Project management in network organizations
DeGroote School of Business
As competition in the global marketplace strengthens, and technological change offers better communication infrastructures and collaboration tools through eBusiness technologies, firms are more frequently teaming with other firms that offer complementary expertise. The objective of forming these arrangements is to increase company capacity to organize and compete effectively. Due to the close inter-company relationships employed by such arrangements, they are known as network organizations. Strategically, a network organization is a long-term arrangement among distinct but related organizations that support the included organizations in gaining or sustaining competitive advantage (Jarillo, 1988). Partnerships within network organizations rely on synergies that allow collaborating companies to profit more by pooling complementary resources than by relying on their own independent operations.
Network organizations, sometimes referred to as virtual organizations or strategic alliances, come in a variety of forms and structures: joint ventures, minority equity alliances, joint R&D and production, co-marketing, licensing, long-term supply agreements, and consortia, among others. These forms differ primarily in the nature and closeness of their inter-organizational linkages and are determined by the strategic objectives of the participating firms. There has been a phenomenal growth in applications of network organizations during the last decade with, for example, the number of U.S. (United States) business alliances increasing at the rate of 25% per year during much of the 1990s (Harbison & Pekar, 1998). Network organizations now underlie the productive power of many industries, including automobiles, electronics, aerospace, pharmaceuticals, biotechnology, and apparel. Overlaying network organization structures reflect patterns of social relationships among sets of people, positions, groups, or organizations.
Emerging trends in organization and technology are changing the way projects are organized and managed, creating new challenges in project management research and practice. To take advantage of complementary expertise, projects in network organizations are frequently undertaken by joint teams formed by members from two or more of the organizations. In fact, network organizations are often formed with the primary objective of undertaking joint projects (Lam, 1997). We will refer to such projects as distributed network projects. As an example, construction projects have long been associated with such inter-organizational arrangements (Cheng, Li, Love, & Irani, 2001; Dahlgren & Söderlund, 2001), where formal agreements assign general contractors the linchpin responsibility to oversee and coordinate projects. In turn, project deliverables are contracted to other firms with the appropriate expertise. Such projects are also common in new product development (Huang, Mak, & Humphreys, 2003; Bal & Gundry, 1999), production process innovations (Wagner, 2003), and other arenas.
The objective of this paper is to explore and explain issues relevant to project management in network organizations and to contrast these with similar internal projects. We begin by examining the relative levels of interdependencies among organizations and classifying the relative complexities and risks that can arise from structures, uncertainties, and organizations when selecting and managing both multi-site internal projects and distributed network projects. We also identify techniques for mitigating such risks, and we note the differences between internal and distributed network projects that tend to magnify the risks of network projects. Inter-organizational relationships, knowledge exchange, and virtual project management are addressed, since these topics are highly relevant to distributed network projects. Related issues in project portfolio selection in network organizations are also covered. Finally, we summarize the issues that relate to project selection and management in network organizations.
Network organizations can be characterized as having complex adaptive systems, no central hierarchical control, and many agents acting in parallel. These organizations have an innate capacity of being adaptive, and thus avoid the tendency of complex hierarchical systems (large centrally controlled organizations) to approach long-term equilibrium and deteriorate in the face of innovative competitors (Pascale, 1999). If business networks are self-organizing, they can lead to multiple spontaneous innovations, where the locus tends to be on networks of learning, composed of companies, universities, and research and development centers. These networks of learning often readily exchange knowledge (Powell, Koput, & Smith-Doer, 1996). Specific examples are in the biotechnology and pharmaceutical industries. Evolving service industries such as call centers are also good examples (Archer, 2003). Network organizations span the range from self-organizing (characterized by spontaneous innovation) to networks that are carefully planned and organized for the purpose of implementing specific projects or processes in new product development, supply chain management, construction, or other joint endeavors. In this paper we will focus on the planned and organized forms of network organization.
A critical determinant in managing distributed network projects is the degree of organizational interdependence among the collaborating organizations. For such comparisons, the three forms suggested by Thompson (1967) are useful:
- i) Pooled or generalized interdependence, where each partner renders a discrete contribution to the whole, and each is supported by the whole. Coordination of activities is through standardization of inputs, and only a broad alignment of the partners towards a joint objective is required. An example is the development of standards or bodies of knowledge by a central body, or the creation of a database to support a frequently asked question (FAQ) online system. Provided that all contributors use the same format, there is no need to organize their contributions either in time or space.
- ii) Sequential interdependence, where the activities of each partner are distinct and serial, so the activities of one partner precede the other. Here, the product, service, or knowledge moves from one partner to the other as it is transformed. An example is a construction project, where certain project components are outsourced to different firms, who make their contributions more or less independently in parallel, or in some specified sequence.
- iii) Reciprocal interdependence, where organizations work together and exchange simultaneous outputs. Each organization is simultaneously dependent on the other, because its inputs are the other organizations’ outputs. An example is a group of organizations exchanging knowledge that is of value to the others, as in automobile manufacturing network groups where member organizations exchange both tacit and explicit knowledge about their production processes. An additional example is new product design, where the client and supplier work together to create and refine a product design that can be manufactured and marketed efficiently.
Of the three forms, pooled interdependence tends to be the least costly in terms of communication and decision effort because there is no real coordination involved, other than an agreement on standard input protocols. Sequential interdependence is more costly because of the need to plan and manage the sequential nature of activities. Reciprocal interdependence is the most difficult to manage effectively, and has the highest coordination costs of all three forms. As the level of interdependence rises from pooled through reciprocal, so does the ability of one organization to inflict damage on another through neglect or misuse of its responsibilities (Kumar & van Dissel, 1996). Higher levels of interdependency also lead to greater complexity in joint projects. For example, one cause of the increase in reciprocal interdependencies is the increased use of concurrent engineering (Williams, 1999). Additionally, most projects are multi-objective, with conflicting goals for cost, schedule, and performance. Complexity also arises because of the trade-offs that must be considered in balancing the effects of project activities on these goals.
High levels of collaboration (i.e., at the reciprocal level) also require high levels of trust. This tends to imply that long-term relationships among network firms are important, because trust is typically built up over a period of time. Practices developed for lower levels of interdependence in industries like the construction industry (e.g., open tendering) tend to require lower levels of trust and may not be effective in situations (e.g., new product development) where reciprocal interdependence is the norm and trust levels must be high.
Higher levels of interdependence among network firms tend to lead to higher orders of complexity in distributed network projects. To increase project completion speed in these circumstances requires attention to complexity reduction and management in order to improve the likelihood of project success. Project complexity considerations derive from two sources: the organization that is undertaking the project, and the complexity of the project itself. Project complexity can be further classified into structural and uncertainty aspects. Contributions to structural aspects include the number and variety of elements and the interdependence of these elements, while uncertainty can arise from goals and methods (Jones & Deckro, 1993; Williams, 1999). Organizational complexity arises from considerations such as multiple stakeholders, including the client, project manager(s), project team, owners of participating organizations, the project champion, the public, and sometimes public bodies, among others (Williams, 1999).
Product complexity plays a key role in addressing issues related to project complexity, as with new product development (Simon, 1969; Singh, 1997). The project management literature focuses on the separation of activities, while the product development literature tends to advocate the crossing of functions and knowledge bases, as with concurrent engineering. Tightly coupled organizational solutions, such as those needed with concurrent engineering, tend to increase complexity and require a higher degree of project coordination. As the complexity and scale of projects increase, the ability to bring projects to a successful conclusion tends to decrease. Complexity and risk may be exacerbated when projects are joint efforts carried out by network organizations that span organizational and geographic boundaries. The various reasons for this will be discussed below. In this section, we present evidence from the literature of complexity issues and proposed solutions that are classified into project structural complexity, product uncertainty, and organizational complexity. Note that many of the solutions described are for complex multi-site internal projects. These may be adaptable to network organizations. The main differences between multi-site internal and network projects are described in a following section.
Structural Complexity in Projects
There are a number of distinct classifications of distributed projects (Evaristo & van Fenema, 1999). First, a single project may encompass several sites, due to scarcity or complementarity of resources, convenience, cost, monitoring capacities, and quality. When talent resources are scarce, all the experts from each of the involved sites could move to a common site to develop the product, or they could work at a distance on the (virtual) project. If a virtual project management approach is chosen, a focus on formal communications and coordination must be stressed among the sub-project leaders at the participating sites, responsible for different modules or functions. Coordination is required to schedule activities and resources in an integrated manner. Team members, of course, also need to communicate among other group members at each site (O’Sullivan, 2003). A critical difference between distributed projects and traditional projects of various types is the focus on the coordination mechanisms. The former project type tends to concentrate on inter-site coordination or boundary spanning across sites; the latter tends to concentrate on intra-site coordination mechanisms, or boundary spanning across projects (Evaristo & van Fenema, 1999).
Large complicated projects are more likely to fail, since they are usually developed by a series of teams working along parallel tracks. Unless all major contingencies are planned for in advance, these tracks will not converge at the end of the project. To overcome this problem, Matta and Ashkenas (2003) suggest injecting a series of rapid-results initiatives into the overall plan, with each contingency having its own version of the overall project goal. Many companies, particularly in the software industry, follow a process for large projects that iterates among design, component development, testing, and customer interaction (Cusumano, 1997). It is generally acknowledged that small teams of talented people are better than large teams for complex projects because smaller teams experience less difficulty in maintaining consistency and communication among team members. They can also simplify scheduling and solve interdependency issues more easily. However, smaller teams have limitations when facing large projects and short deadlines. Consequently, a large project may need to involve hundreds of talented professionals working for many months to complete it within a reasonable timeframe.
Project complexity is usually managed by hierarchical decomposition, breaking projects into separate but related subsystems or modules that can be worked on by smaller teams and then assembled into the final product. However, modularization must also minimize the number of module interconnections, or these interconnections will add unnecessarily to complexity. While simplifying the development process, modularization with standard interconnections also aids in maintenance of the final product. The reverse of the modularization process, usually in parallel with module development, is to integrate and test the modules through the work of multiple teams into a smoothly functioning final product. This obviously requires focused project management attention and resources.
Microsoft handles this “divide, conquer, and integrate” process in major software development projects by a “synchronize and stabilize” style of product development (Cusumano, 1997). Two principles describe this process. The first focuses on evolving features and “fixing” resources; the second is about doing everything in parallel, with frequent synchronization among the different subsystem teams. A similar example of new product development in the aerospace industry (O’Sullivan, 2003) describes a case analysis of work patterns in a virtual multilateral (multi-organization, multi-team) development organization that was composed of a lead firm and its suppliers. In this case, a complex product was successfully co-developed across significant geographical and organizational boundaries, by a joint team that had limited prior experience of working together. One key to success appeared to be the lead firm's imposition of administrative standards. This included standards for work content and timing in relation to managing the resolution of task interdependencies, thus encouraging integrative work patterns to emerge. Another key to success involved regularly scheduled inter-team communications through sub-project managers, including frequent visits to the lead firm's site.
Construction projects have their own characteristics that relate to how complexity can be managed at different stages. Component products used in this industry are relatively standardized, leading to less complexity. But project processes are less standardized (Austin, Newton, Steele, & Waskett, 2002). Further standardization of construction project processes would help reduce the level of complexity and project risk levels.
The decoupling principle states that if two project activities are highly interdependent, these should be carried out by the same organization and under the same authority. According to this principle, if it is not possible for one organization to manage project activities, activities should be made independent (decoupled) (Stinchcombe, 1985). Dahlgren and Söderlund (2001) point out that large engineering projects often result in interdependencies among organizations and phases in the project, resulting in a violation of the decoupling principle. They suggest hierarchy matching at similar levels between the participating organizations as a way to resolve the problem.
Kessler and Chakrabarti (1999) state that new product development speed in R&D organizations can be increased through a contingency approach. In a study of 75 such projects, they found that clear time-goals, longer tenure among team members, and parallel development would increase speed. Designing for manufacturability, frequent product testing, and computer-aided design systems, however, decreased speed. They also found that factors that speed up radical innovation—such as concept clarity, champion presence, and co-location—were found to slow down incremental innovation. Industrial multi-firm projects involve interdependencies that create uncertainties in areas such as planning operations and information. These must be resolved through negotiations between the contracting parties, where the level of coordination required is a function of the degree of interdependence (Dahlgren & Söderlund, 2001). Although a contractual framework provides a structure within which the participating organizations can work, high complexity creates a need for more organic structures (Dahlgren & Söderlund, 2001).
An example of interdependency is re-work, when project components rejected due to low quality must be re-designed and re-developed. This causes knock-on effects in other project element designs, counter to the sequential nature of project management methodologies that assume steady progress. Re-work is a common feature of methodologies such as concurrent engineering that are designed to shorten the project life cycle. One model that has been proposed to address this problem (Pillai, Joshi, & Rao, 2002), integrates the key factors of all the phases of an R&D project life cycle and indicates overall performance through an integrated performance index. Issues specifically addressed include the changing nature of the project as it progresses; and managing risk. Terwiesch, Loch, and De Meyer (2002) have developed a time-dependent model for managing interdependent tasks, with two alternative strategies: iterative and set-based coordination. Set-based coordination requires an absence of ambiguity and should be followed if the cost of pursuing multiple design alternatives in parallel is low. Iterative coordination should be used if the downstream task faces ambiguity or if re-work costs are low.
In concurrent engineering, project tasks usually involve the establishment of multifunctional teams in which team members from different functional departments interact in every phase of development tasks in order to design and develop products and processes concurrently. However, complexity of the product development and design processes may result in the formation of large interdependent task groups that require tight coordination. This makes it difficult for team organization and may delay project completion. To reduce this complexity, Chen and Lin (2003) suggest transforming the binary task relationships into quantifiable task coupling strengths, and decomposing large interdependent task groups into smaller and manageable sub-groups. To accomplish this, they applied a design structure matrix, the analytic hierarchy process, and cluster analysis to represent task relationships, quantify task couplings, and decompose the task groups.
Laufer, Denker, and Shenhar (1996) noted that project managers today are under pressure to complete complex projects under conditions of uncertainty in less time, without sacrificing cost and quality, and without leaving customers and users dissatisfied. Managers who are able to meet this challenge have to move beyond classical scheduling, teamwork, and concurrent engineering approaches. They have to develop a style of simultaneous management in order to manage contending demands.
A study by Gassmann and von Zedtwitz (2003) of virtual project teams in 37 technology intensive multinational companies developed four classifications of project organizations: decentralized self-organization (e.g. independent business units); coordination by a system integrator; core team as system architect; and centralized venture team. These authors suggested a contingency approach for organization selection based on the type of innovation (radical/incremental), systemic (highly interdependent tasks/autonomous nature of project), mode of knowledge (tacit/explicit), and degree of resource bundling (complementary/redundant). They further suggested that the centralization of R&D projects is necessary when the project involves radical innovation, systemic project work, prevalence of tacit knowledge, and presence of complementary resources. On the other hand, they suggested that decentralization of R&D projects is possible when the project involves incremental innovation, autonomous project work, prevalence of explicit knowledge, and presence of redundant resources.
Söderlund (2002) proposed a framework with four possible types of project organization on two dimensions: separated or coupled project phases, combined with either separated or coupled subsystems. He suggested choosing a modular project organization if the project phases were coupled but the subsystems were separated. However, if coupling occurs on both dimensions, project management must deal simultaneously with complex issues involving multiple interfaces and relationships between upstream and downstream activities. The two aspects of time addressed in this paper include the overall deadline and the synchronization of activities within the project. The organization most suited to meeting the overall deadline should be chosen.
Ambos and Schegelmilch (2004) found that foreign R&D units are a heterogeneous group faced with a variety of tasks and increasingly equipped with regional or global market mandates. In their survey of 49 leading German multi-national corporations, Ambos and Schegelmilch classified market mandates as 54% global, 34% regional and only 12% local. Orchestrating international projects adds to the complexity of teamwork because of the diversity among team members of national and managerial culture, educational background, and language (natural and professional). These can become major barriers to meeting project objectives (Lam, 1997). Cultural and management differences tend to magnify project management problems in such environments.
A major study of global R&D (Chiesa, 2000) found two main classifications of project management structures: specialization (through a center of excellence or supported organization) and integration (a more dispersed model involving either a network or dispersed contributors). Both approaches have their advantages and disadvantages. For example, although the integration approach requires a higher degree of coordination, it ensures a broader base of expertise. These two factors are the essence of distributed network projects. In choosing projects for such organizations, a balance is desirable between the coordination effort and the gains from collective expertise that are available through a network organization approach. The different degrees of interdependence make this more evident. For example, where interdependence is either pooled or sequential, coordination is relatively straightforward. When distributed network projects are integrated, interdependence is reciprocal. This gives rise to more problems than the other forms of interdependence because coordination becomes complex and more difficult to manage.
Management techniques must adapt to the new environment; higher levels of interdependence tend to increase the possibility of conflict (Kumar & van Dissel, 1996). Jones and Deckro (1993) explained how an increase in project complexity leads to an increase in internal conflicts within the project, so management methods and style must adapt to deal with this. Changes need to be made to internal management structures within projects; in particular, the use of multi-disciplinary teams is becoming more widespread. Cheng et al (2001) note that projects within the fragmented construction industry are often characterized by disputes that arise from poor communications. To address this issue, they suggest partitioning communication among construction alliance parties into several aspects. These include inter-organizational communication within the alliance team and communication channels that are created for either close or distant connections, with the choice of channels depending on the amount, speed, and required efficiency and effectiveness.
In an analysis of four virtual project cases, Croasdell, Fox, and Sarker (2003) found that successful utilization of communication tools helps to develop strong social connections among team members. Social bonds and strong leadership help team members develop the trust and joint purpose that is needed to work together to resolve issues. Lack of bonding and regular communications may inhibit the successful completion of projects. The importance of trust in the management of virtual network projects is very aptly described as “Only trust can prevent the geographical and organizational distances of global team members from becoming psychological distances” (Jarvenpaa & Leidner, 1999).
Jarvenpaa and Leidner (1999) explored issues around communication and trust in small global virtual teams. These teams worked in an environment that paired groups at universities around the world, separated by location and culture. Jarvenpaa and Leidner report on a series of case studies where the teams were challenged by a common collaborative project. For these teams, the only economically and practically viable communication medium was asynchronous or synchronous computer-mediated communication. Their results suggest that such teams may experience a form of “swift” trust based on positive stereotypes, but this form of trust appears to be very fragile and temporal. They suggest attention to organized and scheduled communications behavior that might facilitate trust in this environment. In particular they contrast individualist and collectivist cultures and their behaviors. People from individualistic societies might be more ready to trust new acquaintances than those from collectivistic societies, in computer-mediated communication environments. Also, social dialog might be easier and tend to generate trust when the team members had previous international experience.
Factors contributing to smooth coordination early in a team's existence include a clear definition of responsibilities. Lack of clarity may lead to confusion, frustration, and disincentive (Jarvenpaa & Leidner, 1999). If the work is only part of the team's responsibility, guidelines on how often and what pattern to use in communication will improve predictability and reduce uncertainty in coordination (O’Sullivan, 2003). Ensuring that team members have a sense of complementary objectives and share in the overall aim of the team helps encourage an appropriate level of participation. Strategies to handle conflict include addressing perceived discontent as soon as it is noticed. Another is to directly address only the concerned individual and avoid sending the entire team messages concerning the conflict. Managers should choose team members carefully, based on characteristics such as responsibility, dependability, independence, and self-sufficiency. These are also desirable in face-to-face situations and critical in virtual teamwork. Although it is not critical to meet in person, it is important to engage in an open and thoughtful message exchange at the beginning of the project. This will help to build a foundation for the team to overcome obstacles as they arise in the virtual environment (Jarvenpaa & Leidner, 1999).
Risk Mitigation and Management
Any increase in project complexity tends to increase risk. For this reason, risk mitigation strategies are an important consideration in distributed network projects. In Table 1a (next page) we have listed the sources of risk, classified by type of project complexity, including some risk mitigation strategies suggested by the literature. Note that some of these risks are endemic to both distributed internal projects and distributed network projects. There are differences among these project types, as we address in the following section. Project structural complexity may arise from either size and/or steps taken to reduce the development cycle time. There are a variety of approaches to mitigating the risks from project size, including the logical decoupling of activities by modularization and a high degree of specification if product design can be entirely specified in advance. Communication, especially among sub-teams, cannot be over-emphasized. The use of standardized components and processes also tends to reduce complexity and risk. Finally, rapid results initiatives and matching hierarchies have been mentioned as management techniques that can help mitigate risk from project size.
|Complexity Type||Source of Risk||Mitigation Strategies||Reference|
|Project (structural)||Project size||Decouple activities; iterative development; high degree of specification; high degree of communication among work teams via sub-project managers; rapid results initiatives; matching hierarchies; standardized components, processes||(Cusumano, 1997; Stinchcombe, 1985) |
(Matta & Ashkenas, 2003) (O’Sullivan, 2003) (Dahlgren & Söderlund, 2001) (Austin et al., 2002)
|Reduction in development cycle time||Concurrent engineering: |
(Design structure matrix;
decomposition of task groups;
iterative & set-based coordination;
integrated performance index)
Concept clarity, champion presence; high degree of communication among sub-teams
|(Chen & Lin, 2003) |
(Terwiesch et al., 2002)
(Pillai et al., 2002) (Kessler & Chakrabarti, 1999)
|Project (uncertainty)||Radical innovation||Co-location||(Kraut, Fussell, Brennan, & Siegel, 2002) |
(Gassmann & von Zedtwitz, 2003)
|Complexity and uncertainty||Separated vs. coupled project phases and sub-systems; simultaneous management||(Söderlund, 2002) (Laufer et al., 1996)|
Table 1a. Project complexity, risk, and mitigation strategies for distributed projects
Reduction in development cycle time is an important aspect of remaining competitive, but this tends to increase project risk. Normally, cycle time is reduced through concurrent engineering approaches, but this introduces additional complexity and risk (particularly due to re-work) because of the parallel work processes. Techniques to overcome these problems include concept clarity, close involvement of the project champion, a high degree of communication, the use of a design structure matrix accompanied by decomposition of task groups, iterative set-based coordination, and an integrated performance index to keep the project on track.
Uncertainty also increases project complexity and risk, because of an inability to specify particular components and outcomes. Radical innovation is typically highly uncertain, so much so that co-location of team members is critical because it allows rapid sharing of experiential/tacit knowledge as the project evolves. Decisions on coupled or separated project phases and sub-systems can usually decrease project complexity, but this may be at the expense of development life cycle length (as in concurrent engineering). Simultaneous management that represents continuous awareness and rapid response on interdependent components is essential in reducing risk in such situations.
As summarized in Table 1b, network organizations contribute to organizational complexity and risk in a variety of ways. If a substantial proportion of knowledge transfer related to the project is tacit, then trust development through long term relationships and the co-location of the core team at one site are two potential solutions. Organizational culture mismatch can be addressed by social interaction and advance planning to address potential problems. Low levels of trust take time to overcome. Techniques include inter-organizational member exchange, strong leadership, social interaction, clear assignment of responsibilities, and scheduled and frequent communications. In the situation of a global development strategy, specialization or integration options are contingencies that should be considered. Frequently, business partners lack common work practices or common working and professional languages. These problems must be addressed. Direction often comes from the linchpin or most influential organization. Low productivity and conflict may result from a lack of communication or interaction among team members. This should be managed by advance planning and potential co-location of the core team.
|Complexity Type||Source of Risk||Mitigation Strategies||Reference|
|Organizational||High degree of tacit knowledge transfer||Co-location; long term relationships||(Gassmann & von Zedtwitz, 2003) (Wagner, 2003)|
|Organizational culture mismatch (e.g. collectivist vs. individualist)||Social interaction; formal planning and agreement to address related issues||(Jarvenpaa & Leidner, 1999) (Lam, 1997)|
|Low trust||Inter-organizational member exchange; strong leadership; social interaction; clear assignment of responsibilities; scheduled, frequent communications||(Jarvenpaa & Leidner, 1999) (Croasdell et al., 2003)|
|Global development strategy||Contingency approach: specialization or integration||(Chiesa, 2000)|
|Lack of common work practices, working and professional language||Direction from linchpin organization||(O’Sullivan, 2003)|
|Potential conflicts||Advance planning; partitioned inter-organizational and other communication channels||(Cheng et al., 2001) (Jones & Deckro, 1993)|
|Low productivity||Co-location||(Olson, Teasley, Covi, & Olson, 2002)|
Table 1b. Organizational Complexity, Risk, and Mitigation Strategies for Distributed Projects
Differences between Multi-Site Internal and Distributed Network Projects
Project complexity and its management is an issue that has been handled in a variety of ways by large integrated firms as well as network organizations. Some of these approaches are suitable for both types of organization. Much of the literature we have discussed, relevant to complexity and risk, has drawn upon research in multi-site internal projects. However, there are significant differences between these two project classes that tend to add even more complexity to distributed network projects. Some of these differences are summarized in Table 2. These are based on our discussions with project managers in firms involved in network organization projects, in the aerospace, semiconductor, and information technology industries. All these complexities tend to slow project completion. The important point is that unless the related risks can be mitigated through careful project planning, these complexities may be so high that they may eliminate projects from consideration.
|Characteristic||Multi-Site Internal||Distributed Network|
|Ownership of intellectual property under development||Not an issue||Requires detailed negotiation and agreement|
|Corporate and professional culture||Rarely an issue||Mismatch can cause severe problems|
|Trust||Occasionally an issue||For new partnerships, takes time to cultivate. Not a serious problem for existing long term relationships|
|Knowledge transfer||Limited by need to know||Limited by terms of project agreement|
|Working language and communication||Usually not an issue||Standards and processes must be negotiated|
|Work procedures, documentation standards||Not an issue||Standards and processes must be negotiated|
|Selection of software packages, languages, etc.||Not an issue||Standards must be negotiated|
Table 2. Major differences between multi-site internal projects and distributed network projects
Ownership of intellectual property that is contributed by each organization, or is to be developed through the joint project, is not an issue for an internal project, but is one that must be addressed contractually in great detail for a distributed network project. Corporate and professional culture mismatch can cause severe problems and high risk in a distributed network project. This is rarely an issue internally, although it may occasionally be a problem for collaborating business units in a loosely managed international corporation. For new partnerships, trust takes time to build, especially between organizations that have differing national cultures (e.g., the individualistic culture of western societies and the collectivistic culture of eastern societies). Knowledge transfer within an organization is based on need to know; it is limited between collaborating organizations based on the terms of the project agreement. Working language, communications, working procedures, documentation standards, and selection of software packages and languages, among other concerns, all have little effect internally because of existing practices within the company. But these may require extensive negotiation, re-training, and development in organizations involved in a distributed network project.
Burn, Marshall, and Barrett (2002) have noted some of the inherent tensions in distributed network projects that do not exist for projects in integrated organizations. These come about from the strengths of distributed network projects: entrepreneurial, risk taking; mutual trust, shared risk; opportunistic. As they point out, these strengths can also become weaknesses. Incentives and rewards for risk taking can lead to difficulties in coordination and cooperation; there are usually few established procedures for negotiation and conflict resolution; and opportunistic behavior by individual organizations can fragment the network.
Inter-Organizational Relationships and Knowledge Exchange
Inter-organizational relationships are not easy to build. Most managers believe that, once these relationships are in place, it is essential to expend substantial efforts in maintaining such relationships. As pointed out in the previous section, working with trusted business partners and building and exchanging relevant knowledge are important aspects of distributed network projects. This section discusses these issues.
There are subtle differences between formal arrangements when a contract is specifically for a well-defined project, typical of the construction industry, as compared to long-term relationships among network firms where projects arise from time to time and project teams are formed from among participating firms. Examples include new product development teams where the linchpin firm may take on the task of product design, working closely with manufacturing, testing, and distribution firms to specify, develop, and test the product design throughout the product development process. Relationships among organizations and people engaged in joint projects can leverage the application of knowledge by providing a path for its distribution through distributed network projects. Information and knowledge transfer, learning, and application that takes place in such environments are critical determinants of network organization effectiveness, as they involve the sharing of customer, operational, and product information among the network organizations (Passiante & Andriani, 2000).
A key strength of a network enterprise is that partnering companies bring their core competencies to share with the multi-firm team. Competency exists in different forms and is generally referred to as the intangible assets of the company. The challenge in a network enterprise is how to make best use of these assets to produce profitable outcomes for all the participating firms. Effective knowledge management (KM) in these situations requires leveraging the organization's knowledge, creating new knowledge or promoting innovation, and increasing collaboration and hence enhancing the skill levels of employees. The most common KM programs include development of a common repository and forming and nurturing communities of practice (Wenger, McDermott, & Snyder, 2002). If KM is implemented across organizations, a useful measurement tool is a KM index based on the balanced score card (Arora, 2002). Practices for sharing specific knowledge depend on the possible degree of codification. Codified (explicit) knowledge can be shared more easily at a distance, while tacit knowledge may require face-to-face contact and interaction. Avadikyan, Llerana, Matt, Rozan, and Wolff (2001) showed that the degree of knowledge codification — ranging from high (contractual) to medium (explicit but not contractual) to tacit (routine or habitual) — of inter-organizational agreements had an impact on inter-firm cooperation.
Generating, identifying, and propagating new ideas and best practices is promoted by inter-organizational learning (Dyer & Nobeoka, 2000; Powell, 1998), including knowledge transfer through social exchange in network organizations (Hall, 2001). Firms are reluctant to share intellectual property and core competencies without prior legal agreements relating to trade secrets, patents, and copyrights (Dyer & Nobeoka, 2000), as well as with arrangements for reciprocal exchanges (Parise & Henderson, 2001). These agreements can cover processes, products, and other intellectual property.
The Toyota automotive network organization is often used as an example to demonstrate inter-organizational knowledge sharing (Dyer & Nobeoka, 2000). Toyota takes the view that the key role of the organization is creating, storing, and applying knowledge rather than simply reducing transaction costs. It uses organizational learning to achieve sustainable competitive advantage through continual learning, adapting, and upgrading. The intention is to increase the capabilities of itself and its partners. Innovation-intensive fields such as biotechnology and pharmaceuticals also rely on collaborations through network organizations. Their key challenge is to develop organizational routines for learning and transferring knowledge that are robust, flexible, and durable (Powell, 1998).
Kazanjian, Drazin, and Glynn (2000) define technological learning as the development of novel technological knowledge that is organizationally accessible for creative problem solving. Knowledge creation and learning extends existing technical knowledge through the design of new products and services. This represents a focused and specific application of organizational learning. It is considerably more of a challenge to extend technological learning across organizational and geographic boundaries, which exhibit barriers such as culture (national as well as managerial), language (professional and natural), distance (time and space), and legal (intellectual property ownership and risk assignment), among others. But this is precisely why some network organizations are formed: to take advantage of complementary specialized knowledge and expertise (Dyer & Nobeoka, 2000).
The idea of communities of practice (Wenger et al., 2002) is easily and frequently extended to network organizations, to assist in the transfer of knowledge among individuals collaborating on specific topics and projects (Archer, 2003), through such tools as knowledge repositories and portal technologies.
Virtual Project Management
Collaboration among different business units in the same organization, or among different organizations for the purpose of undertaking projects, almost invariably involves geographical separation of sites. Projects that involve multiple sites are often called virtual projects. Due to modern technology, there are a multitude of technologies available to support multi-site communication, so the lack of technology support is rarely an issue in virtual projects. However, such issues as knowledge transfer go well beyond technology support. In this section we outline some of the related issues, both from technology and from other perspectives, as they impact the management of distributed network projects.
A virtual team is defined as a group of people, possibly including sub-teams, who interact through interdependent tasks, guided by common purpose and work, across space, time, and organizational boundaries, with links strengthened by information, communication, and transportation technologies (Lipnack & Stamps, 1997). Members must contribute to some common goal. Their participation may be temporary; they may or may not meet face-to-face from time-to-time. Team boundaries may vary according to project requirements. Virtual teams that are linked through advanced computer and telecommunication technologies are prevalent in network organizations today (Dube & Pare, 2001). Global virtual teams differ from more localized virtual teams in several respects. Team members are dispersed around the world and rarely meet face-to-face during the course of a project. They represent different cultures and speak different languages. And they face particular technological dilemmas around accessibility and compatibility.
As virtual teams have become more widely used, their use is becoming increasingly adapted to parallel activities, resulting in an increased need to interact as project schedules are shortened. A virtual project office provides flexibility and cost benefits, but also brings challenges to project managers. These include: accommodating a geographically diverse workforce; facilitating communication to disparate organizations; saving on bureaucratic costs by automating reporting, updating, and distributing data; encouraging and formalizing knowledge transfer and communication among team members; and respecting schedules that may vary across time zones and among different nationalities and cultures (Elkins, 2000). Although there may be substantial cost benefits from virtual teams, the distributed environment is not likely to be as successful as co-located teams (Kraut et al., 2002; Olson et al., 2002). There are also some unique advantages to co-located work (Olson et al., 2002), including increased interactivity and continuous communication, resulting in easier coordination and learning. These advantages can result in substantial increases in work group productivity, especially when tacit knowledge is to be shared.
Technologies for supporting distributed network projects have advanced rapidly in the past few years, including Web based workflow management systems for large scale global projects (Badir, Founou, Stricker, & Bourquin, 2003), portals, knowledge repositories, and communities of practice. A common language is critical to the ability of organizations to be able to exchange knowledge during the development process (Koen, Ajamian, Burkart, & Clamen, 2001). These authors developed a new concept development model that provides a common language and insights into the “fuzzy front end” of innovation and found that highly innovative companies were more proficient in this “fuzzy front end” than others. Bengtsson and Soderholm (2002) see technology development as a boundary-spanning activity where insights and discoveries from different organizations or organizational units emerge as new products or technical solutions. In some cases, technology development projects are organized within networks through cooperation between independent companies possessing unique resources that can be utilized for project components. These authors focused on the relationship between tacit and explicit knowledge and on the different distances inherent in the development effort. Two different bridging processes were proposed as a means to overcome distances: a separating-integrating process and a linking-formalizing process.
Virtual projects have fewer opportunities for face-to-face meetings, so choosing a technology to accomplish a task at the right time becomes a matter of project survival. How and when to use these technologies may require trial and error solutions. Major disadvantages of virtual project management include the lack of physical interaction, nonverbal cues, and synergies that often accompany face-to-face communications, raising issues of trust. However, this concern has other dimensions, such as the type of knowledge and the meaning transferred (tacit or explicit) and the history of interaction among the participants. Videoconferences may help alleviate a lack of physical interaction, providing the ability to interact that is closer to face-to-face, although unstructured videoconferences can easily run astray, resulting in reduced confidence in the project's success and a loss of commitment to the project by team members (Dube & Pare, 2001). Each communication or collaborative tool has strengths and weaknesses. For example, teleconferencing or videoconferencing are synchronous and require higher levels of commitment, flexibility, and discipline than asynchronous technologies such as e-mail and online portals. As with any communication mechanism, virtual projects may involve cultural differences, communication, language barriers, and discrepancies in technological proficiency among team participants.
Project Portfolio Selection
Optimizing the choice and management of a portfolio of projects has always been a difficult issue due to the interaction of multiple resource constraints. As network organizations grow in influence, and international and cultural boundaries are encountered, selecting individual or multiple projects requires an ability to understand and mitigate the effects of both organizational and project complexity. Complexity in the organizational environment tends to complicate portfolio selection because of the interdependencies among participating organizations. And it is important to match the organizational structure to the project type. For example, large and/or innovative R&D projects are best carried out by a co-located organization, but projects that can be broken into separate modules with well-defined interfaces can be handled well in a distributed organization (Gassmann & von Zedtwitz, 2003). Matching measures of organizational structure complexity to appropriate project type may therefore help to mitigate risks associated with a project portfolio. There are a number of techniques for evaluating and comparing the risks and other attributes of projects during the portfolio selection process (Dickinson, Thornton, & Graves, 2001; Archer & Ghasemzadeh, 1999).
A portfolio is balanced if there is a suitable distribution of projects on dimensions such as technology and market risk, completion time, and return on investment. Resource limitations require an organization to strategically allocate resources to a subset of possible technology projects. Currently available tools and methods to select the optimal set of projects are only applicable when projects are independent and are evaluated in a common funding cycle. Methodologies that can be helpful when participating organizations are interdependent include a dependency matrix model to optimize a portfolio of product improvements, balancing risk, overall objectives, and the cost and benefit of the entire portfolio (Dickinson et al., 2001). Distributed network projects require not just optimizing portfolio performance, but negotiating with business partners, suppliers, and customers to achieve goals that are acceptable to all the stakeholders.
In this paper we have addressed a variety of issues that affect both internal multiple site projects and distributed network projects. For network organizations, these issues become more difficult to resolve as the level of interdependence among the organizations increases. Complexities can arise from structural and uncertainty aspects of the project itself, and from a range of organizational complexities. To manage complexity and its associated risks, a variety of mitigating strategies have been proposed in the literature, and we have described some of these. There are multiple sources that risk the outcome of a project, including technology, market, schedule, cost, legal, and political, among others. However, these risks tend to be exacerbated in distributed network projects. To emphasize this point, we outlined some of the differences reported in a limited survey of network project managers. A key issue in distributed network projects is building and maintaining inter-organizational relationships and supporting knowledge exchange between organizations involved in common projects. Because of the distributed nature of network projects, virtual project management is the norm. The availability of supporting communications technology is rarely an issue, but planning and managing joint projects requires a significant increase in attention to risk mitigation, including major decisions on issues such as co-location of the project team.
When distributed network projects are being considered, there are sophisticated methodologies to assess risk amount and impact prior to project commitment, through risk identification, quantification, response development, and control. These are areas in which major advances have occurred recently (Pinto, 2002). A key criterion for successfully applying risk evaluation in portfolio selection is that risk assessment and quantification should be uniformly applied across all projects and teams (both internal and network projects) in order to distinguish among projects that have acceptable and unacceptable levels of risk. For example, companies that rely heavily on joint R&D for new products need to be able to take risks without compromising the profitability of the company. Company strategy must be linked to portfolio development when it involves both a high degree of innovation and a high rate of growth. These, however, tend to be conflicting objectives. An emphasis on joint projects requires considerations that overlap corporate boundaries, involving business partners in strategic decisions of project portfolio choice.
In managing distributed network projects, large companies with multiple sites tend to have an advantage over smaller single site companies, as a result of the maturity gained by handling internal multi-site projects. Expanding their reach to distributed network projects should focus on issues raised from differences we outlined in this paper. Assigning project managers with experience in multi-site projects, combined with international experience, can significantly improve the prospects for distributed network projects. Based on our analysis of distributed network projects, we believe that the risks tend to be higher than with internal projects, although such joint projects may provide considerable leverage for a company's expertise and other resources. To take advantage of this leverage requires experienced project managers and risk mitigation strategies implemented through joint planning.
Ambos, B., & Schlegelmilch, B. B. (2004). The use of international R&D teams: An empirical investigation of selected contingency factors. Journal of World Business, 39, 37-48.
Archer, N. (2003). Some perspectives on communities of practice (Working Paper No. 460). Hamilton, ON: School of Business, McMaster University.
Archer, N., & Ghasemzadeh, F. (1999). An integrated framework for project portfolio selection. International Journal of Project Management, 17(4), 207-216.
Arora, R. (2002). Implementing KM - A balanced score card approach. Journal of Knowledge Management, 6(3), 240.
Austin, S., Newton, A., Steele, J., & Waskett, P. (2002). Modelling and managing project complexity. International Journal of Project Management, 20, 191-198.
Avadikyan, A., Llerana, P., Matt, M., Rozan, A., & Wolff, S. (2001). Organisational rules, codification and knowledge creation in inter-organisation cooperative agreements. Research Policy, 30, 1443-1458.
Badir, Y. F., Founou, R., Stricker, C., & Bourquin, V. (2003). Management of global large-scale projects through a federation of multiple Web-based workflow management systems. Project Management Journal, 34(3), 40-47.
Bal, J., & Gundry, J. (1999). Virtual teaming in the automotive supply chain. Team Performance Management, 5(6), 174-193.
Bengtsson, M., & Soderholm, A. (2002). Bridging distances: Organizing boundary-spanning technology development projects. Regional Studies, 36(3), 263.
Burn, J., Marshall, P., & Barnett, M. (2002). E-business strategies for virtual organizations. Oxford: Butterworth Heinemann.
Chen, S.-J. G., & Lin, L. (2003). Decomposition of interdependent task group for concurrent engineering. Computers & Industrial Engineering, 44, 435-459.
Cheng, E. W. L., Li, H., Love, P. E. D., & Irani, Z. (2001). Network communication in the construction industry. Corporate Communications, 6(2), 61-70.
Chiesa, V. (2000). Global R&D project management and organization: A taxonomy. Journal of Product Innovation Management, 17, 341-359.
Croasdell, D., Fox, A., & Sarker, S. (2003). Systems development by virtual project teams: A comparative study of four cases. In M. Kosrow-Pour (Ed.), Annals of Cases in Information Technology (Vol. 5). Hershey, PA: Idea Group Publishing.
Cusumano, M. A. (1997). How Microsoft makes large teams work like small teams. MIT Sloan Management Review, 39(1), 9-20.
Dahlgren, J., & Söderlund, J. (2001). Managing inter-firm industrial projects - on pacing and matching hierarchies. International Business Review, 10, 305-322.
Dickinson, M. W., Thornton, A. C., & Graves, S. (2001). Technology portfolio management: Optimizing interdependent projects over multiple time periods. IEEE Transactions on Engineering Management, 48(4), 518-527.
Dube, L., & Pare, G. (2001). Global virtual teams. Communications of the ACM, 44(12), 71-73.
Dyer, J. H., & Nobeoka, K. (2000). Creating and managing a high-performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21(3), 345-367.
Elkins, T. (2000). Virtual teams connect and collaborate. Industrial Engineer, 32(4), 26-31.
Evaristo, R., & van Fenema, P. C. (1999). A typology of project management: Emergence and evolution of new forms. International Journal of Project Management, 17(5), 275-281.
Gassmann, O., & von Zedtwitz, M. (2003). Trends and determinants of managing virtual R&D teams. R & D Management, 33(3), 243-262.
Hall, H. (2001, April 2001). Social exchange for knowledge exchange. Paper presented at the Managing knowledge: Conversations and critiques, Leicester, England.
Harbison, J. R., & Pekar, P. (1998). Smart alliances: A practical guide to repeatable success. San Francisco, CA: Jossey-Bass Publishers.
Huang, G. Q., Mak, K. L., & Humphreys, P. K. (2003). A new model of the customer-supplier partnership in new product development. Journal of Materials Processing Technology, 138, 301-305.
Jarillo, J. C. (1988). On strategic networks. Strategic Management Journal, 9, 31-41.
Jarvenpaa, S. L., & Leidner, D. E. (1999). Communication and trust in global virtual teams. Organization Science, 10(6), 791-815.
Jones, R. E., & Deckro, R. F. (1993). The social psychology of project management conflict. European Journal of Operational Research, 64, 216-228.
Kazanjian, R. K., Drazin, R., & Glynn, M. A. (2000). Creativity and technological learning: The roles of organization architecture and crisis in large-scale projects. Journal of Engineering and Technology Management, 17(3/4), 273.
Kessler, E. H., & Chakrabarti, A. K. (1999). Speeding up the pace of new product development. Journal of Product Innovation Management, 16, 231-247.
Koen, P., Ajamian, G., Burkart, R., & Clamen, A. (2001). Providing clarity and a common language to the “fuzzy front end”. Research Technology Management, 44(2), 46-55.
Kraut, R. E., Fussell, S. R., Brennan, S. E., & Siegel, J. (2002). Understanding effects of proximity on collaboration: Implications for technologies to support remote collaborative work. In P. H. S. Keisler (Ed.), Distributed work (pp. 137-162). Cambridge, MA: MIT Press.
Kumar, K., & van Dissel, H. G. (1996). Sustainable collaboration: Managing conflict and cooperation in interorganizational systems. MIS Quarterly, 20(3), 279-300.
Lam, A. (1997). Embedded firms, embedded knowledge: Problems of collaboration and knowledge transfer in global cooperative ventures. Organizational Studies, 18(6), 973-997.
Laufer, A., Denker, G. R., & Shenhar, A. J. (1996). Simultaneous management: The key to excellence in capital projects. International Journal of Project Management, 14, 189-199.
Lipnack, J., & Stamps, J. (1997). Virtual teams - Reaching across space, time, and organizations with technology. New York, NY: Wiley.
Matta, N. F., & Ashkenas, R. N. (2003). Why good projects fail anyway. Harvard Business Review, 81(9), 109.
Olson, J. S., Teasley, S., Covi, L., & Olson, G. (2002). The (currently) unique advantages of collocated work. In P. H. S. Keisler (Ed.), Distributed work (pp. 113-135). Cambridge, MA: MIT Press.
O’Sullivan, A. (2003). Dispersed collaboration in a multi-firm, multi-team product-development project. Journal of Engineering and Technology Management, 20, 93-116.
Parise, S., & Henderson, J. C. (2001). Knowledge resource exchange in strategic alliances. IBM Systems Journal, 40(4), 908-924.
Pascale, R. T. (1999). Surfing the edge of chaos. Sloan Management Review, 40(3), 83.
Passiante, G., & Andriani, P. (2000). Modelling the learning environment of virtual knowledge networks: Some empirical evidence. International Journal of Innovation Management, 4(1), 1-31.
Pillai, A. S., Joshi, A., & Rao, K. S. (2002). Performance measurement of R&D projects in a multi-project, concurrent engineering environment. International Journal of Project Management, 20, 165-177.
Pinto, J. K. (2002). Project management 2002. Research technology management, 45(2), 22-37.
Powell, W. W. (1998). Learning from collaboration: Knowledge and networks in the biotechnology and pharmaceutical industries. California Management Review, 40(3), 228-241.
Powell, W. W., Koput, K. W., & Smith-Doer, L. (1996). Interorganizational collaboration and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41(1), 116-145.
Simon, H. A. (1969). The sciences of the artificial. Cambridge, MA: MIT Press.
Singh, K. (1997). The impact of technological complexity and inter-firm cooperation on business survival. Academy of Management Journal, 40(2), 339-368.
Söderlund, J. (2002). Managing complex development projects: Arenas, knowledge processes, and time. R&D Management, 32(5), 419-430.
Stinchcombe, A. (1985). Project management in the Nordic Sea. In A. L. S. a. C. A. Heimer (Ed.), Organization theory and project management. Bergen, Norway: Norwegian University Press.
Terwiesch, C., Loch, C. H., & De Meyer, A. (2002). Exchanging preliminary information in concurrent engineering: Alternative coordination strategies. Organization Science, 13(4), 402-421.
Thompson, J. D. (1967). Organizations in action: Social science bases of administrative theory. New York, NY: McGraw-Hill.
Wagner, B. A. (2003). Learning and knowledge transfer in partnering: An empirical case study. Journal of Knowledge Management, 7(2), 97.
Wenger, E., McDermott, R., & Snyder, W. M. (2002). Cultivating communities of practice: A guide to managing knowledge. Boston, MA: Harvard Business School Press.
Williams, T. M. (1999). The need for new paradigms for complex projects. International Journal of Project Management, 17(5), 269-273.