The Organizational Life of an Idea: Integrating Social Network, Creativity and Decision-Making Perspectives

The Organizational Life of an Idea: IntegratingSocial Network, Creativity and Decision-MakingPerspectives*

Bob Kijkuit and Jan van den EndeRSM Erasmus University

abstract Existing theories on the influence of social networks on creativity focus on ideageneration. Conversely, the new product development literature concentrates more on theselection of ideas and projects. In this paper we bridge this gap by developing a dynamicframework for the role of social networks from idea generation to selection. We applyfindings from creativity and behavioural decision-making literature and present an in-depthunderstanding of the sociological processes in the front-end of the new product developmentprocess. Our framework builds on the importance of mutual understanding, sensemaking andconsensus formation. The propositions focus on both network structure and content andhighlight the need to have strong ties and prior related knowledge, to incorporate decisionmakers, and to move over time from a large, non-redundant and heterogeneous to a smallerand more cohesive network structure. We conclude with a discussion on empirical validationof the framework and possible extensions.

INTRODUCTION

An effective ‘front end’ of the new product development process is important for theinnovative performance of firms. The front end (FE) is the process during which ideas areborn and further developed, ending with the go/no-go decision for the start of a project(Khurana and Rosenthal, 1998). Because of its importance, many firms put effort inorganizing the front end of their product development process (Kim and Wilemon,2002). A typical example is Shell that has created its ‘GameChanger’ suggestion andreview system (Hamel, 1999; Van Dijk and Van den Ende, 2002). The dominant viewbehind such endeavours is that firms should collect as many ideas as possible, organizean effective review and selection process, and provide appropriate feedback to ideasubmitters (Wheelwright and Clark, 1992).

In this paper we develop a social network perspective on the front end. We proposethat the networks of employees surrounding an idea affect the quality of that idea and its

Address for reprints: Bob Kijkuit, Management of Technology and Innovation, RSM Erasmus University,Room T10-34, PO Box 1738, 3000 DR Rotterdam, The Netherlands ([email protected]; [email protected]).

© Blackwell Publishing Ltd 2007. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UKand 350 Main Street, Malden, MA 02148, USA.

Journal of Management Studies 44:6 September 2007doi: 10.1111/j.1467-6486.2007.00695.x

chances of adoption. Our perspective is based on the view that ideas in the front end areadapted and improved before they are actually reviewed by management. We develop aframework on how the structure and content of the network of the idea, and its dynamics,affect the success of this adaptation process.

This paper builds on, and contributes to, the social network and new product devel-opment (NPD) literatures. Recently the social network literature has brought theoretical(Perry-Smith and Shalley, 2003) and empirical (Burt, 2004; Perry-Smith, 2006) expla-nations for the generation and, to a limited extent, the development of new ideas. Thegeneral assumption in these studies is that many infrequent social relations with peopleoutside your own social circle can provide people with unique information that, ifcombined, can lead to new creative insights. However, the focus of these authors is on theinitial phase of the FE process. We address the question how social networks develop inlater phases of idea development and evaluation. Amongst others, we show that networkconditions that enhance the novelty of the generated ideas may at the same time impedethe further development of the ideas and their actual transfer into projects (Reiter-Palmon and Illies, 2004).

The new product development (NPD) literature focuses more on the selection of ideasand the product development projects after selection (Brown and Eisenhardt, 1995;Wheelwright and Clark, 1992). With respect to selection this literature reflects a classicdecision-making perspective, in which decision makers are assumed to make consistentchoices that maximize the value for the firm, and that result from systematic assessmentsof all alternatives in comparison to predetermined criteria (Cooper et al., 1997; Rousselet al., 1991). As a consequence, the NPD literature has hardly addressed the socialprocesses involved in decision-making on new product development projects. We con-tribute to the NPD literature by applying a network perspective on the front end of thenew product development process, the need for which was already proposed by Van deVen (1986, p. 592) in his much cited paper on central problems in management ofinnovation: ‘As these ideas surface networks of individuals and interest groups gravitateto and galvanize around the new ideas. They, in turn, exert their own influence on theideas by further developing them . . .’.

The aim of this paper is to address the gap between the social network literature,which focuses on idea generation, and the NPD literature, which focuses on ideaevaluation, by developing a theoretical framework on the front end of the new productdevelopment process that concentrates on the transition from idea generation to evalu-ation. Our unit of analysis is the network of an idea, which we define as all people thatdiscuss a particular idea with each other. We build on recent trends within the socialnetwork literature to go beyond a pure structuralist view of networks by considering bothstructure and content of networks (Adler and Kwon, 2002) and by applying a dynamicnetwork perspective, which has thus far hardly been applied in the context of creativityand innovation (Perry-Smith, 2006). We distinguish three phases in the front end, thegeneration, development and evaluation phase, and we propose that the structure andcontent of the network of the idea should change over these phases for the network tocontribute to the quality of the idea. In developing our framework we apply insights fromcreativity (Kurtzberg and Amabile, 2001; Lubart, 2001; Woodman et al., 1993) andbehavioural decision-making literature (Daft and Lengel, 1986; Smith et al., 1994;

B. Kijkuit and J. van den Ende864

© Blackwell Publishing Ltd 2007

Weick, 1995). We particularly address the conditions to create the appropriate levels ofmutual understanding, sensemaking and consensus formation in the network of an idea.

Below we first define more clearly the front end of the product development process.Next we present an illustrative example of the generation, development and evaluationof an idea. The example concerns two ideas in the same unit, and shows how the networkof each idea can change over time, how the network and the idea affect each other, andhow this dynamics affect the result of the front end process. In the core part of the paperwe develop our framework. Finally, we make suggestions for possible empirical settings,data collection techniques and extensions to the current model.

A PROCESS VIEW OF THE FRONT-END

Process models of the NPD process are common in NPD literature and are often referredto as ‘stage-gate’ models. Extensive models include all activities from ‘idea generation’ to‘post-implementation review’. In this study we explicitly concentrate on the pre-development activities, which refer to all activities from idea generation until thego/no-go decision to execute an NPD project (Khurana and Rosenthal, 1998). Authorsfocusing on these activities advocate that the FE process is geared towards reducingthe uncertainty surrounding an idea to a point where it meets with a firm’s set ofpre-determined selection criteria (Kim and Wilemon, 2002; Moenaert et al., 1995).

Process models are also common in the creativity and decision-making literatures.According to the creativity literature, the creative process consists of activities, such asproblem identification, problem construction, and response generation (Lubart, 2001).The decision-making literature identifies similar processes, but includes the decisionprocess in its models. Simon’s classic decision-making model (Simon, 1965) identifiedthree phases, namely intelligence, design and choice. In response to empirical studies,Mintzberg et al. (1976) later built on this model and highlighted that the decision-makingprocess is highly complex and dynamic, surrounded by both uncertainty and ambiguity(Cohen et al., 1972). Mintzberg et al. (1976) again identified three phases: the identifi-cation phase, which consisted of recognition and diagnosis routines; the alternativedevelopment phase, which consisted of search and design routines; and the selectionphase, which consisted of screening, evaluation-choice and authorization routines.

In FE models the analogy of three phases is also found in, for instance, the model ofCooper (1988), who distinguishes between idea generation, product definition andproject evaluation. Khurana and Rosenthal (1998) go further and identify various sub-processes such as opportunity identification, project strategy formulation and projectpreplanning. For this paper we follow the analogy of three main phases and define thefront-end of new product development to consist of three phases, namely ‘idea genera-tion’, ‘idea development’ and ‘idea evaluation’.

The most important activities in the idea generation phase are problem identification,problem structuring and idea formulation (Khurana and Rosenthal, 1998; Leifer et al.2000; Schwenk, 1984). This phase involves recognizing gaps or flaws with the currentstate of thinking (Lubart, 2001), which is often the result of questioning the status quo, theneed to solve a problem or dissatisfaction with the current state of affairs (Dasgupta,1996), resulting in an initial creative idea. We would like to stress here that problem

Organizational Life of an Idea 865

© Blackwell Publishing Ltd 2007

identification and structuring are not always explicit, since the generation of an ideaoften takes place on the ‘fringe of consciousness’ (Dasgupta, 1994, p. 34).

In the development phase, response generation (Amabile, 1996) and concept devel-opment (Urban and Hauser, 1993) are the most important activities. During this phase,the idea moves from a one-liner into a detailed proposal. People that generated the ideamay dive into relevant literature or consult colleagues and friends to clarify key issues.This may lead to exploring alternatives and searching in new directions, making the ideamore robust and perhaps even resulting in a redefinition of the idea.

During the idea evaluation phase, the most important activities are screening (Cooper,1988) and decision making (Frederickson and Mitchell, 1984). The idea is alreadydescribed in detail and will only be refined on minor details. The evaluation will be basedon the decision maker’s personal opinion and in part on information provided byrelevant experts and their management peers. The most important groups of decisioncriteria in the NPD literature refer to market prospects, technological feasibility andcompany fit (Cooper et al., 1997; Roussel et al., 1991). In this paper, we speak of‘decision makers’ when we refer to the people who have the authority to make orparticipate in a go/no-go decision on the idea. We speak of ‘other employees’ when werefer to all other employees in the organization who are involved in idea generation,development and evaluation. Moreover, as mentioned before and in line with previousresearch (Khurana and Rosenthal, 1998), we consider the front end to finish when thego/no-go decision has been made.

Finally, it is important to note that the phases during the front end are interdependentand not necessarily sequential (Khurana and Rosenthal, 1998). This has also beenhighlighted by Mintzberg et al. (1976) and Saunders and Jones (1990), who have empha-sized the importance of viewing the decision process as a dynamic, open-system processsubjected to interferences, feedback loops and dead ends. Further development of anidea can, for instance, lead to an almost completely new idea or a negative evaluation cansend an idea back into the idea development phase. This does not, however, change thefact that the three phases do represent the major phases that all ideas go through beforethey are considered for funding.

AN ILLUSTRATION OF SOCIAL NETWORKS IN THE FRONT END

To illustrate our approach, we use an example from the fast moving consumer goodsindustry and describe how social networks can play a role in the FE, drawing on findingsfrom the creativity and decision-making literature. This example concerns the FE in anR&D laboratory, but the framework developed in this paper also applies to other productand service development contexts.

Current Perspectives on Networks and Creativity

The current applications of network structure to creativity focus on idea generation andargue that ‘good’ ideas are the result of having non-redundant, heterogeneous contactsthat enable a person to generate ideas by combining diverse information (Burt, 2004;Perry-Smith, 2006; Perry-Smith and Shalley, 2003). Non-redundant contacts are

B. Kijkuit and J. van den Ende866

© Blackwell Publishing Ltd 2007

contacts which are only connected to the individual in question and not to each other.Heterogeneous contacts are contacts that represent different functional backgrounds orlevels of tenure (Perry-Smith, 2006). The underlying assumption is that non-redundantand heterogeneous contacts increase the range of skills, knowledge and perspectivesavailable to an individual or a group, positively impacting creative performance (Pelledet al., 1999). This reasoning is in line with general creativity theories (Amabile, 1983,1996; Woodman et al., 1993) and is also recognized by scholars in decision-makingliterature (Eisenhardt and Schoonhoven, 1990; Haleblian and Finkelstein, 1993).

At any given time numerous ideas will be floating around in a given R&D lab, onlysome of which will be accepted and turned into funded projects. The question is: Whyare some ideas funded and others rejected, and what role do networks play in thisprocess? Consider the following example at multinational ‘Food Inc.’. In one of FoodInc.’s central R&D labs a series of presentations are organized for R&D scientists aboutnew market development. Topics that are covered include obesity in western countries,intestinal diseases in developing countries and so on. Later, various R&D scientists fromFood Inc. present at the ‘new market development’ presentations attend a scientificconference on the future of microbiology. During the conference they learn about newresearch on the ability of bacteria to cure certain intestinal diseases, filter saltwater andconserve foods over a longer period of time. The research on intestinal diseases sparks anidea with one of the R&D scientists of Food Inc., who is currently working on a projectto incorporate digestion stimulating bacteria in butter without affecting smell, colour ortaste. Based on this idea he writes a proposal to incorporate ‘anti-intestinal disease’bacteria in butter for developing countries, which he puts up for review.

Mutual Understanding in Networks

Previous research (Burt, 2004; Perry-Smith, 2006; Perry-Smith and Shalley, 2003) wouldargue that the R&D scientist in the example above was able to come up with his idea byinteracting with non-redundant groups of people. Although this view provides a com-pelling argument, creativity research would suggest that there are limitations to theapplicability of that theory, especially in an R&D setting. This research stream empha-sizes the need for mutual understanding to enable individuals to build effectively andcreatively on diverse knowledge (Kurtzberg and Amabile, 2001; Mumford andGustafson, 1988).

Mutual understanding refers to the ability to understand and build on each other’sknowledge base. At the firm level this concept is similar to ‘absorptive capacity’ (Cohenand Levinthal, 1990). Empirical research from creativity and decision-making literaturesuggests that performance of groups consisting of diverse members requires mutualunderstanding (Miura and Hida, 2004; West and Anderson, 1996). This implies that innon-redundant networks a sufficient level of mutual understanding is crucial.

The aforementioned example can illustrate this point. It is, for instance, highly ques-tionable if people without a degree in biochemistry would recognize the potential of thescientific findings presented at the conference. Furthermore, it is also very likely that theR&D scientist of Food Inc. will need to discuss the findings of the researchers in detail todetermine whether the ‘anti-intestinal disease’ bacteria would also work in a butter

Organizational Life of an Idea 867

© Blackwell Publishing Ltd 2007

solution. The example shows that non-redundant contacts alone do not suffice in anNPD context. The knowledge complexity, which is an essential part of the daily routine(Von Hippel, 1994), also requires a sufficient level of mutual understanding.

Uncertainty, Ambiguity and Sensemaking

Besides neglecting the importance of mutual understanding, previous research oncreativity and social networks has also not considered uncertainty to be relevant. ‘Wealso assume here that uncertainty and insecurity are relatively low’ (Perry-Smith andShalley, 2003, p. 94). This view contradicts with NPD literature, which focuses onuncertainty and advocates that whether an idea is accepted is dependent on the abilityof R&D scientists to reduce the uncertainty sufficiently to meet the selection criteria (Kimand Wilemon, 2002; Moenaert et al., 1995). Furthermore, decision-making literaturehighlights that ambiguity is equally important in a decision-making context (Daft andLengel, 1986; March, 1987; Thomas and Trevino, 1993). While uncertainty refers to alack of information, ambiguity refers to the existence of multiple and conflicting inter-pretations regarding an organizational situation (Daft and Lengel, 1986).

It is under circumstances of high uncertainty and ambiguity that sensemaking isconsidered a crucial process that enables organizational members to function (Weick,1995). It is defined as the process through which individuals develop meaning of theirsurrounding and act accordingly (Drazin et al., 1999). As Weick (1995) pointed out, thisprocess is not only about ‘reading’, but also about ‘shaping’ the environment, whichmakes it distinctly different from such activities as understanding and interpretation.Previous applications of sensemaking to the study of creativity and innovation haveessentially focused on how the creative individual or unit makes sense of the diverseinformation with which he or she is confronted (Dougherty and Hardy, 1996; Drazinet al., 1999; Hill and Levenhagen, 1995). Other research on sensemaking has focused onthe uncertainty and ambiguity that is faced by decision makers (Daft and Lengel, 1984,1986; Thomas and Trevino, 1993). Preferences of managers are often vague and con-tradictory and develop over time (March, 1987). We would therefore argue that ‘reading’and ‘shaping’ in the FE is a joint process of decision makers and other employeesinvolved in generating, developing and evaluating an idea.

Whether an idea is accepted is thus not only dependent on whether a generated ideameets some predetermined criteria, but also on the shaping of the idea and the criteriaduring the FE. This co-development may ensure that an idea fits current practice betterand is thus more easily adopted by risk avoiding managers (Christensen, 1997). In thislight, social networks are not only relevant for generating solutions, but maybe even moreimportantly, for identifying problems and opportunities that fit in the organization.

This point of ‘reading’ and ‘shaping’ the environment can be illustrated as follows.One of the other R&D scientists present at both the ‘new market development’ presen-tations and the scientific conference also considers the applicability of the researchfindings, but does not immediately write a proposal. Instead, she decides to contact oneof the R&D managers that organized the ‘new market development’ presentations anddiscusses the various scientific findings from the scientific conference. The ‘anti-intestinal’ bacteria and ‘new conservation method’ findings seem most interesting to the

B. Kijkuit and J. van den Ende868

© Blackwell Publishing Ltd 2007

R&D scientist, but the manager is wary of health regulations. The manager is moreinterested in the biological saltwater filter option, but since it is not a food product hedoes not see a business application within Food Inc. In an attempt to at least use one ofthe findings, the R&D scientist decides to look into the idea of the filter and contacts theresearch group from the scientific conference and peers from product design. After somediscussion with the researchers and her peers, the R&D scientist again contacts the R&Dmanager to discuss an idea for a cheap, disposable filter that can be sold along with ricebags already sold by Food Inc. The manager sees more potential and contacts a mar-keting and production manager responsible for developing markets. The managers areenthusiastic, but wary about the costs. The R&D manager relays this information to theR&D scientist, who writes a proposal and puts this up for review.

The example above highlights that social networks can reduce uncertainty and ambi-guity through sensemaking. The sensemaking process works through gathering informa-tion on the idea and the criteria of decision makers, and can result in adaptations of theidea and refinement of the criteria. The example also shows that networks, which includedecision makers and rely on repeated communication, can facilitate this sensemakingprocess.

Consensus Formation

A final aspect considered in decision-making literature, which may influence the socialnetwork structure is the need to create consensus. As Whyte (1989) notes: the first task ofall decision-making groups is to ‘produce consensus from the initial preferences of itsmembers’. Although creativity literature does not claim to address the evaluation deci-sion, several authors in this field have also emphasized that an idea is only valuable ifcollectively desired (Sternberg and Lubart, 1991) or has gained social acceptance(Simonton, 1989). This supports the view that an evaluation decision based on a certaindegree of consensus creates the required support in the organization for the actualproject.

Empirical research from the decision-making field has considered both demographicas well as group process variables. Demographic research suggests that consensus for-mation is negatively correlated with group size (Smith et al., 1994) and group diversity(Knight et al., 1999). Moreover, decision-making research on group processes has founda positive correlation between social integration (Lott and Lott, 1961; Smith et al., 1994)and consensus formation. The definition of social integration employed by Smith et al.(1994, p. 417) is based on the work of O’Reilly et al. (1989) and includes ‘attraction to thegroup, satisfaction with other members of the group and social interaction among groupmembers’. Communication frequency has also been considered to improve consensusformation (Lott and Lott, 1961), but more recent empirical findings have been mixed.Smith et al. (1994) and Ancona and Caldwell (1992) have, therefore, argued that fre-quent communication may only be crucial in conditions in which there is some diversityof insights or disagreement regarding decisions.

The main argument against the positive influence of intense communication in groupsis the risk of groupthink ( Janis, 1972). However, groupthink research has been incon-clusive and has led Smith et al. (1994) to conclude that social integration does not

Organizational Life of an Idea 869

© Blackwell Publishing Ltd 2007

negatively affect decision quality. More importantly, the whole issue of groupthink isunlikely to negatively affect the FE process if divergent thinking has already beenincorporated earlier in the FE process and if group composition is fairly temporal.

An example of actions facilitating consensus formation in a network can be illustratedas follows. To align the views and resolve any final conflict, the R&D scientist coulddecide to start organizing joint meetings in which the marketing, production and R&Dmanager and the product design specialists are all present. These meetings would enablethe R&D scientist to overcome differences, align the views and reach an agreement.

In short, in this section we have used an illustrative example to demonstrate thatprevious applications of social network theories to creativity require substantial exten-sions to be useful in the entire FE process, including the development and evaluationphase. More specifically, the complexity, uncertainty and ambiguity that are inherent inthe FE process make mutual understanding, sensemaking and consensus formation keyrequirements of social interaction for success in the FE. Moreover, the example showshow the idea is adapted and improved, and how the criteria are refined as a result ofsocial interaction. After a brief introduction to the social network literature, we will usethis refined understanding to formulate new propositions in the next section.

NETWORK STRUCTURE AND CONTENT

Using networks as a means to acquire resources has received a considerable amount ofattention in the management literature over recent years (see Borgatti and Foster, 2003for an overview). The classic literature on networks has employed a ‘structuralist’ per-spective advocating that the types of resources that can be acquired from a networkdepend on the structure of network relations (Adler and Kwon, 2002). This perspectivehas considered both network level and relationship level characteristics. Next to struc-ture, the network literature has recently showed increased attention for the influence ofthe content conveyed through ties on resource acquisition.

First, the structuralist perspective related to the network level reveals essentially twodistinct though not necessarily opposing views on how structure affects the benefits thatcan be obtained from social networks (Gargiulo and Benassi, 2000). The first view isbased on the ‘structural holes’ theory (Burt, 1992) and proclaims that a network ofnon-redundant contacts can provide an individual, the ‘tertius’, with information andcontrol benefits. Most scholars in the network literature that have focused on creativityand knowledge exchange have built on this or similar lines of reasoning (Burt, 2004;Cummings, 2004; Perry-Smith and Shalley, 2003). The second view on network struc-ture was introduced by Coleman (1988) who stresses the importance of social cohesion.This network concept refers to the extent to which relations between people are sur-rounded by third-party connections, the so-called ‘mutual friends’. In this networkstructure, control or action does not come from brokerage, which is the focus of Burt’s(1992) tertius strategies, but from trust, norms of cooperation and reputation (Coleman,1988; Obstfeld, 2005). A cohesive network structure creates benefits such as support,coordinated action and clear expectations (Coleman, 1988; Obstfeld, 2005; Reagans andZuckerman, 2001).

B. Kijkuit and J. van den Ende870

© Blackwell Publishing Ltd 2007

The structuralist perspective relating to the relationship level started with the pioneer-ing work of Granovetter (1973) on tie strength. However, over the years, the discussionamong network theorists has mainly focused on the network level and has only recentlystarted to reconsider the importance of tie strength. Literature relating to tie strength looksat the dyadic level and generally defines two extremes, weak and strong ties. Weak ties canprovide search benefits (Hansen, 1999), autonomy (Perry-Smith and Shalley, 2003) anddiverse information (Granovetter, 1973) at a low cost in terms of time and effort, makingit possible to maintain many ties. Strong ties, on the other hand, are characterized byfrequent two-way interactions (Hansen, 1999). These ties require more effort, but createtrust (Reagans and McEvily, 2003) and mutual understanding (Gilsing and Nooteboom,2005), which facilitate the transfer and construction of knowledge, especially morecomplex knowledge (Handley, 2006; Hansen, 1999; Roberts, 2006; Uzzi, 1999).

Finally, the ‘content perspective’ focuses on how the organizational position of actorsin a network and the knowledge of actors related to that position can influence the extentto which resources can be acquired. Consider Podolny and Baron’s (1997) study onorganizational mobility, in which they suggested that ‘perhaps it can be said that not allstructural holes are of the same color’ (p. 689). They suggested that whether ties spanningstructural holes are beneficial depends on the content conveyed through those ties. Otherexamples of network studies focusing on content rather than structure have consideredthe diversity of knowledge resulting from demographic differences or actors belonging todifferent functional disciplines, geographical locations or business units (Cummings,2004; Reagans and McEvily, 2003; Reagans and Zuckerman, 2001). The differencebetween a structuralist perspective and a content perspective is that the latter accountsfor the attributes of actors. A strong tie between two actors from the same unit may, forinstance, not be as useful as a strong tie between two actors of different units or betweena decision maker and an R&D scientist. Network content thus accounts for the work-related knowledge or influence that is the result of people’s background or that comeswith certain positions within organizations.

We thus do not only apply the ‘structuralist’ approach of social networks to thefront-end, but also include the role of network content.

PROPOSITIONS

Previous studies on the role of networks in creativity (Burt, 2004; Perry-Smith, 2006;Perry-Smith and Shalley, 2003) focus on the individual. Conversely, our unit of analysisis the network of an idea, which, as noted earlier, refers to all persons that discuss aparticular idea with each other.

In the initial phase this would refer to personal networks of employees. These personalnetworks affect the creation of the idea. Once an initial opportunity or idea is identified,we argue that it is the social interaction that is conducted with respect to an idea thatdetermines the further development of the idea and its evaluation. Taking the ‘network-of-an-idea’ as unit of analysis acknowledges that the network that is created around anidea may come from contacts of the idea generator(s), but can also expand to includeother people who get involved during the FE, such as the R&D manager and productdesigners in the example.

Organizational Life of an Idea 871

© Blackwell Publishing Ltd 2007

The previous studies on the role of social networks and creativity have advocated anddemonstrated that non-redundant networks of employees that include people fromdifferent functional backgrounds enhance creativity (Burt, 2004; Perry-Smith, 2006). Inline with these studies, we consider this network structure a prerequisite for high qualityideas. The propositions in this paper take non-redundant, heterogeneous network struc-tures as a starting point and describe how these networks should evolve during the FE toensure that rough ideas will evolve into high quality proposals that are worth funding.We discuss the propositions in the same sequence as the ‘example’ section starting withmutual understanding.

Mutual Understanding

The implication of creativity and decision-making literature on the dynamics of thenetworks in the NPD is the need for mutual understanding (Miura and Hida, 2004;Mumford and Gustafson, 1988; West and Anderson, 1996) in non-redundant andheterogeneous networks. During idea generation, mutual understanding is required torecognize the value of diverse and complex knowledge, as we illustrated in our example.Mutual understanding is crucial during idea development to actively transfer complexknowledge and thereby improve market prospects and technical feasibility, and ensurethat the idea fits within the company. In the final phase, mutual understanding is mainlyimportant to get support from decision makers.

We would thus argue that mutual understanding is important throughout all threephases. Moreover, the means through which mutual understanding is accomplishedshould come from two different sources, namely ‘prior related knowledge’ and strongties. Prior related knowledge refers to basic skills, a shared language (Boland andTenkasi, 1995) and knowledge of the latest scientific or technological developments in agiven field (Cohen and Levinthal, 1990). This aspect of mutual understanding is impor-tant to not only transfer, but also recognize the value of new information. It is thereforeimportant during both idea generation and development to improve the quality of anidea and thereby increase the chances of its acceptance:

Proposition 1a: Prior related knowledge in networks of employees and ideas during theidea generation and development phases increases the probability of idea acceptance.

Research on tie strength has shown that relatively weak ties suffice to search forknowledge, whereas strong ties support the actual transfer of knowledge (Hansen, 1999).As a consequence, we expect that strong ties are important during idea development.Moreover, during idea evaluation the focus shifts from the more complex and technicalconcept development to building support and overcoming any final conflict makingstrong ties essential. Finally, strong ties are especially important in situations where thediverse information comes from a minority which would otherwise not be acknowledgedor recognized (Nemeth and Staw, 1989).

Proposition 1b: Strong ties in networks of ideas during the idea development andevaluation phases increase the probability of idea acceptance.

B. Kijkuit and J. van den Ende872

© Blackwell Publishing Ltd 2007

Sensemaking

Another implication of decision-making theory concerns the need for ‘sensemaking’ in acontext filled with uncertainty and ambiguity (Weick, 1995). The key requirementfacilitating sensemaking is information processing involving the relevant actors, in thiscase employees involved in the idea and decision makers. Lean information-processingmechanisms, such as written reports, could suffice to reduce uncertainty (Thomas andTrevino, 1993). However, empirical research has shown that for ambiguity resolutionface-to-face interaction is crucial, which facilitates immediate feedback so that inter-pretations can be checked (Daft et al., 1987).

The creative foundation of an idea is laid in the generation phase of the FE. Sense-making, in this phase, mainly takes place between employees and possible outsidespecialists involved in the generation of the idea. Research in the innovation field hashighlighted that high management involvement will be counterproductive in this initialphase (Benner and Tushman, 2003). On the other hand, some basic sense of the keymarket developments in the organization can also facilitate creativity, as shown in theexample. Low management involvement, therefore, seems most productive.

Proposition 2a: Networks of employees in the generation phase that include weak ties todecision makers will increase the probability of idea acceptance.

In the development phase, ideas move ‘into’ the organization and go from an initialsketch to a more detailed project proposal (Smith and Reinertsen, 1991). In this phase,idea developing employees and decision maker interaction is most effective. The idea isstill wide open, enabling the employees involved in the development of the idea to assesshow the idea may be adjusted to fit with the need of the organization, while ‘shaping’(Weick, 1995) the future selection criteria. At the same time, decision makers are offereda sense of what is possible and allowed to convey what they consider critical andappropriate. This process is also likely to influence the way in which decision makers mayframe their criteria to be able to assess ideas.

Proposition 2b: Networks of ideas in the development phase that include strong ties todecision makers will increase the probability of idea acceptance.

While not downplaying the importance of interaction between decision makers andother employees involved in the idea in the evaluation phase, it will most likely focus on‘support gathering’, because the major outlines of ideas have already been ‘locked-in’.

Proposition 2c: Networks of ideas in the evaluation phase that include strong ties todecision makers will have no substantial impact on the probability of acceptance.

Consensus Formation

The final propositions in this paper relate to consensus formation. As we discussedearlier, a certain degree of consensus creates the required organizational support for the

Organizational Life of an Idea 873

© Blackwell Publishing Ltd 2007

actual project and can be created through social integration and informal communica-tion (Smith et al., 1994). As discussed in the section on network structure and tie strength,the network that is associated with these attributes is a cohesive or redundant structure.This network structure functions by means of reputation and group norms (Krackhardt,1999; Reagans and McEvily, 2003). These mechanisms ensure that people are morelikely to demonstrate cooperative behaviour and facilitate the development of grouprules or shared values. Support for a cohesive network structure is also found in theinformation-processing literature, which views group meetings as the most effectivemeans to reach a collective decision (Daft and Lengel, 1986). Such a structure optimizesthe chances that the range of perspectives regarding the idea are brought in line witheach other (Haleblian and Finkelstein, 1993) and with decision-making criteria, and thusthat the quality of the idea is further improved. The optimal network structure duringevaluation is thus more cohesive than the network structure that facilitates generation.The network surrounding an idea should evolve from a non-redundant structure withlittle joint discussions to a cohesive structure creating social integration, clear expecta-tions and a common frame of reference.

Proposition 3: An increasing level of cohesion in idea networks from the developmentto the evaluation phase increases the probability of idea acceptance.

Group size is also considered an essential element influencing group performance(Ancona and Nadler, 1989). Although the increase in the range of perspectives that isassociated with larger groups is considered positive (Haleblian and Finkelstein, 1993), itis also considered to create problems of coordination and control in decision-making(Seashore, 1977; Smith et al., 1994; Thomas and Fink, 1963). Large networks of ideaswill, therefore, be mainly important during development to get additional input andcriticism (Mizruchi and Stearns, 2001). Smaller groups, conversely, allow for a form ofteam work, which is considered critical for decision-making in NPD (Ancona andCaldwell, 1992). This is supported by empirical studies that found a negative indirecteffect of group size on informal communication and social integration, in line witharguments set forth by Seashore (1977) and Thomas and Fink (1963). Smaller networkswould therefore seem critical in the evaluation phase to frame the idea in a direction thatfacilitates the creation of consensus.

Proposition 4: Networks of ideas that decrease in size from development to the evalu-ation phase increases the probability of idea acceptance.

In short, our framework starts from the assumption from the literature that thegeneration and development of ideas benefit from non-redundant, heterogeneous net-works with many weak ties. Such networks can provide the diverse knowledge that iscrucial to the creative act of generating an idea. Our framework adds the need of priorrelated knowledge even during idea generation to guarantee mutual understanding.Second, the framework stresses that, during the development and evaluation phase,strong ties become important, including those to decision makers. Lastly, the frameworkadvocates that networks should develop into a smaller more cohesive structure fromdevelopment to evaluation. Such a network structure combined with strong ties provides

B. Kijkuit and J. van den Ende874

© Blackwell Publishing Ltd 2007

the conditions for the convergence of the idea towards its final form, and for the furtheradaptation of the idea to company requirements. The framework is summarized inTable I.

RESEARCH AGENDA

The obvious first step in further advancing the understanding of social interaction duringthe FE is an empirical validation. This is no small task since the process is dynamic andrequires longitudinal data. In this section we deal with possible empirical settings, datacollection techniques, alternative explanations, data collection issues and possible exten-sions to the current model.

Empirical Settings

The most obvious setting for empirical validation would be an R&D lab of an organi-zation in a highly competitive environment that relies on its employees to submit ideasand that employs some sort of formal decision-making gate that determines whether anidea will receive funding or not. Success in this case refers to whether an idea is fundedand to what extent. Other possible research settings could be NPD or ‘Business Devel-opment’ groups, which are either standalone or physically co-located with business units.

Data Collection Techniques

Collecting the data on the social interaction in the settings described above could be donethrough a survey questionnaire with ‘name generators’. Name generators are questionsthat focus on gathering names by asking respondents to list the people with whom theydiscussed the idea (see Marsden, 1990, 2004 for more details). Attribute data on contacts,such as unit membership, geographical location, educational expertise, could be gath-ered by additional questions or archival data from company records.

One of the problems of network surveys, and even more so in a longitudinal setting,is the time required of respondents. A way to overcome this problem is to combine thesurvey approach with the collection of ‘email’ data. This method of using emailexchanges to map social interaction is relatively new, but already shows interestingresults (Kossinets and Watts, 2006; Loch et al., 2003) and may prove most useful to test

Table I. The dynamics of structure and content of the network of an idea

Phase Structure Content

Size Cohesion Tie strength Prior related knowledge Decision makers involvement

Generation Large Low Weak High WeakDevelopment Medium Medium Strong High StrongEvaluation Small High Strong Low (Neutral)

Organizational Life of an Idea 875

© Blackwell Publishing Ltd 2007

the model outlined in this paper. In such a design, interaction between people in thenetwork is identified on the basis of email exchanges, for which fully automated softwareprograms exist. Key in such a design will be to filter out those email exchanges that relateto a specific idea. This could be done by scanning for keywords in email texts.

Email data provide some clear advantages. First, there is no interviewee bias. Peoplewill not over- or underestimate certain relations. Second, there is no problem withreactive measurement, which refers to the influence of interviewers mapping a networkon the dynamics of the network. This requires that there is no reasonable connectionbetween the use of the email data for this study and the knowledge of such use by theparticipants in the network of the idea. The obvious disadvantage of email data is theconstruct validity. Measuring social interaction by mapping email exchange will notcover all social interaction relating to a project proposal. Furthermore, emails that do notcontain the keywords or phrases may not be detected. To ensure that email exchange isa good proxy for social interaction, researchers should conduct the research in a settingthat relies heavily on email exchange, such as high-tech or computer industries.

Data Collection Issues

One of the most important issues to deal with is to prevent a ‘survivor bias’ (Singeltonand Straits, 2004), referring to an overrepresentation of successful projects in a sample,which could be especially problematic if the ideas are selected ex-post. This could beprevented if companies employ some sort of archival system in which all submittedproposals are recorded. However, ideas that are abandoned before being submitted arestill not included. To resolve this problem one could consider alternative researchdesigns, such as on-site field studies or ethnographies (Sutton and Hargadon, 1996).

The advantage of these alternative research designs is that ideas can be detected asthey emerge, which does not only prevent a survivor bias, but also allows researchers tomeasure various dimensions of the initial idea and study these over time. Examplesinclude: the perceived initial market prospects, technical feasibility, and company fit andtheir changes over time. Such a longitudinal approach allows one to accurately monitorthe dynamics of networks and assess the extent to which the interaction focused ontechnical, business or even political issues. Finally, a-priori selection would prevent thehalo effect (Pedhazur and Schmelkin, 1991), referring to the inclination of people toemphasize ex-post their relation with successful ideas and understate their involvementin unsuccessful ideas.

Extensions

Empirical validations of the model can also extend the model. One interesting extensionis to distinguish between different types of ideas, for instance between radical andincremental ideas (Garcia and Calantone, 2002). The involvement of people frommarketing units will be more important for incremental ideas than for radical ideas, sincecustomers and marketing representatives have difficulty judging the value of radical ideas(Song and Montoya-Weiss, 1998). Moreover, the involvement of current customers mayeven frustrate the generation of radical ideas (O’Connor, 1998). Other issues surround-

B. Kijkuit and J. van den Ende876

© Blackwell Publishing Ltd 2007

ing more radical ideas are the higher levels of uncertainty and ambiguity, the need toovercome internal cultures pressuring people to pursue low risk, incremental innovations(Dougherty, 1992) and the difficulty to convey more complex ideas to decision makers(Reagans and McEvily, 2003). These issues may imply that stronger ties are moreimportant to not only develop, but also generate radical ideas.

Another extension could consider the effect of perceived relational risk on the degreeof social interaction in the front end. Relational risk can refer to the risk people run of lossof face and reputation, or even to promotion and raise, when exchanging knowledge orasking questions (Edmonson, 1999). Relational risk can also refer to lock-in risk, whichprevents them from investing time in a relation to build mutual understanding (Bogen-rieder and Nooteboom, 2004). This might imply that in organizations with a competitiveculture, strong ties and a larger power base are more important. Finally, this extensioncould also be dynamic by not considering the networking effect on the idea, but also onthe people involved. Building on the interactionist perspective advocated by Woodmanet al. (1993), it would be interesting to see how, for instance, ‘networking experience’ onprevious ideas influences people’s motivation and networks on new ideas.

DISCUSSION AND CONCLUSION

In this paper we have developed a dynamic framework for the development of the socialnetworks of ideas in the front end of the innovation process in the firm. According to ourframework, networks of ideas should evolve from a non-redundant, heterogeneousstructure with many weak ties and a weak degree of decision maker involvement, into asmaller more cohesive network with, in the development phase, stronger ties to decisionmakers. Such a network provides the conditions for the convergence of the idea devel-opment towards its final form, and for the further adaptation of the idea to companyrequirements. Our framework furthermore requires that the successful generation anddevelopment of ideas need sufficient prior related knowledge between members of thenetwork to guarantee mutual understanding, and strong ties with decision makers duringthe development and evaluation phases.

Our framework extends the existing social network literature by addressing the ques-tion of how social networks should develop in the later phases of idea development andevaluation. The framework is based on findings from creativity and decision-makingliterature, which advocate the need for mutual understanding, sensemaking and consen-sus formation. Furthermore, the framework employs a dynamic perspective, which hashardly been applied in the context of creativity and innovation (Perry-Smith, 2006).Whereas prior research has emphasized the dynamics of the structure of networks forthe individual (Perry-Smith and Shalley, 2003), we focus on the network of the idea, theimportance of which was noted by Kurtzberg and Amabile (2001), and propose that thedynamics of the structure and contents of networks affect the development and evalua-tion of the idea itself. Finally, in line with recent trends (Adler and Kwon, 2002), theframework goes beyond a pure structuralist view of networks by adding the contentdimension of networks, particularly the need for prior related knowledge and intensecommunication to facilitate the exchange of diverse and complex knowledge and align-ment with company requirements.

Organizational Life of an Idea 877

© Blackwell Publishing Ltd 2007

Our study contributes to the NPD literature by developing a framework on how socialnetworks can play a key role in the front end process. Moreover, our framework employsan adaptation perspective, unlike most scholars and managers who have considered thefront end mainly from a selection perspective. Previous research has advocated the needto generate as many ideas, good and bad ones, as possible (Wheelwright and Clark,1992), and to select the best ones from this set. Not only does this conceptualization of thefront end rely on rather unrealistic assumptions with respect to the prognostic anddiscriminating capabilities of managers to find the few very good ideas from the pile ofmediocre ones, but such a process also requires a time consuming and costly system toprovide feedback to the idea submitters (Van Dijk and Van den Ende, 2002) and runs therisk of frustration among the many employees whose ideas are rejected.

Our description of the front end puts more emphasis on the adaptive effects originat-ing from proper networks of the idea over time. As argued, such adaptation may lead toan improved quality and company fit of the idea, and may increase the chances ofacceptance. Improper networks of ideas will generally not lead to improvement of thequality of the idea, and may thus already lead to cancelling out the idea by the networkparticipants before it is actually evaluated. This will be even more so if experienced ideagenerators, who recognize the insufficient dynamics of the network of the idea, are partof the networks. Summarizing, in evolutionary terms the traditional view of the front endis characterized by a selection perspective, whereas our description emphasizes theadaptative effects of the networks of ideas (Hodgson, 2001; Lewin and Volberda, 1999).

Our framework has its limitations. As noted earlier, the distinction between incremen-tal and radical ideas and the influence of perceived relational risk on the optimal networkcharacteristics need further research. Furthermore, the importance of the intrinsic valueof an idea relative to network effects should be included in future studies. Apart fromsuch extensions, the obvious priority should be on empirical validation, which wehighlighted in our discussion above on a research agenda.

The paper has managerial implications. Most importantly, managers acting in accor-dance with this framework should encourage idea generating employees to discuss theseideas with others before submitting the idea for review. Moreover, they should nothesitate to give some direction to the idea with an eye on company requirements. Moregeneric actions that management can take to improve the FE process include: reconsiderrecruitment policies, employ a job rotating system, and develop guidelines for proposalsthat stimulate networking.

First, company recruitment could take the networking potential of employees involvedin idea generation and development into consideration. This implies that such employeescould be recruited because of their extensive existing network in the scientific or orga-nizational world or for their networking skills. Research on this last topic is limited, butinitial work indicates that, for example, self-monitoring individuals, which have thewillingness and ability to monitor and control their self-expression in social situations, arebetter in networking (Kilduff and Tsai, 2003, pp. 81–4). Testing of such character traitscould be made part of the psychological tests that are now commonly used in recruitmentor assessment procedures.

Second, the benefits of job rotating systems in a research and NPD setting have beenadvocated in previous research (Griffin and Hauser, 1996), but not with the specific

B. Kijkuit and J. van den Ende878

© Blackwell Publishing Ltd 2007

purpose of ‘building’ networks. The degree of rotation could vary from regular projectrotation to R&D-marketing rotation (Griffin and Hauser, 1996). Although rotation haslimitations resulting from the lack of required specialist knowledge and training, rotationpolicies can give scientists and product developers a broader frame of reference andaccess to both scientific and technical as well as managerial knowledge and relations.

Lastly, management should create guidelines for the submission of ideas for newprojects that stimulate information gathering. Guidelines for the submission of a proposalcould, for instance, ask for evidence of input from a number of different experts bothinside and outside the company and at least one management sponsor. Such actions canform a start for management to harness the potential of networks in the front end.

NOTE

*We are most grateful to Geert Duysters, Michael Jensen, Bart Nooteboom, Raf Jans and the threeanonymous reviewers of JMS whose comments have substantially improved this paper, and the NetherlandsOrganization for Scientific Research (NWO) and Vereniging Trustfonds Erasmus Universiteit Rotterdamfor financial support.

REFERENCES

Adler, P. S. and Kwon, S. W. (2002). ‘Social capital: prospects for a new concept’. Academy of ManagementReview, 27, 17–40.

Amabile, T. M. (1983). The Social Psychology of Creativity. New York: Springer-Verlag.Amabile, T. M. (1996). Creativity in Context. Boulder, CO: Westview.Ancona, D. G. and Caldwell, D. F. (1992). ‘Bridging the boundary: external activity and performance in

organizational teams’. Administrative Science Quarterly, 37, 634–51.Ancona, D. G. and Nadler, D. A. (1989). ‘Top hats and executive tales: designing the senior team’. Sloan

Management Review, 31, 19–28.Benner, M. J. and Tushman, M. L. (2003). ‘Exploitation, exploration, and process management: the

productivity dilemma revisited’. Academy of Management Review, 28, 238–56.Bogenrieder, I. and Nooteboom, B. (2004). ‘Learning groups: what types are there? A theoretical analysis

and an empirical study in a consultancy firm’. Organization Studies, 25, 287–313.Boland, R. J. and Tenkasi, R. V. (1995). ‘Perspective making and perspective taking in communities of

knowing’. Organization Science, 6, 350–72.Borgatti, S. P. and Foster, P. C. (2003). ‘The network paradigm in organizational research: a review and

typology’. Journal of Management, 29, 991–1013.Brown, S. L. and Eisenhardt, K. M. (1995). ‘Product development: past research, present findings, and

future directions’. Academy of Management Review, 20, 343–78.Burt, R. S. (1992). Structural Holes. Cambridge, MA: First Harvard University Press.Burt, R. S. (2004). ‘Structural holes and good ideas’. American Journal of Sociology, 110, 349–99.Christensen, C. M. (1997). The Innovator’s Dilemma: When New Technologies Cause Great Firms To Fail. Boston,

MA: Harvard Business School Press.Cohen, W. M. and Levinthal, D. A. (1990). ‘Absorptive capacity: a new perspective on learning and

innovation’. Administrative Science Quarterly, 35, 128–52.Cohen, M. D., March, J. G. and Olsen, J. P. (1972). ‘A garbage can model of organizational choice’.

Administrative Science Quarterly, 17, 1–25.Coleman, J. S. (1988). ‘Social capital in the creation of human capital’. American Journal of Sociology, 94,

S95–120.Cooper, R. G. (1988). ‘Predevelopment activities determine new product successes’. Industrial Marketing

Management, 17, 237–47.Cooper, R. G., Edgett, S. J. and Kleinschmidt, E. J. (1997). ‘Portfolio management in new product

development: lessons from the leaders I–II’. Research Technology Management, 40, 16–28; 40, 43–52.

Organizational Life of an Idea 879

© Blackwell Publishing Ltd 2007

Cummings, J. N. (2004). ‘Work groups, structural diversity, and knowledge sharing in a global organization’.Management Science, 50, 352–64.

Daft, R. L. and Lengel, R. H. (1984) ‘Information richness: a new approach to manager informationprocessing and organization design’. In Staw, B. M. and Cummings, L. L. (Eds), Research in OrganizationalBehavior. Greenwich, CO: JAI Press.

Daft, R. L. and Lengel, R. H. (1986). ‘Organizational information requirements, media richness andstructural design’. Management Science, 32, 554–71.

Daft, R. L., Lengel, R. H. and Trevino, L. K. (1987). ‘Message equivocality, media selection, and managerperformance: implications for information systems’. MIS Quarterly, 11, 355–66.

Dasgupta, S. (1994). Creativity in Invention and Design. New York: Cambridge University Press.Dasgupta, S. (1996). Technology and Creativity. New York: Oxford University Press.Dougherty, D. (1992). ‘Interpretive barriers to successful product innovation in large firms’. Organization

Science, 3, 179–202.Dougherty, D. and Hardy, C. (1996). ‘Sustained product innovation in large, mature organizations: over-

coming innovation-to-organization problems’. Academy of Management Journal, 39, 1120–53.Drazin, R., Glynn, M. A. and Karanjian, R. K. (1999). ‘Multilevel theorizing about creativity in organiza-

tions: a sensemaking perspective’. Academy of Management Review, 24, 286–307.Edmonson, A. (1999). ‘Psychological safety and learning behavior in work teams’. Administrative Science

Quarterly, 44, 350–83.Eisenhardt, K. M. and Schoonhoven, C. B. (1990). ‘Organizational growth: linking founding team, strategy,

environment, and growth among U.S. semiconductor ventures, 1978–1988’. Administrative Science Quar-terly, 35, 504–29.

Frederickson, J. W. and Mitchell, T. R. (1984). ‘Strategic decision processes: comprehensiveness andperformance in an industry with an unstable environment’. Academy of Management Journal, 27, 399–423.

Garcia, R. and Calantone, R. (2002). ‘A critical look at technical innovation typology and innovativenessterminology: a literature review’. Journal of Product Innovation Management, 19, 110–32.

Gargiulo, M. and Benassi, M. (2000). ‘Trapped in your own net? Network cohesion, structural holes, and theadaptation of social capital’. Organization Science, 11, 183–96.

Gilsing, V. and Nooteboom, B. (2005). ‘Density and strength of ties in innovation networks: an analysis ofmultimedia and biotechnology’. European Management Review, 3, 179–97.

Granovetter, M. S. (1973). ‘The strength of weak ties’. American Journal of Sociology, 78, 1360–80.Griffin, A. and Hauser, J. R. (1996). ‘Integrating R&D and marketing: a review and analysis of the literature’.

Journal of Product Innovation Management, 13, 191–215.Haleblian, J. and Finkelstein, S. (1993). ‘Top-management team size, CEO dominance, and firm perfor-

mance: the moderating roles of environmental turbulence and discretion’. Academy of Management Journal,36, 844–63.

Hamel, G. (1999). ‘Bringing Silicon Valley inside’. Harvard Business Review, 77, 70–84.Handley, K., Sturdy, A., Fincham, R. and Clark, T. (2006). ‘Within and beyond communities of practice:

Making sense of learning through participation, identity and practice’. Journal of Management Studies, 43,641–53.

Hansen, M. T. (1999). ‘The search-transfer problem: the role of weak ties in sharing knowledge acrossorganization subunits’. Administrative Science Quarterly, 44, 82–111.

Hill, R. C. and Levenhagen, M. (1995). ‘Metaphors and mental models: sensemaking in innovative andentrepreneurial activities’. Journal of Management, 21, 1057–74.

Hodgson, G. M. (2001). ‘Is social evolution Lamarckian or Darwinian?’. In Laurent, J. and Nightingale, J.(Eds), Darwinism and Evolutionary Economics. Cheltenham: Edward Elgar, 87–118.

Janis, I. L. (1972). Victims of Groupthink: A Psychological Study of Foreign-Policy Decisions and Fiascos. Boston, MA:Houghton Mifflin.

Khurana, A. and Rosenthal, S. R. (1998). ‘Towards holistic “front ends” in new product development’.Journal of Product Innovation Management, 15, 57–74.

Kilduff, M. and Tsai, W. (2003). Social Networks and Organizations. London: Sage.Kim, J. and Wilemon, K. (2002). ‘Focusing on the fuzzy front end in new product development’. R&D

Management, 32, 269–79.Knight, D., Pearce, C. L., Smith, K. G., Olian, J. D., Sims, H. P. and Smith, K. A. (1999). ‘Top management

team diversity, group process, and strategic consensus’. Strategic Management Journal, 20, 445–65.Kossinets, G. and Watts, D. J. (2006). ‘Empirical analysis of an evolving social network’. Science, 311, 88–90.Krackhardt, D. (1999). ‘The ties that torture: Simmelian tie analysis in organizations’. Research in the Sociology

of Organizations, 16, 183–210.

B. Kijkuit and J. van den Ende880

© Blackwell Publishing Ltd 2007

Kurtzberg, T. R. and Amabile, T. M. (2001). ‘From Guilford to creative synergy: opening the black box ofteam-level creativity’. Creativity Research Journal, 13, 285–94.

Leifer, R., McDermott, C. M., O’Connor, G. C., Peters, L. S., Rice, M. and Veryzer, R. W. (2000). RadicalInnovation: How Mature Companies Can Outsmart Upstarts. Boston, MA: Harvard Business School Press.

Lewin, A. Y. and Volberda, H. W. (1999). ‘Prolegomena on coevolution. A framework for research onstrategy and new organizational forms’. Organization Science, 10, 519–34.

Loch, C. H., Tyler, J. R. and Lukose, R. (2003). ‘Conversational structure in email and face-to-facecommunication’. Unpublished paper, HP Labs.

Lott, A. J. and Lott, B. E. (1961). ‘Group cohesiveness, communication level, and conformity’. Journal ofAbnormal and Social Psychology, 62, 408–512.

Lubart, T. I. (2001). ‘Models of the creative process: past, present and future’. Creativity Research Journal, 13,295–308.

March, J. (1987). ‘Ambiguity and accounting: the elusive link between information and decision making’.Accounting, Organizations and Society, 12, 153–68.

Marsden, P. V. (1990). ‘Network data and measurement’. Annual Review of Sociology, 16, 435–63.Marsden, P. V. (2004). ‘Recent developments in network measurement’. In Carrington, P. J., Scott, J. and

Wasserman, S. (Eds), Models and Methods in Social Network Analysis. New York: Cambridge University Press.Mintzberg, H., Raisinghani, D. and Theoret, A. (1976). ‘The structure of “unstructured” decision processes’.

Administrative Science Quarterly, 21, 246–75.Miura, A. and Hida, M. (2004). ‘Synergy between diversity and similarity in group-idea generation’. Small

Group Research, 35, 540–64.Mizruchi, M. S. and Stearns, L. B. (2001). ‘Getting deals done: the use of social networks in bank

decision-making’. American Sociological Review, 66, 647–71.Moenaert, R. K., De Meyer, A., Souder, W. E. and Deschoolmeester, D. (1995). ‘R&D/marketing com-

munication during the fuzzy front-end’. IEEE Transactions on Engineering Management, 42, 243–58.Mumford, M. D. and Gustafson, S. B. (1988). ‘Creativity syndrome: integration, application, and innova-

tion’. Psychological Bulletin, 103, 27–43.Nemeth, C. J. and Staw, B. M. (1989). ‘The tradeoffs of social control and innovation in groups and

organizations’. Advances in Experimental Social Psychology, 22, 175–210.Obstfeld, D. (2005). ‘Social networks, the tertius iungens orientation, and involvement in innovation’. Admin-

istrative Science Quarterly, 50, 100–30.O’Connor, G. C. (1998). ‘Market learning and radical innovation: a cross case comparison of eight radical

innovation projects’. Journal of Product Innovation Management, 15, 151–66.O’Reilly, C. A., Caldwell, D. F. and Barnett, W. P. (1989). ‘Work group demography, social integration, and

turnover’. Administrative Science Quarterly, 34, 21–37.Pedhazur, E. J. and Schmelkin, L. P. (1991). Measurement, Design, and Analysis: An Integrated Approach. Hillsdale,

NJ: Lawrence Erlbaum Associates.Pelled, L. H., Eisenhardt, K. M., and Xin, K. R. (1999). ‘Exploring the black box: an analysis of work group

diversity, conflict, and performance’. Administrative Science Quarterly, 44, 1–28.Perry-Smith, J. E. (2006). ‘Social yet creative: the role of social relationships in facilitating individual

creativity’. Academy of Management Journal, 49, 85–101.Perry-Smith, J. E. and Shalley, C. E. (2003). ‘The social side of creativity. a static and dynamic social network

perspective’. Academy of Management Review, 28, 89–106.Podolny, J. M. and Baron, J. N. (1997). ‘Resources and relationships: social networks and mobility in the

workplace’. American Sociological Review, 62, 673–93.Reagans, R. and McEvily, B. (2003). ‘Network structure and knowledge transfer: the effects of cohesion and

range’. Administrative Science Quarterly, 48, 240–67.Reagans, R. and Zuckerman, E. W. (2001). ‘Networks, diversity, and productivity: the social capital of

corporate R&D teams’. Organization Science, 12, 502–17.Reiter-Palmon, R., and Illies, J. J. (2004). ‘Leadership and creativity: understanding leadership from a

creative problem solving perspective’. The Leadership Quarterly, 15, 55–77.Roberts, J. (2006). ‘Limits to communities of practice’. Journal of Management Studies, 43, 623–39.Roussel, P. A., Saad, K. N. and Erickson, T. J. (1991). Third Generation R&D: Managing the Link to Corporate

Strategy. Boston, MA: Harvard Business School Press.Saunders, C. and Jones, J. W. (1990). ‘Temporal sequences in information acquisition for decision making:

a focus on source and medium’. Academy of Management Review, 15, 29–46.Schwenk, C. R. (1984). ‘Cognitive simplification processes in strategic decision making’. Strategic Management

Journal, 5, 111–28.

Organizational Life of an Idea 881

© Blackwell Publishing Ltd 2007

Seashore, S. E. (1977). Group Cohesiveness in the Industrial Work Group. New York: Arno.Simon, H. (1965). The Shape of Automation for Men and Management. New York: Harper and Row.Simonton, D. K. (1989). ‘The chance-configuration theory of scientific creativity’. In Gholson, B., Shadish,

W. R., Neimeyer, R. A. and Houts, A. C. (Eds), The Psychology of Science: Contributions to Metascience.Cambridge: Cambridge University Press.

Singelton, R. A. and Straits, B. C. (2004). Approaches to Social Research. New York: Oxford University Press.Smith, K. G., Smith, K. A., Olian, J. D., Sims, H. P., O’Bannon, D. P. and Scully, J. A. (1994). ‘Top

management team demography and process: the role of social integration and communication’.Administrative Science Quarterly, 39, 412–38.

Smith, P. G. and Reinertsen, D. G. (1991). Developing Products in Half the Time. New York: Van NostrandReinhold.

Song, M. X. and Montoya-Weiss, M. M. (1998). ‘Critical development activities for really new versusincremental products’. Journal of Product Innovation Management, 15, 124–35.

Sternberg, R. J. and Lubart, T. I. (1991). ‘An investment theory of creativity and its development’. HumanDevelopment, 34, 1–31.

Sutton, R. I., and Hargadon, A. (1996). ‘Brainstorming groups in context: effectiveness in a product’.Administrative Science Quarterly, 41, 685–718.

Thomas, E. J. and Fink, C. F. (1963). ‘Effects of group size’. Psychological Bulletin, 60, 371–84.Thomas, J. B. and Trevino, L. K. (1993). ‘Information processing in strategic alliance building: a multiple-

case approach’. Journal of Management Studies, 30, 779–814.Urban, G. L. and Hauser, J. R. (1993). Design and Marketing of New Products, 2nd edition. Englewood Cliffs, NJ:

Prentice-Hall.Uzzi, B. (1999). ‘Social relations and networks in the making of financial capital’. American Sociological Review,

64, 481–505.Van de Ven, A. H. (1986). ‘Central problems in the management of innovation’. Management Science, 32,

590–607.Van Dijk, Ch. and Van den Ende, J. (2002). ‘Suggestion systems. Transferring employee creativity into

practicable ideas’. R&D Management, 32, 387–95.Von Hippel, E. (1994). ‘Sticky information and the locus of problem solving: implications for innovation’.

Management Science, 40, 429–39.Weick, K. E. (1995). Sensemaking in Organizations. Thousand Oaks, CA: Sage.West, M. A. and Anderson, N. R. (1996). ‘Innovation in top management teams’. Journal of Applied Psychology,

81, 680–93.Wheelwright, S. C. and Clark, K. B. (1992). Revolutionizing Product Development: Quantum Leaps in Speed, Efficiency

and Quality. New York: Free Press.Whyte, G. (1989). ‘Groupthink reconsidered’. Academy of Management Review, 14, 40–56.Woodman, R., Sawyer, J. and Griffin, R. (1993). ‘Toward a theory of organizational creativity’. Academy of

Management Review, 18, 293–321.

B. Kijkuit and J. van den Ende882

© Blackwell Publishing Ltd 2007