which, in the context of funding atthe NATLAB, presented a stable source of funding. With regard to integration, several practices wereadopted: monthly presentations of the joint projects by the members outside the department and insome cases joint sharing of offices.

But balancing diversity and integration was not confined within the organizational boundaries ofthe S&T MD. To increase the amount of skills and resources available for the S&T MD, the managerpursued a strategic effort to recruit additional diversity through joint research projects acrossdepartment lines and by extension with other organizations. On the positive side, this allows forgreater flexibility as skill needs shift across time but on the negative side, this increases the problemsof commitment and integration. But the practices discussed above were useful in ameliorating anypotential issues.

Finally, the leadership of the S&T manager was critical. Because the manager played a key role indesigning and establishing the unit, he had considerable say in how the unit was to operate. Further,his considerable background in the NATLAB gave him invaluable insights on what would work andwhat would not work within the organizational framework of the laboratory. In addition, theleadership style of the manager was highly effective in crafting a vision for the unit and employees andwhich also provided the often overlooked cognitive and emotional support needed to gain the loyaltyand commitment of employees.

Table 5Measures of other managerial practices.



Sufficient, stable project funding 4.6 82 3.3 56 3.1 52

Good allocation of internal funds 4.0 70 3.5 59 3.0 50

Comparisons made using only scientists’ responses (MTS): S&T MD (n = 5), COLO (n = 25), and NATLAB (n = 522). Rank indicates

relative ordering among 36 attributes based on percentage time true within each category.


As Kahn and McDonough (1997) point out in their study, while the record of co-location efforts hasbeen mixed, the performance of co-location efforts are typically dependent on the use of collaborationand integration mechanisms. As this study illustrates, the use of such mechanisms is a function of boththe initial decisions in the formation of the co-location effort as well as the leadership style of theprimary manager. Given the paucity of research on co-location in general, a greater level of empiricalresearch is needed to evaluate the impact of these efforts. In particular, an interesting question forfuture research is whether research organizations can select or train the type of leaders to successfulmanage a co-location effort. As this study suggests, the ability of managers to provide cognitivementoring and emotional support should be an important part of any co-location manager’spersonality.

As innovation becomes increasingly critical to a firm’s survival and success, the ability tosuccessfully manage cross-functional efforts and research complexity will be essential. Accordingly,more in-depth knowledge of co-location and similar efforts is vital to understanding the innovationprocess fully. The findings of the present study take us a step in that direction.

Acknowledgements

The authors gratefully acknowledge the support of the U.S. Department of Energy Office of Scienceand the Science, Technology and Engineering Foundations Strategic Management Unit at SandiaNational Laboratories. This research has been performed under contract with Sandia NationalLaboratories, DOE contract DE-AC04-94AL85000. Sandia is operated by Sandia Corporation, asubsidiary of Lockheed Martin Corporation.

References

Abraham, R., 1999. The impact of emotional dissonance on organizational commitment and intention to turnover. Journal ofPsychology 133, 441–455.

Allen, T.J., 1977. Managing the Flow of Technology. MIT Press, Boston, MA.Cesaroni, F., Di Minin, A., Piccaluga, A., 2004. New strategic goals and organizational solutions in large R&D labs: lessons from

Centro Ricerche Fiat and Telcom Italia Lab. R&D Management 34 (1), 45–56.Cobb, S., 1976. Social support as a moderator of life stress. Psychosomatic Medicine 38 (5), 300–314.Cordero, R., Farris, G.F., DiTomaso, N., 2004. Supervisors in R&D laboratories: using technical, people, and administrative skills

effectively. IEEE Transactions on Engineering Management 51 (1), 19–30.Danilovic, M., Winroth, M., 2005. A tentative framework for analyzing integration in collaborative manufacturing network

settings: a case study. Journal of Engineering and Technology Management 22, 141–158.Eisengerger, R., Fasolo, P., Davis-LaMastro, V., 1990. Perceived organizational support and employee diligence, commitment,

and innovation. Journal of Applied Psychology 75 (1), 51–59.Eisenberger, R., Hutchison, S., Sowa, D., 1986. Perceived organizational support. Journal of Applied Psychology 71, 500–507.Ganster, D., Fuilier, M., Mayes, B., 1986. Role of social support in the experience of stress at work. Journal of Applied Psychology

71, 102–110.Green, S.G., 1995. Top management support of R&D projects: a strategic leadership perspective. IEEE Transactions on

Engineering Management 42 (3), 223–232.Hage, J.T., 1999. Organizational innovation and organizational change. Annual Review of Sociology 25, 597–622.Holland, S., Gaston, K., Gomes, J., 2000. Critical success factors for cross-functional teamwork in new product development.

International Journal of Management Reviews 2 (3), 231–259.Hollingsworth, J.R., 2004. Institutionalizing Excellence in Biomedical Research: The Case of Rockefeller University Found in

Creating a Tradition of Biomedical Research: Contributions to the History of the Rockefeller University. RockefellerUniversity Press, New York.

Hollingsworth, J.R., Hage, J.T., Hollingsworth, E.J., in press. The Search for Excellence: Organizations, Institutions, and MajorDiscoveries in Biomedical Science. New York: Cambridge University Press.

Johansen, K., Comstock, M., Winroth, M., 2005. Coordination in collaborative manufacturing mega-networks: a case study.Journal of Engineering and Technology Management 22, 226–244.

Jordan, G.B., Hage, J.T., Mote, J.E., Hepler, B., 2005. Investigating differences among R&D projects and implications for managers.R&D Management 35 (5), 501–511.

Jordan, G.B., 2005. What matters to R&D workers? Research Technology Management 48 (3), 23–32.Jordan, G.B., Streit, L.D., Binkley, J.S., 2003. Assessing and improving the effectiveness of national research laboratories. IEEE

Transactions in Engineering Management 50 (2), 228–235.Jordan, G.B., Streit, L.D., 2003b. In: Shapira, P., Kuhlmann, S. (Eds.), Recognizing the Competing Values in Science and Technology

Organizations: Implications for Evaluation Learning from Science and Technology Policy Evaluation. Edward Elgar,Cheltenham, UK and Northampton, MA, p. 2003.

Kahn, K.B., McDonough III, E.F., 1997. An empirical study of the relationships among co-location, integration, performance, andsatisfaction. Journal of Product Innovation Management 14 (3), 161–178.


Leenders, R.T.A.J., van Engelen, J.M.L., Kratzer, J., 2003. Virtuality, communication, and new product team creativity: a socialnetwork perspective. Journal of Engineering and Technology Management 20 (1-2), 69–92.

McDonough III, E.F., Barczak, G., 1991. Speeding up new product development: the effects of leadership style and source oftechnology. Journal of Product Innovation Management 8 (3), 203–211.

Michlitsch, J.F., 2000. High-performing, loyal employees: the real way to implement strategy. Strategy & Leadership 28 (6),28–33.

Mote, J.M., 2005. R&D ecology: using 2-mode network analysis to explore complexity in R&D environments. Journal ofEngineering and Technology Management 22 (1), 93–111.

Nihtila, J., 1999. R&D–Production integration in the early phases of new product development projects. Journal of Engineeringand Technology Management 16 (1), 55–81.

Nooteboom, B., 2000. Learning and Innovation in Organizations and Economies. Oxford University Press, Oxford.Nooteboom, B., 1999. The dynamic efficiency of networks in interfirm networks. In: Grandori, A. (Ed.), Organization and

Industrial Competitiveness. Routledge, London, pp. 91–119.Oh, K., Kim, Y., Lee, J., 1991. An empirical study of communication patterns, leadership styles, and subordinate satisfaction in

R&D project teams in Korea. Journal of Engineering and Technology Management 8 (1), 15–35.Schaubroeck, J., Cotton, J.L., Jennings, K.R., 1989. Antecedents and consequents of role stress: a covariate structure analysis.

Journal of Organizational Behavior 10, 35–58.Sicotte, H., Langley, A., 2000. Integration mechanisms and R&D project performance. Journal of Engineering and Technology

Management 17 (1), 1–37.Stoker, J.I., Looise, J.C., Fisscher, O.A.M., De Jong, R.D., 2001. Leadership and innovation: relations between leadership, individual

characteristics and the functioning of R&D teams. International Journal of Human Resource Management 12 (7), 1141–1151.

Stokes, D.E., 1997. Pasteur’s Quadrant: Basic Science and Technological Innovation. Brookings Institution Press, Washington, DC.Susman, G.I., Majchrzak, A., 2003. Editorial: Research issues in knowledge management and virtual collaboration in new

product development: an introductory essay. Journal of Engineering and Technology Management 20, 1–5.von Corswant, F., Tunalv, C., 2002. Coordinating customers and proactive suppliers: a case study of supplier collaboration in

product development. Journal of Engineering and Technology Management 19, 249–261.Von Zedwitz, M., 2003. Initial managers of international R&D laboratories. R&D Management 33 (4), 377–393.West, M.A., 2004. Effective Teamwork: Practical Lessons from Organizational Research, 2nd ed. BPS Blackwell, Malden, MA.Wuyts, S., Colombo, M.G., Dutta, S., Nooteboom, B., 2005. Empirical tests of optimal cognitive distance. Journal of Economic

Behavior and Organization 58 (2), 277–302.Wayne, S.J., Shore, L.M., Liden, R.C., 1997. Perceived organizational support and leader-member exchange: a social exchange

perspective. The Academy of Management Journal 40 (1), 82–111.


Documents

Designing and facilitating collaboration in R&D: A case study