Insulating teams are, for the most part, sequestered from the environment. This may be a deliberate choice by the manager or leader; other times, the team may be ostracized by the organization. Managers often want to isolate their teams for security reasons

(e.g., the Los Alamos team that developed the atomic bomb) or intellectual reasons (e.g., the Xerox PARC team, which developed the personal computer). The greatest threat to the effectiveness of insulating teams is dissociation from the organization. Insulating teams may fail to develop a viable product or service because they are out of touch with the rest of the organization and industry, or they may have a great product but lack the support they need for success.

Insulating teams not only have to sequester themselves from others in the company so that they can focus on the work that needs to be done, they also need to insulate themselves from outsiders, who have strategic reasons to learn their secrets. Competitors often want access to information so that they can decipher long-range business plans. Information technology has made insulation much more challenging. One company went to extreme lengths to preserve their project plans.6 They scheduled a particularly important strategic planning conference and had checked into the hotel. The conference room number and the name of their company were listed on the hotel’s daily agenda sign, but no one showed up. At the appointed hour, without fanfare, they boarded a chartered bus designated for a nonexistent “tour group.” From there, they travelled to an unannounced locale and began their meeting far from the eyes of outsiders who wanted to infiltrate the meeting.

Broadcasting teams concentrate on their internal processes and simply inform others what they are doing. More often than not, the broadcasting team has little outside contact; it makes decisions about how to serve its customers from within. The team members let others outside the team know what they are doing after they have already made decisions.

Marketing teams promote their objectives, products, services, and culture actively within their organization. Their objective is to garner support from organizational authorities and receive recognition. They differ from broadcasting teams in that they actively tailor their communication to suit the needs, interests, and objectives of the organization. For example, Google designed the campus of their U.S. headquarters to encourage opportunities for employee networking among their 8,000 employees by constructing clusters to work in and large white boards in open areas to stimulate the public gathering and display of ideas and building multiple cafes and seating nooks.7

Surveying teams have their finger on the pulse of the organization, their clients, and customers. For example, Chicago T-shirt company Threadless tears down barriers between employees and customers by asking customers to submit their own designs that other customers vote for online.8

The disadvantage of surveying, of course, is that the team may spend an undue amount of time and resources surveying clients and customers instead of engaging in the task at hand. In this sense, a surveying team is really a task force. At some point, the surveying team has to make critical decisions about when to stop collecting data and probing for information and develop a product or service.

It is rare for teams to be self-sufficient. Their dependence on others for economic, social, and political support drives many of the strategies for interacting with the environment. Xerox PARC is an example wherein a team used its insulation to advantage in developing groundbreaking technology. However, the team members were unsuccessful in selling their ideas to their own management (because virtually no broadcasting or marketing had been done). Therefore, Xerox did not obtain the benefits of its own special research group.

X-teams are highly externally oriented, and their members reach out across the organization to forge dense networks.9 X-teams enable rapid execution using a three-phase process: exploration, exploitation, and exportation.10 During exploration , X-team members try to understand their task from a novel vantage point, generating as many insights and ideas as possible. During exploitation , X-teams settle upon one product they wish to create, using rapid prototyping to move from possibilities to reality. Finally, during exportation , they find ways to move their product and their knowledge and excitement into the broader organization and market. X-teams may find themselves crossing what might be considered competitive organizational borders to achieve a shared goal. For example, in 2012, two fierce organizational competitors, Roche, a Swiss manufacturer, and Seaside Therapeutics, a U.S. pharma company, forged an X-team with the mission of providing drug therapy with much faster access to more than 100,000 sufferers of Fragile X syndrome. Previously, the companies specialized in drugs for mental retardation and autism; prior to forming the X-team, each company carefully guarded its pharmaceuticals and protected its patent plans. But in a novel move to combat Fragile X syndrome, the competitors pooled resources and shared technologies and patents. By traversing heretofore protected organizational borders, the team was ultimately able to get life-saving drugs to market more quickly.11

X-team behaviors involve selecting team members based on their networks, prioritizing external outreach as the modus operandi, and utilizing established internal processes to facilitate external work. Other X-team behaviors consist of working with management leadership for project commitment, resources, and support; setting goals and measurable results for exploration, exploitation, and exportation activities; and ensuring the team focuses on scouting, ambassadorship, and task coordination.12

The relationship a team has with its environment evolves, sometimes from direct managerial intervention, but more often from organizational norms, culture, and situational constraints. For example, consider how founding teams in organizations influence organizational development. The development of all organizations begins with a single (founding) team. Founding teams have a profound impact on the organization as well as the community, by creating new populations. When these organizations disband, they create a large amount of employment volatility through job creation and destruction. In one investigation of 830 entrepreneurs of founding teams, two factors strongly affected the composition of these entrepreneurs’ founding teams: homophily and network constraints based on strong ties.13 Specifically, homophily or similarity was a key determinant of team composition. The existence of strong ties (such as marriage and long-term friendship) was also key.

Our discussion of team management vis-à-vis the external environment may seem to be overly politicized. There is a trade-off between the team members actually doing their work and their efforts at influencing management and other relevant external entities. If leaders fail to effectively manage the relationship between the team and the external environment, the team may be regarded as ineffective, regardless of its actual productivity. However, by placing constraints and barriers upon the team’s ability to control and gain access to resources, such as education, knowledge, and economic resources, team effectiveness may be hindered.

The interaction between a team and the organization is similar to the interaction between the members and the team itself in that there are struggles for power, status, and roles.14 Just as members spend time during the formation of a team to determine what role they will play and which members will have power, teams play out these same issues with other organizational entities. For example, top management may set constraints and provide direction. Teams may react in a variety of ways: some welcome the direction, some try to shape the new directives, and others reject direction.

Just as people develop reputations and make impressions on others in the organization within the first few months of their arrival, so do teams. On the basis of this initial impression, a negative or positive escalating cycle of team reputation and team performance begins. Reputations may persist over time, despite efforts to change them. Teams that do not overcome a negative initial evaluation may be perceived as failures—even if they eventually achieve their goals. “The inability to influence top management early on can be devastating to a team. In short, labeling creates self-fulfilling prophecies.”15 Often, teams want to get down to work and not waste time navigating their external relationships. However, these very issues—if ignored early on—can haunt the team in the future. Therefore, even though teams must fulfill both internal and external activities, the point in time at which they undertake those activities is not trivial. Obtaining support from the environment is critical if a team is to enter a positive escalation cycle.16

In an ideal organizational environment, there is clear and consistent communication among the different functional and geographic units. Communication quickly disperses innovation, reduces unnecessary duplication of effort, and facilitates the implementation of best practices.

However, the reality of organizational communication may be a far cry from the ideal. Communication among different functional and geographically dispersed units does not occur as frequently as it should. Who talks to whom is much more a function of informal, social networks than a bureaucrat’s organizational chart might lead one to believe.

Most organizations are composed of informal communication networks, in which communication is incomplete and information is not ubiquitously dispersed throughout the company. Information may be controlled or held only by certain people, as opposed to being directly accessible to everyone in the organization.

Why is it that some teams are singled out and win more approval from senior management than other teams? The typical explanation centers upon human capital : Inequalities result from differences in individual ability. People who are more intelligent, educated, and experienced rise to the top of their organization; those who are less qualified do not.

However, there is another explanation that accounts for the existence of inequality. Social capital is the value managers add to their teams and organizations through their ties to other people.23 Social capital refers to the resources available through social networks and elite institutional ties—such as club memberships—that a person can use to enhance his or her position. Social capital is the value that comes from knowing who, when, and how to coordinate through various contacts. Whereas human capital refers to individual ability, social capital refers to opportunity created through relationships. Managers with more social capital get higher returns on their human capital because they are positioned to identify and develop more rewarding opportunities.24 Certain managers are connected to certain others, trust certain others, feel obligated to support certain others, and are dependent on exchange and reciprocity with certain others, above and beyond those in their immediate functional unit.

To understand the value of social capital within the organization and how it affects team performance, it is necessary to consider the broader organizational environment. Organizational charts are rather crude depictions that reveal the chain of command and report relationships. However, as many managers can attest, the way that work gets done and information gets spread within an organization is a far cry from published organizational charts. Instead, informal systems of connections and relationships, developed over time, guide the flow of information between people and teams. Let’s explore these informal networks in detail.

Exhibit 10-3 is a somewhat crude depiction of two social networks in an organization. The dots represent people. The lines that connect the dots are the communication networks between people in the organization—simply, who trusts whom. To be sure, members of a given department or functional unit are more interconnected than are members of completely different functional units. However, even within particular departments, there is a high variation in communication. Moreover, there is high variation among organizational members in terms of who reaches beyond the walls of the functional teams and communicates with others.

Perceived networks may predict performance more than actual networks.25 The accuracy of perceived advice networks is related to increased power in organizations.26 People who have an astute knowledge of network links are in a position of power. First, this information provides a good assessment of who is powerful in the organization. Second, this information can be used to identify where the coalitions are in an

organization, their size, and their sources of support. Third, an accurate assessment of the network can expose the weaknesses in other groups by exposing holes, gaps, and lack of support.

Consider the two different panels in Exhibit 10-3. Panel A depicts the network (or communication) structure of a manager, Gregory, who has a network of relatively close colleagues—most likely from the same functional unit. This type of close-knit, self-contained network is a clique network and is reminiscent of the traditional family unit. In the clique network, groups of people, all of whom know one another quite well, share largely redundant communication structures. At the extreme, members of clique networks are only aware of others with whom they have direct contact.

When we contrast Gregory’s communication network to Berta’s in Panel B, we notice some striking differences. First, Berta’s network is much less tightly knit than is Gregory’s network. Second, Berta’s network spans what appears to be more functional units than does Gregory’s network. In a sense, Berta knows more people who don’t know each other. Third, Berta’s network is structurally more unique than is Gregory’s network. Simply stated, Gregory’s network is highly similar to all the other people in his clique. In contrast, Berta’s network does not look like anyone else’s, in terms of the connections she has. Before reading further, how do you think these differing network structures affect the performance of the team, the individuals involved, and the larger organization?

Individuals (and teams) who span organizational divides and integrate the knowledge and best practices from different areas of the organization (who otherwise have little incentive to do so) are extremely valuable for the organization. These people are known as boundary spanners . Boundary spanners bridge the functional gaps, or the structural holes , that exist in organizations. Structural holes separate nonredundant social contacts in the organization. A person who bridges or spans a structural hole fills a unique spot in the organizational network: bringing together people, knowledge, and information that would otherwise not be connected. This, of course, is critical for maximizing diversity of ideas and setting the stage for creative thinking in the organization—not to mention avoiding duplication of effort and speeding along organizational innovations to relevant units. Teams whose members have access to different social networks, independent of their individual knowledge and skills, may be more likely to learn through relationships than teams whose members’ social ties are redundant.27 The value of a boundary spanner is the ability to capitalize on the social structure of the organization that, as we have seen, is structurally imperfect.

When we again compare the networks of Gregory and Berta, we see that Gregory has a network that spans one structural hole (i.e., the relatively weak connection between a cluster reached through contacts 1, 2, and 3 versus the other cluster that is reached through contacts 4 and 5). In contrast, Berta preserves the connections with both clusters in Gregory’s network but expands the network to a more diverse set of contacts. Berta’s network, adding three new clusters of people, spans 10 structural holes.

The structural holes that exist between people, functional units, and teams represent opportunities for teams and their leaders. Boundary spanners broker the flow of information between people on opposite ends of a structural hole and control the nature of projects that bring people together on opposite ends of the structural hole. In interlocking teams, the functional teams are networked to other teams in different functions via a supervisor who coordinates activities. In this way, the structure of a network is a competitive advantage for certain people and certain teams in the organization.

Managers with contact networks rich in structural holes are the people who know about, have a hand in, and exercise more control over rewarding opportunities. They have broader access to information because of their diverse contacts. They are more aware of new opportunities and have easier access to these opportunities than do their peers—even their peers of equivalent or greater human capital! For this reason, they are also more likely to be discussed as suitable candidates for inclusion in new opportunities. They are also likely to have sharpened and displayed their capabilities because they have more control over the substance of their work defined by relationships with subordinates, superiors, and colleagues. A study of 64 software development teams revealed that boundary spanning, buffering, and boundary reinforcement were positively related to team performance and psychological safety.28

In Exhibit 10-3, Gregory is in a tightly constructed, dense clique network. In contrast, Berta’s network is much less dense, more unique, and more varied. Members of clique networks consider one another to be their closest contacts, and because they focus their efforts at primarily internal communications, they are often sequestered from the larger organization. The loyalty and cohesiveness of inner circles can be both comforting to those situated securely in them and intimidating to those who stumble into them without a real ally.

Berta’s entrepreneur network is a less tightly knit group, with contacts in a variety of disparate organizational areas. In fact, Berta does not appear to be housed in any particular network. On the surface it might seem that Gregory would feel more secure, nestled in his cohesive group, and, hence, be more successful. Berta is a boundary spanner, a link between different subgroups and functional units that without her would not be connected. Berta occupies a unique position in her network as she single-handedly bridges these separate groups. In this sense, Berta is an information broker , because she alone is at the critical junction between these networks and serves the important role of brokering information. The people in Berta’s functional group are more dependent on Berta than on Gregory for information. In a very crude sense, Gregory is an organizational clone—expendable—at least on a sociostructural level. In contrast, Berta is a critical player; remove her and the organization may suffer serious consequences and lost opportunities. Berta serves an important team and organizational function by garnering information that would otherwise be unavailable to the team or the organization.

Gregory is in a highly cohesive group, which can be advantageous when it comes to managing the internal team environment. The problem is that Gregory does not really learn anything new by interacting with the members of his group. In contrast, because Berta’s contacts do not know one another and, therefore, cannot apply social pressure on each other—as they can in clique networks—Berta is potentially privy to a greater amount of accurate and nonredundant information. Berta’s position as a structural hole is an indicator that people on either side of the hole circulate in different flows of information. The structural hole between two clusters (or teams) does not mean that people in the two clusters are unaware of one another. Rather, people are so focused on their own activities that they have little time to attend to the activities of people in the other cluster.

In comparing Gregory’s and Berta’s networks, the information benefits in Berta’s network are enhanced in several important ways that go a long way toward furthering individual, team, and organizational goals—the very things that we outlined in Chapter 2 to be critical measures of performance. From Berta’s view, there is more benefit reaped because more contacts are included in the network. The diversity of contacts enhances the quality of benefits, because each cluster of contacts is an independent source of information. One cluster (e.g., a team), no matter how numerous its members, is only one source of information because people connected to one another know about the same things at approximately the same time. Because nonredundant contacts are linked only through the leader at the center of the network, the leader is assured of being the first to see new opportunities created by the need of one team that could be served by skills in another team. Berta has the opportunity of bringing together otherwise disconnected individuals. Furthermore, having more diverse contacts means that the manager is more likely to be among the people discussed as suitable candidates for inclusion in new opportunities. Because people communicate through Berta, she can adjust her image with each contact. Berta is able to monitor information more effectively than is possible with typical bureaucratic control; she is highly mobile relative to bureaucracy. For a summary of the advantages and disadvantages of cliques and boundary-spanning networks, see Exhibit 10-4.

In comparison with others who are equal in human capital, boundary spanners enjoy more organizational success. Managers with larger networks of disconnected contacts get promoted earlier than comparable managers with smaller networks of more interconnected contacts.29 In an investigation of 3,000 senior managers in a high-technology firm with over 100,000 employees, those promoted early had more social capital—as determined by their social network analysis.30 CEOs’ compensation is positively predicted by the social capital of the chair of the compensation committee and the CEO’s social capital relative to the chair.31 The success of Broadway musical teams from 1945 through 1989 revealed that the “small world” (connected quality) of

creative artists directly affected the financial and artistic success of the show as well as the percentage of hits and flops and raves.32 Similarly, the success of scientific teams in social psychology, ecology, and astronomy is also directly predicted by the structure of the creative collaboration network of the scientists in these communities.33

However, a study of 525 companies that went public reveals that their boards were best when comprised of a majority of the original top management team (TMT) members, rather than independent outsiders.34 Board members possess valuable tacit knowledge of the company; outsiders should provide resources for the TMTs to execute strategies, rather than monitor them.

The network structure of team members affects how much information they share in common with others in a team. For this reason, boundary spanners are more cognitively central (i.e., aware of what other people know) than are non–boundary spanners. This affects their ability to influence the team. The more information team members share with others, the more cognitively central they are in the team. Also, cognitively central members acquire pivotal power in a team and exert more influence on consensus than members who are not cognitively central.35 Thus, those who are connected are more influential.

Team social capital is the configuration of team members’ social relationships within a group and in the social structure of the broader organization.36 Although boundary-spanning activities can benefit team performance, pursuing social relationships outside one’s team might decrease the group’s internal cohesiveness,37 which can in turn negatively affect performance.38 Some teams have greater social capital “liquidity” because their members have positions in the overall social structure of the organization. For example, rapid access to information and political support is greater for teams whose members socialize during their free time with organization’s upper management. In an investigation of work-related ties that extend into social realms outside of work, the optimal configuration was a moderate level of internal closure within the group and a large number of “bridging” relationships to other groups’ leaders.39 It is actually counterproductive for a team when all of the members socialize informally outside of the organization. It is better for the team members to socialize outside of the team and outside of work. Such ties are particularly critical because the shift in focus from work to social interaction invites a shift in the types of resources that are transferred among members. A meta-analysis of 37 studies of teams in real organizations indicated that teams with densely configured interpersonal ties attain their goals better and are more committed to staying together.40 In another investigation involving 51 teams working in two wood composite plants, external coordination (extra-team relationships) had a significant, positive impact on team performance.41

The network structure of team leaders is important for team performance. Teams with leaders who are central in their teams’ intragroup networks and teams that are central in their intergroup network perform better.42 According to the leader centrality–performance hypothesis , team leaders from whom subordinates seek advice or friendship tend to have relatively comprehensive views of the social structures of their teams. Such a perspective helps them make better decisions. Central leaders occupy structurally advantageous positions in their social networks and often act as gatekeepers and regulators of resource flow.43 In a study of 300 self-managing teams at a large manufacturing plant of a Fortune 500 corporation, leaders who contributed most to their team’s success excelled at managing the boundary between the team and the larger organization.44 This boundary-management behavior involved four key skills: relating, scouting, persuading, and empowering (see Exhibit 10-5).45 Teams with more prestigious formal leaders (i.e., leaders whom a high proportion of subordinates sought out for advice) experienced lower levels of conflict and had higher levels of team viability.46 In contrast, leaders who had advice ties with subordinates (who did not have advice ties with each other) had elevated levels of team conflict and lower levels of team viability. Coworkers tend to place more trust in fellow coworkers who are trusted by the team’s formal leaders than in those who are less trusted by the leader.47

Organizational actors who hold an accurate perception of their own social network are more effective than those who hold inaccurate perceptions.48 People’s perceptions of their organizational networks are not necessarily independent perceptions. An investigation of teams in the public and private organizations reveals that team members’ perceptions of organizational support are similar to those of coworkers with whom they seek advice and those who hold equivalent positions in the organizational network.49

Everyone cannot be a boundary spanner or information broker. Managers who are aware of their information brokers are more effective in using their own networks. (As an exercise, follow the three-step guide in Exhibit 10-7.) Just because you know who the brokers are in your network does not automatically entitle you to rely on them. Some people have been able to rely on networks by putting forth a meaningful goal or mission. For example, song composer Missy Mazzoli had a goal of funding a recording of her opera Songs from the Uproar, inspired by the turbulent life of Swiss explorer Isabelle Eberhardt. So, she went to Kickstarter, and a network of 103 friends, fans, and art patrons interested in giving life to a genre generally too unprofitable for record companies to take an interest in and raised nearly $8K in 10 days.50

Structural holes are the gaps that are created when members of cliques (closed networks) do not connect. Identify those gaps and find a way to bridge those gaps. When a manager identifies structural holes, he or she recognizes and overcomes the clique network system. Clique networks contain a number of disadvantages for the organization and the team. People in clique networks fall prey to the homogeneity bias. For example, men in clique networks include significantly fewer women among their contacts.51 This can stifle creativity, propagate prejudice, and hinder the benefits of diversity.

Expanding the size of the network does not mean increasing the size of the team but rather, increasing the number of people that the manager and the team come into contact with. Speed interviewing is an example of maximizing contacts. Companies can’t evaluate finance candidates over weeks or months because others will hire them first. In one survey, 100 executives reported spending an average of 60 minutes when interviewing staff-level candidates and 103 minutes for management-level job candidates. The face-to-face interview was almost inevitably followed by a much shorter phone interview.52 In our simulations of people who have experienced job threat—such as loss of employment—those with higher self-rated socioeconomic status report having broader and more expansive networks than those with lower socioeconomic status.53

The homogeneity bias pulls people toward developing cliques of like-minded people—which is exactly the wrong type of network to foster. It may be more comfortable in the short run, but it will have negative, long-term consequences for the individual, the team, and the organization. A better alternative is to develop a diversified network, which means crossing organizational boundaries and functional areas. People who have a close friend in an ethnically different group are more likely to perceive an association of the friend’s ethnicity with the self and have positive expectations about the out-group.54

Uzzi distinguished embedded ties (social ties) from arm’s length (nonsocial, purely business) ties.55 According to Uzzi, at the network level, the optimal composition of ties is achieved when a network has an integrated mix of embedded as well as arm’s length ties.56 This is because embedded ties and arm’s length ties can offer complementary benefits when combined, just as the overall value of a portfolio increases when it is composed of complementary assets that offset each other’s inherent weaknesses and strengths. Networks dominated by one type of tie produce fewer benefits, which partly accounts for the negative effects of overly embedded (e.g., “old-boy”) networks, or overly disembedded exchange networks (e.g., whipsawing of suppliers by large manufacturers).57 A study of 375 managers over an 8-month period revealed that the more embedded managers perceived themselves to be, the more likely they were to show declines in social capital development over time, and declines in social capital behaviors were directly related to declines in human capital development.58 Moreover, another investigation of the strategic and operational effectiveness of 96 self-managing teams in a large multinational organization revealed that a combination of autonomy and external knowledge allowed teams to capture the benefits while offsetting their risks. Specifically, teams with high levels of autonomy and external knowledge delivered more strategically and operationally effective projects than did teams with high autonomy but low external knowledge or high external knowledge but low autonomy. However, the complementarity between autonomy and external knowledge depended on what the teams were doing and what they knew: When content knowledge was scarce (e.g., country knowledge), autonomy and knowledge improved strategic and operational effectiveness, but not when content knowledge was common (e.g., technical knowledge).59

Consider Joshi’s model for understanding the networking of teams based on their organizational demography (see Exhibit 10-8).60 Organizational demography

refers to the composition of teams in terms of the proportion of majority and minority members in the team.61 The majority is an identity group that has typically enjoyed access to power and authority in the organization. The minority represents an identity group that has been marginalized from a position of power and authority. The model distinguishes three types of team demographics: homogeneous majority teams, homogeneous minority teams, and diverse teams. Homogeneous majority teams are predominantly composed of people belonging to a high-status majority identity group. Homogeneous minority teams are composed of members belonging to a single-identity group that is not that of the larger organization. Diverse or heterogeneous teams represent both majority and minority members equally. According to the model, homophilous ties may limit the extent to which team members have access to diverse perspectives in homogeneous majority teams. As organizations move toward pluralistic settings, these teams may lose support and information from minority peers. In terms of homogenous minority teams, lack of networking opportunities with peers and upper management can be detrimental; such teams need to devise strategies to compensate for the numerical scarcity of minority members. Diverse teams can gain centrality in external networks through homophilous interactions of its majority team members. To maximally use the benefits of their external networks, diverse teams need to focus on building trust and cooperation internally in monolithic and pluralistic settings.

In bureaucratic organizations, boundary spanners need to network not only laterally but also hierarchically. When Nokia attempted to change their reputation for bureaucracy, new employees were introduced to ten or more people both inside and outside of their respective departments, and both up and down the management hierarchy.62 An improvisational theater group in Denver, Colorado, Playback Theatre West, takes this idea literally. The group’s tool, called sociometry , which they introduce at company retreats, asks each employee to line up behind the supervisor they trust the most (other than the boss). Thus, the “real” hierarchical network, a map of the informal networks inside the company, appears.63

Quite often, business opportunities are conducted in the context of social relationships.64 Many social relationships are common and therefore highly scripted or routinized, such as conducting a business meeting on the golf course. These scripts are significant because they define the conditions under which networks are most effective.65 For this reason, it is difficult to form close relationships across gender lines if they are built through socializing activities such as playing golf, going to the theater, or evening dinner gatherings, because these practices often have a different meaning between men and women than they do between persons of the same gender.66 One study investigated why women scientists with human capital equal to, or better than, their male counterparts do not do as well in graduate school and are thereafter disadvantaged by a lack of contacts to the resources and tacit information important for identifying, developing, and following through on leading-edge research projects.67 The differential performance of women followed from their disadvantaged position in a social structure of relations, not their human capital endowments. By actively managing social networks, key barriers to the advancement of women in science can be overcome.

Multiteam systems (MTS) are two or more teams that interface directly and interdependently in response to environmental contingencies toward the accomplishment of shared goals.68 Although the teams pursue different proximal goals, they share at least one goal. In MTS, the effectiveness of cross-team processes is important for MTS effectiveness.69

Cross-business unit collaboration is central to the creation of value in large organizations. An investigation of 6 organizations examined how business leaders create valuable collaborations. Overall, a business unit–centric process led by multibusiness teams of general managers led to better collaborations than corporate-centric process.70

Teams can introduce structural strategies for integrating with other teams. Exhibit 10-9 reviews seven structural solutions for integrating across teams within an organization. Three of the strategies focus on integrating between teams. Four of the strategies focus on integrating across multiple teams and components of a business unit.

In one investigation, some team members left their own team to temporarily join another group. Even though this type of boundary spanning often increases cognitive growth, teams may exhibit a cognitive territoriality in response to the presence of others. For example, old-timers in teams use a greater number of their own ideas after newcomers arrive than before. Moreover, after they return, itinerant members are perceived as highly involved, but they are also perceived as more argumentative than they were

before leaving, and although they produce more unique ideas than indigenous members, their contributions to the team project are perceived as less valuable. As a result, itinerant group members have less direct influence after their boundary-spanning stint than they did prior to it!71