Managing Collaboration at the Point of Execution:
Improving Team Effectiveness with a Network Perspective
University of Virginia and Visiting Research Professor Grenoble Ecole de Management
379 Rouss and Robertson Hall
Charlottesville, VA 22904
One Rogers St
Cambridge MA 02142
Advanced Human Technologies
580 California Street, 16th Floor
San Francisco CA 94104
Whether selling products or services, making strategic decisions, delivering solutions, or
driving innovation, most work of any substance today is accomplished by teams.
However, since the early 1990s, teams have evolved from more stable groups where
members were co-located, dedicated to a common mission, and directed by a single
leader to more matrixed entities with colleagues located around the world, juggling time
between several projects, and accountable to multiple leaders. As teams have become
more fluid, substantial challenges have been posed to traditional advice on team
formation, leadership, roles, and process. In this article, we describe how leaders at all
levels within an organization can obtain innovation and performance benefits by shifting
focus from forming teams to developing networks at key points of execution.
Managing Collaboration at the Point of Execution
“Not everything is or will be a team. Increasingly we don’t have teams here so much as
groups that need to form, get their work done and disband or move onto the other three
teams they are on. This flies in the face of a lot of the advice on building team harmony,
vision and other things that we just don’t have time for. Its like we call them teams but
they aren’t really in the conventional sense of the word…we need new ways of working
with these groups…” Executive in a Global Healthcare Organization
Many executives have turned to team- or matrix-based structures over the past 15
years,i and this trend does not appear to be slowing. A recent Gartner report concluded that
in the future, “[t]he primary work unit in the enterprise will be the virtual ‘matrixed’ team,
which is composed of diverse competencies, knowledge and capabilities, and assembled to
meet specific project goals or ongoing process deliveries.”ii Other research across a large
number of organizations reveals that 63% of new product development teams will be
geographically distributed within the next few years, with 22% expected to be globally
dispersed.iii Still other studies show that more and more corporations are turning to
teams as a way to organize white-collar and professional work.iv
Yet although teams have become increasingly prevalent, evidence is mounting
that they also generate hidden costs of collaboration, lengthy decision cycles, and
diffusion of focus throughout an organization.v A part of this problem is driven by the
changing nature of work. Over the past two decades, waves of restructurings have pushed
work and the coordination of work into informal networks within and between
organizations.vi At the same time, globalization has substantially changed how and where
work gets done as well as introduced myriad cultural and logistical challenges for teams
distributed around the globe. And although broadly available communication
technologies support virtual collaborations, they also carry a cost in the myriad and
instantaneous demands these technologies place on team members’ time and attention.vii
On the backdrop of this new work environment, a core problem is emerging in that
advice on team effectiveness comes from an era where people could commit substantial
time and focused effort to one team. With sufficient boundaries protecting team members’
time on specific projects, more traditional advice on team formation,viii leadership and
roles,ix group process/dialogue,x and organizational designxi makes sense. Yet as this
reality has passed, the utility of these ideas has fallen off. Teams today are frequently
formed and disbanded rapidly, distributed across multiple sites, and composed o
simultaneously working on myriad projects, with different bosses competing for their
attention. Further, these teams’ work increasingly demands substantial coordination and
integration of specialized expertise within and outside of the team.
In this context, shifting from interventions that create cohesive teams to ones that
enable rapid formation and dissolution of networks at the point of need can produce
dramatic performance impact. Instead of broad brush interventions to improve overall
team cohesion, a network lens enables managers to make changes at network inflection
points to, for instance, re-align the flow of information and decisions with the strategic
goals of the team, or connect internal communication clusters with external sources of
knowledge. Our work with 20 organizations and more than 50 teams has demonstrated
consistent ways network analysis yields unique insights on team performance.xii In the
next section of this article we introduce network analysis and then review six critical
relational dimensions that enable team leaders to visualize their team and intervene in
very different ways than conventional advice on team-building would suggest.xiii
Applying a Network Lens to Team Effectiveness
ONA, also known as Social Network Analysis (SNA), is an established set of
methods and statistics for eliciting and analyzing relationships between people such as
“who obtains information from who”, ”who trusts who”, or “who is aware of who’s
expertise”. Data are typically collected through surveys in which each person in a team is
asked questions such as, “Please indicate the degree to which you typically turn to each
person below for information to get your work done.” A single survey might ask one or
several of these kinds of questions along with demographic information such as
hierarchical level, company tenure, work location, job function and other individual
descriptors of team mates. For each network question an analyst can generate a diagram
showing where there are connections between people and conversely, where those
connections are lacking. In addition to visualizations, a highly advanced body of
statistics can be used to measure group cohesiveness, dispersion of information or level
of reliance on a small number of people (See Appendix 1 for a summary of steps entailed
in a typical ONA).
By focusing on the right internal and external networks, this perspective gives
leaders a very granular means of promoting team effectiveness. Below (and summarized
in Table 1) we outline six key questions a network perspective enables teams to address
to improve effectiveness.
|Editor’s Note: Insert Table 1 about Here|
Are the right voices influencing team trajectory? Traditional advice suggests
that performance results when the right expertise is on a team with strong leadership and
well-defined process and content roles.xiv With sufficient time and predictable problem
domains, leaders can cultivate and gain commitment to a team vision as well as match
content and process roles and accountabilities with team member expertise. Yet few team
leaders have this luxury anymore. Most are dealing in very nebulous problem domains
with little surety on the outcome or ability to even predict who will be on or off the team
in coming weeks and months. Rather than designing a vision, process and structure for a
known future, team leaders are better served by applying a network lens to ensure that the
right expertise is being brought to bear at the right point in time – a difficult challenge as
teams have become larger, virtual, cross-functional and frequently staffed with members
not dedicated to one effort.xv
There are two basic challenges to team effectiveness on this front. First, how does
a leader in a distributed or large team know who is influential and if the right expertise is
being brought to bear? All too often, certain people – typically those who are loud, who
have the leader’s ear, or whose expertise was good for past purposes – become too
prominent in myriad but seemingly invisible collaborations. Typically many interactions
occur outside of formal meetings and drive a team on a solution trajectory that
undermines results possible from more balanced collaboration leveraging the full team’s
expertise. Cliques also rapidly form and preclude integration of expertise – creating an
invisible barrier to innovation and execution that the team was formed to bridge in the
first place. In these cases, mapping information flow and problem-solving collaborations
and then coloring nodes in the network by technical competencies allow a leader to
ensure the right expertise is influential in ideation and execution.
Second, how does a leader ensure the right balance of reliance on formal structure
(to ensure consistency and efficiency) and informal structure (to ensure innovation)?
Although climate or team development surveys can indicate that a team has become too
rigid or hierarchical, these assessments often do not give sufficient insight on what to do
outside of platitudes on participative leadership and delegation. Network analysis lets a
leader see where a team is falling into a routine of relying on roles – whether the team
leader or others – and so potentially not leveraging the best expertise or running into
invisible bottlenecks. Now more than ever, leaders need to ensure that influence and
decision authority flows to the right people in a network depending on needs at a given
point in time. A network view helps determine if this is happening, provides insights on
where role redesign or coaching can decrease the team’s reliance on voices good for past
purposes, and ensures that leaders are letting go and followers are taking courageous
action (rather than elevating all things up the hierarchy) when they need to be influential.
Is the team “appropriately” connected for the task at hand? Although
executives acknowledge the importance of collaboration, the tendency is to take either a
“more is better” or “ad hoc” approach to collaboration. Each philosophy can hamper team
effectiveness via unproductive network patterns. First, one of the legacies of the advice
industry built up around teams lies with a heavy emphasis on consensus and participative
leadership. These approaches were intended to expand the pool of ideas and expertise
brought to bear as well as encourage member’s personal commitment to team goals.
However, rather than the right voices being heard, excessive consensus building too often
results in slow decision making, too many meetings consuming people’s time, and a
sense of entitlement to participate in all aspects of decision making. A network
perspective can allow leaders to visualize information flow and time spent in
collaborations in order to make targeted decisions on where excessive collaborations are
draining group effectiveness.
The second problem arises from allowing collaboration to occur in an ad hoc way
that can lead to invisible barriers to team effectiveness. Research over the past 20 years
has consistently shown that people who make targeted investments in relationships
perform better than those who simply build ever larger networks.xvi The same general
results apply to teams as well.xvii Yet unfortunately team leaders too often do little to
build the right patterns of connectivity and so allow team networks to fall into
unproductive collaborations constrained by formal structure,xviii demographic similarity
(or homophily),xix and personality.xx Although leaders cannot make people become
friends, they can use network information to change staffing, team meetings, and a range
of communication vehicles to ensure that homophily, lack of time, organizational
pressures, and inertia do not drive teams into biased or ineffective networks.
A network perspective helps ensure that the right collaborations are occurring rather
than allowing overly connected or ad hoc networks to evolve. To do this, leaders (often in
conjunction with the team) first identify the ideal network that needs to be in place at a
given point in a team’s lifecycle. Then, by comparing existing collaborative patterns to the
desired network, they can make targeted shifts that both build out needed relationships and
decrease time spent on unproductive collaborations. The ideal network can be identified in
several ways: 1) in smaller teams, a leader can run a facilitated exercise to brainstorm the
network that needs to be in place at a given point in time; 2) in larger or distributed groups,
a leader can embed survey questions into a network diagnostic to identify the ideal pattern;
and 3) more systematically – as some pharmaceutical and electronics companies have done
in their new product development efforts – organizations can profile high-performing team
networks at key points in a process and so provide teams with an archetype for success.
Regardless of process, the ability to adjust connectivity in a targeted fashion can
dramatically improve team effectiveness.
Has the team cultivated important external relationships? Another legacy
from decades of focus on team process and roles is that most advice on teams is inward-
focused and promotes insularity. Of course, leaders intuitively know that teams live in a
context demanding effective coordination with key stakeholders. They also all know that
there are better ideas and practices outside the team and organization that they need to
find and leverage. But when the rubber meets the road, the pervasive tendency is to focus
inside the team on tasks, roles, and process. More recent research has shown that
effective teams pay targeted attention to external relationships critical for expertise,
decision making, resources, and political support.xxi A network perspective – applied in a
way that enables a leader to see key external ties – helps ensure the right range of
relationships are developed and that a team is not overly focused on specific
First, an external view of collaboration can improve ideation and the quality of
solutions a team generates by ensuring the right external ties are cultivated and the best
and most relevant expertise is brought to bear. This is a process of building networks that
enable a desired innovation to occur. It begins by ensuring that the right external ties have
been created and that no substantial gaps exist in the external network to make sure that
the right kind and caliber of expertise is being sought. Far too often, external sourcing of
expertise is done in an ad hoc way through personal relationships, which dramatically
sub-optimizes both the creativity side of an innovation as well as the implementation
where resources and other partners can speed execution and time to market.
Second, a network lens helps ensure coordinated effort in execution. For example,
consider a common problem most organizations have that results from multiple touch
points and lack of coordinated effort in delivering offerings to key accounts. In the
traditional view of key account management, an account manager acts as a central
coordinator of an account team, which is composed of specialists who represent different
products, services, or divisions across the organization. However, misaligned priorities
and diverse reporting lines mean that communication is often poor and that account
strategies are frequently not well executed or understood. A network perspective to
account management recognizes that appropriate resources across the entire organization
need to be accessed and brought to bear, often on an ad hoc basis, to capitalize effectively
on opportunities that emerge at the client. Taking a network approach to account teams
also facilitates the balancing act between sufficient team communication and over-
communication. Team members must be familiar with each others’ activities on the client
account in order to present a coordinated approach to the client and to offer relevant
products and services; however, excessive meetings and other traditional approaches to
account team structure result in inefficiencies and can impact morale.
Finally, an external network focus ensures efficiency in execution via knowledge
transfer from similar projects. Transfer from external groups is best accomplished
through a combination of weak ties to search for new knowledge and strong ties to
facilitate the transfer of complex knowledge.xxii Similarly, people who straddle multiple
groups are in an advantageous position from which to select and then transfer good ideas
from external sources.xxiii An external focus is as important for routine project execution
as it is for innovation. Thus, although it may be tempting to have teams be internally
focused to meet deadlines, in the long run too few external connections can put the team
at a disadvantage for obtaining important information. Instead of leveraging existing
knowledge, internally focused teams may turn to published sources that may not be as
relevant or may be harder to associate with the team’s work. Or, in an attempt to be
focused on the work, they may ignore other information, ultimately reducing the quality
of the team’s work.
Are value-added collaborations occurring in the team network? Traditional
team advice focuses on building commitment to a goal and then relies on roles, process and
structured meeting formats to ensure information flow and decision making are occurring
as required in a team. In general, the tendency is to believe that more communication is
better – but there is rarely any focus on the quality of the information or knowledge moving
in the team. As outlined previously, a network perspective can be specifically applied to
these dimensions (information and decision making) to help a team leader target
inefficiencies. But network analysis can also help a leader go a step further in assessing and
then promoting effectiveness of collaborations within a team.
First, from a pure performance perspective strengthening networks isn’t about
simply increasing interactions; it’s a product of increasing productive interactions and
reducing unproductive ones. By applying a network lens to their teams, leaders can assess
where value and costs from collaboration reside and improve connectivity at points that
add economic value. Depending on the kind of team and its focus in the organization, this
typically occurs in one of two key ways. First, team leaders can assess value creating
interactions with network questions that 1) identify and facilitate productivity via best
practice transfer and/or knowledge sharing and/or 2) uncover collaborations that underlie
revenue growth and can be replicated at targeted points to generate greater value. Second,
and just as importantly, team leaders can assess time spent in interactions and convert this
into an understanding of collaborative costs and their drivers. Natural solutions emerge
with this lens that let leaders see where decision rights, accountabilities, and meeting
format shifts can move a team from costly gridlock to execution.
But beyond ensuring that interactions in a network are value-added from a
performance perspective a network lens is also very powerful in showing where
interactions with teammates serve to either engage or disengage people from team goals
and objectives. To this end, team leaders often will map at least one more emotional
aspect of a network – aspects such as personal support, career advice, a sense of purpose,
or, most commonly, enthusiasm in a group. Spend some time in most any organization
and you are sure to hear people talk about the level of enthusiasm or energy associated
with different people or projects. In some instances, a team in which ideas flow freely
and its members build effortlessly on one another’s work will be described as “high
energy.” In others, a particularly influential person may be known as an “energizer” –
someone who can spark progress on projects or within groups. And, of course, on the flip
side are the people who have an uncanny ability to drain the life out of a group.
Applied to this concept of enthusiasm or energy, a network perspective provides a
vantage to help leaders take the pulse of their group by simply asking a network question
to the effect, “When you interact with this person how does it typically affect your energy
level (positive, neutral or negative)?” In this way, leaders can find those people who
energize a group and/or find where enthusiasm for a given course of action exists. As a
team leader vying for people’s time and attention, it provides a non-trivial way of
understanding those who are capturing the hearts (as well as the minds) of the team, those
who are becoming less engaged, and those who may unintentionally be having a negative
impact by virtue of how they interact with the group. Importantly, this is not just an
interesting concept but something that team leaders can and have had an impact on. It
turns out that managing energy comes down to a set of finite behaviors that can be
developed in a team and that can have a substantial impact on performance and
innovation once a leader can visualize where opportunities for improvement exist by
taking a network perspective.xxiv
Do underlying relationship qualities yield effective collaboration at the point
of need? Traditionally, team communications have focused on face-to-face and virtual
communications that first serve to build harmony and joint commitment to team goals
and then serve to ensure execution against a project plan. Early on, team-building
activities such as ropes courses or falling into teammates’ arms are meant to inspire trust
and commitment to colleagues. Although important to creating relationships, outside of
the general term of team- or trust-building, we have little evidence of what they actually
build into a network that improves collaboration. By mapping the dimensions of
relationships that precede or lead to effective knowledge sharing, team leaders have a
much better view of ways to improve collaboration at the point of need. Specifically, a
network lens can help team leaders focus on development of two relational
characteristics: 1) awareness of teammates’ expertise and 2) trust in teammates’ abilities.
First, newfound teammates can be useful in solving problems only if the team
develops an awareness of their expertise. This awareness determines whether and how
expertise on a team is leveraged when a new problem or opportunity comes along. Creating
awareness of “who knows what” and “who knows who knows what” helps a team respond to
opportunities more seamlessly.xxv And importantly, rather than increasing communications
or the volume of meetings once a leader knows where lack of awareness of colleague
expertise is hurting team collaboration they can take very targeted and efficient actions to
improve. Rather than consume everyone’s time in more meetings, assessing and then
building awareness of expertise through staffing, skill profiling, and other vehicles create a
latent network of awareness where people can tap relevant expertise at the right point in time.
Second, in addition to meta-knowledge of expertise in a group, another key
determinant of whom a person seeks out and listens to in the face of a new problem is
trust. Research shows that trust leads to increased overall knowledge exchange,xxvi makes
knowledge exchanges less costly,xxvii and increases the likelihood that knowledge
acquired from a colleague is sufficiently understood and absorbed that a person can put it
to use.xxviii In a team context, two dimensions of trust are important to knowledge
creation and sharing: benevolence (“You care about me and take an interest in my well-
being and goals”) and competence (“You have relevant expertise and can be depende
upon to know what you are talking about”).xxix Benevolence-based trust is what most of
the team-building activities actually build and is important in a network because it allow
one to query a colleague in depth without fear of damage to self-esteem or reputati
addition, though – and frequently overlooked by traditional team-building activities –
people must also trust that the person they turn to has sufficient expertise to offer
solutions. Competence-based trust allows one to feel confident that a person sought out
knows what he or she is talking about and is worth listening to and learning from.
Mapping this dimension of trust in a network allows leaders to identify gaps and then
consider a range of behavioral interventions to improve connectivity on this front.xxx
Does organizational context support collaboration and momentum? With
minimal fixed-asset or geographic constraints in knowledge-based work, teams and team-
based designs often seem a natural and easy solution for improving collaboration and
productivity. However, experience has proven the transition to teams a difficult one,
requiring infrastructure realignment and development of appropriate cultural values and
leadership skills. Drains on productivity and morale will occur if a team-based design is
inconsistent with an organization’s strategy, information and performance management
systems, leadership style, or employee skill base. Yet altering any of these elements
requires significant time and cost to develop, implement, and administer.
The single greatest factor driving the degradation of organizational networks is
staff turnover. However morale, overwork, internal competition, and other factors largely
driven by leadership can also negatively impact people’s ability or willingness to connect
with others in the support of organizational objectives. Factors and initiatives that support
network development include formal structure, work processes, development activities,
and culture, as shown in Table 2. Effectively implementing the relevant initiatives will
create positive network development and team execution.
|Editor’s Note: Insert Table 2 about Here|
A Network View of Three Prevalent Types of Teams in Organizations
In the following vignettes we have relied on ONA to highlight a different means
of visualizing team improvement opportunities. In Table 3, below, we summarized the
key drivers and actions typically taken for the three types of teams we will review.
|Editor’s Note: Insert Table 3 about Here|
Sales Teams. Since the 1980s, many industrial and professional services
organizations have implemented account team structures to serve key clients. Typically,
these teams cross divisions, product groups, service lines, geographic location (often
across multiple countries), industry specialization (for complex clients), and levels of
seniority and are considered the center of complex business-to-business sales.xxxi Clearly,
multiple and diverse resources need to be brought to bear to make sales and build
relationships. Yet traditional account team structures and account planning practices have
somehow not matched the complexity of identifying and exploiting emerging
opportunities in the highest potential accounts. In fact, research has shown no correlation
between the use of teams and sales performance on key accounts.xxxii
We typically find that the difference between high- and low-performing account
teams in an organization is not a result of product knowledge, sales behaviors, or more
traditional account planning practices. In most organizations, key account teams have the
same training and avail themselves to the same technologies. As a result, team-building
or sales-training interventions often pale in comparison to the improvement potential
uncovered by visualizing network drivers of account team success. Instead, our results
more generally show that the difference between high and low-performing account teams
can be traced to the pattern and quality of the teams’ relationships with others inside and
outside the team.
On a network pattern level it turns out to be very important that teams not evolve
into ‘cliques’ or sub-groups, that they not over-rely on a given person (either expert or
leader) and that they manage a balance of external ties bridging to key stakeholders. On
a network content level our results show that high performing teams are distinguished not
just by more active information flow networks but also networks showing greater
awareness of other’s expertise, more fluid decision-making interactions and competence-
based trust. While individual competency and training matter to keep teams from landing
in the bottom 20% of performers, it is almost always the above network dimensions that
turn out to be key predictors in distinguishing the top 20% of performers.
With account teams, ONA helps leaders assess three high leverage points for
1) Quality of relationships between the account team and client. The appropriate
level of high-quality client connections across relevant team members allows for
better client service, discovery of cross-selling opportunities, and decreased
susceptibility to departure of key individuals on both sides of the relationship.
2) Quality of relationships within the account team. A team fluidly connecting
key expertise and roles is much more efficient and effective in identifying and
capitalizing on sales and delivery opportunities.
3) Quality of relationships connecting the team with the host organization. A
broad network of connections back into the organization allows an account team
to leverage scale in a large organization and so materialize the right products,
services, and expertise for clients in a timely and efficient fashion.
Consider one Fortune 250 company’s efforts to raise effectiveness of low-
performing account teams by assessing and replicating networks that high-performing
teams naturally cultivated. In this organization, cross-functional teams were assigned to
top accounts, with the mission of improving customer satisfaction and increasing sales by
optimizing existing product lines, introducing new products, and improving processes
between the organization and its key clients. These teams were assigned to the top 18
accounts at this company, each of which generated 2% to 5% of total firm revenues. In
this case, a sales-effectiveness program focused on defining network differences of high-
and low-performing account teams and putting in place tools (e.g., collaboration analysis
processes, technology) and support mechanisms (e.g., training, account planning
processes) to replicate important networks dimensions in lower-performing teams.
Let’s compare the network of one account team generating top revenue and
consistently strong customer satisfaction ratings with the network of another that
underperformed significantly across most performance measures. First, when evaluating
the relationships between the account teams and the client, in each case, client contact
was heavily dependent upon two salespeople. However, in the more successful account,
there were several supporting relationships with other members of the team. As seen in
Exhibit 1, the high-performing team had formed collaborative relationships with multiple
clients, frequently turning to them for support or product information. In stark contrast,
the underperforming team reached out to only two clients regularly for informational
purposes (the 12 clients identified along the upper left-hand side had informational
relationships with the account team only on a very infrequent basis).
|Editors Note: Insert Exhibit 1 about Here|
And it was more than just the volume of connections that distinguished the high-
performing team. This unit also had qualitatively better relationships with the client,
which could be seen with a network lens. The most significant benefit that both account
teams received from their extended networks (people outside of their team) was the
ability to assist with specific sales opportunities. However, an almost equally important
benefit cited by the successful team was “contacts to other relevant parties in the
organization.” Not only were the members of the higher-performing team better
connected among themselves and with their clients, but they also relied much more
heavily on their networks to make connections that provided access to additional sales
opportunities within the client.
|Editors Note: Insert Exhibit 2 about Here|
Second, moving from the client interface to the internal network within the team,
the ONA revealed that the underperforming team was much more formal and structured,
with interactions conducted via telephone and planned meetings, whereas the successful
team communicated much less formally, through impromptu meetings, e-mail, and instant
messages. This more seamless collaboration in the high-performing team was enabled by
members having greater awareness of colleagues’ skills and expertise. Although both teams
were small, the ONA revealed that very few people on the underperforming team were
aware of the knowledge, expertise, and capabilities of others on the team. In contrast, the
successful team had extremely high levels of awareness across all roles and so naturally
morphed to client sales opportunities much more effectively.
On both teams, the two most central people were important in holding the
network together. But the higher-performing team had greater lateral connectivity, which
helped to better serve the customer. These non-hierarchical networks also decreased
susceptibility should highly connected team members leave or be promoted. For example,
on the underperforming team, if the two most central people left the team, connectivity
would fall off by 76%. The effect on the successful team is less pronounced—losing the
two most central people would reduce connectivity by 55%.
This is a common predicament for account teams that focus too heavily on leaders
or certain experts and unwittingly create susceptibilities and hurt overall performance.
For example, take the case of a global information technology (IT) services company that
used ONA to optimize touch points with a key client across dimensions such as seniority,
roles, functions, and expertise to more effectively deliver services. In this case, the ONA
surprised the IT services account executive on two important fronts. First, client contact
was spread across many more people than expected, at all levels of the team. In fact, the
account executive was astonished at the number of people with client relationships and
the urgent need for an effective communication plan to ensure that everyone was aligned
with the client’s objectives and delivering a consistent voice. In this case, the very
externally focused players turned out to be desktop technicians who provided support to
key client stakeholders – revealing a lever of influence that was entirely unknown to the
account manager and a lack of influence at more senior levels on his team that
dramatically concerned him.
A second point of interest for the IT services account manager lay with the
networks of the seven project managers charged with cultivating relationships with other
areas of the IT services organization and helping to cross-sell services and deepen
account penetration and revenue. Of these seven leaders, two were barely connected back
into the IT services organization, and four were clearly overloaded by demands from
internal team members but did not realize the impact their workload and accessibility
were having on the rest of the team. In this case, when the account manager saw who one
of his top network enablers was, he panicked, as this person was about to go on a leave of
absence. With the ONA, however, he was able to avoid disruption by understanding this
person’s relationships and the benefits she provided through them and then pairing her
with several “high-potential” team members ready to take on client responsibilities
during her leave.
Innovation Teams. History teaches us that most breakthrough innovations are re-
combinations of existing ideas or technologies, the integration of which occurs through
teams and informal networks.xxxiii Although traditionally these networks have formed in
very serendipitous ways, it is increasingly important to cultivate and “manage” lateral
and external connections for more effective innovation. Success often comes from
targeted initiatives ensuring connectivity among those with the right expertise in a given
domain and those with the right influence in the organization—the people who have a
unique ability to get things done by virtue of their position in the network. Rather than
sequestering small teams with a charge to generate a blinding insight or engaging in yet
another corporate restructuring to break down silos, leaders can use ONAs to mobilize
networks of relevant expertise. Network dimensions such as the following can
dramatically improve innovation success:
1) Staffing innovation teams with brokers from broader informational
networks. By mapping networks of groups charged with innovation (e.g., R&D
units or learning-oriented alliances) we can staff innovation teams with brokers –
those well positioned to be able to take an idea from one domain and see its
potential for application in another domain – and greatly facilitate integration of
expertise in breakthrough innovations.
2) Targeted development of external ties for decision purposes. Obtaining
relevant buy-in and decision acceptance from both formal and informal leaders in
the broader organizational network dramatically improves speed and fidelity of
execution as an innovation evolves from ideation to implementation.
3) Team networks that produce creative friction and innovation breakthroughs
through recombination of existing expertise and resources. Mapping the
quality of ties among those with core expertise ensures that the right combinations
of knowledge and resources are integrating to generate new product or service
Consider a leading food organization that produced a dramatic innovation
breakthrough by working through networks more effectively. Producing substantial
innovation in the confection category is a real challenge, as the top 20 brands in the
category have remained remarkably consistent over the last 35 years. It is rare for a new
confection to last past 3 to 5 years. It is even rarer for a new business model to be
introduced into the category. The MyM&Ms business of Masterfoods – the group that has
developed and now markets customized M&Ms – provides an example of how mobilizing
internal and external networks can efficiently produce such breakthrough innovations.
The spark for this innovation occurred in the Advanced Technology team in
R&D, which was charged with developing new technologies for the Chocolate Business
Unit. The team was intrigued by the opportunity to innovate the appearance of the unit’s
products, inspired by the legendary “m” printed on M&Ms. The first innovation came
with learning how to ink jet a picture onto a tablet of chocolate. The team was successful
in developing a food-grade system to print digital photos directly on the surface of a
chocolate bar. However, although a technical success, beta testing with consumers did
not show enough volume to justify launching the product on a large scale.
In parallel, though, another group within the same team was charged with printing
the faces of the M&M characters onto M&Ms. The “faces” were printed on M&Ms by
the use of offset lithographic technology. This technology is traditionally used to print
onto confections, but not to the level of resolution required to print high-quality faces on
the curved M&M surface. In this case the market tests showed high interest in the
product, but the cost of creating new print wheels prohibited small runs, keeping the
Each team had an incremental success that was not viable in the market; yet their
combined insights had the potential to generate a substantial breakthrough. However,
although co-located, the teams did not communicate in a way that allowed them to share
their work in sufficient detail to see the connection between the new technology and the
opportunity to print different logos on M&Ms. The “ah ha” moment on the technology
side of MyM&Ms came with the realization that the ink jet technology from chocolate
tablet printing could be combined with the concept of printing on M&Ms. This created
the ability to print text on M&Ms at low volumes, allowing the company to cheaply
customize the print on the M&Ms piece and create the innovative business concept of
selling custom-printed M&Ms direct to consumers. The catalyst for the breakthrough
occurred when the manager, a broker in network terms who was familiar with both
projects, brought the two groups together by “translating” how the technology innovation
could be applied to printing on M&Ms.
Several other networks needed to be built before the product could enter the sales
pipeline. First, the development team needed an external network of suppliers to provide
key elements of technology. To address this issue, the team created a unique external
partnership network composed of an ink jet print head supplier, an ink company, and a
printer frame fabricator. Two of these partners were already working with Masterfoods in
other areas; one was completely new. Getting the innovation accepted and to market
would not have been possible without this external network that helped to make the
technical dream a reality. Second, an additional internal business network was created at
Masterfoods that would take the technical innovation from prototype to market. Most
innovations falter getting from prototype to a marketable product. However, several
factors made this case a success. While the technology was being developed, the
Masterfoods management team was looking for a new venture to be an example of
business model innovation for the rest of the firm to emulate.
MyM&Ms was a completely new type of business model for Masterfoods, one that
promised high profit potential. However, resistance to the new business model from the
financial community in this firm was deep and well entrenched, as the financial model was
completely different to the firm’s core business. Traditionally, the CEO would have had to
dictate that all functions support the initiative; however, the team knew that this sort of
mandate often entrenched resistance. Instead, the team took a different approach to
implementing the innovation, one that leveraged individual networks. The executive
sponsor of the venture deliberately avoided a hierarchical dictate and reached out to the
personal network he had created with the finance leadership to get buy-in. Specifically, he
asked them to lead beta teams throughout the enterprise in testing out the new business
model internally. The use of the finance network as part of the effort to test the new
business resulted in reduction of internal barriers and enabled the beta test of the new
business to be implemented in 60 days, a breathtaking improvement over the normal 2.5-
year new product introduction time.
Execution teams. Execution teams can include project teams in professional
services, software development teams in technology, and service teams designed to
provide technology support to internal IT customers. These teams generate value through
the creation and delivery of services and so depend on effective and efficient
coordination to execute on time and within budget.xxxiv In most cases, coordination issues
are addressed by developing and applying a process that governs how work will be done
across individual roles and responsibilities. Yet process maps alone do not accommodate
the informal collaboration necessary for coordination and are often inflexible when it
comes to exceptions or unexpected changes. As a result, a network perspective applied to
execution teams helps to reveal opportunities for improvement on key network
dimensions, such as the following:
1) Building mutual awareness of current work and expertise. Greater awareness
of skills and expertise helps in execution and also informs members of colleagues’
work, speeding coordination of effort.
2) Formation of cohesive, specialized subgroups knit together by technical
brokers. Large execution teams form into largely autonomous subgroups to focus
on specialized work. In these cases, broker roles evolve and are critical to
bridging sub-teams to allow for efficient execution and then integration of highly
3) External relationships for product/service adaptation. External ties provide a
way to get independent evaluation and calibration of the work to improve
acceptance and satisfaction of the product.
Effective networks are especially important for software development teams, who
need to coordinate work smoothly to build a single coherent system whose parts fit
seamlessly together. Increasingly, these teams are staffed globally by people from
multiple countries to take advantage of cost structure and available expertise. Traditional
approaches to software teams utilize software development processes and tools that
decompose the work and work assignments to manage the complexity of the project and
accommodate work being done in multiple locations. As a result, the overall structure of
the software is handled through a high-level design and architecture, but responsibility
for the detailed coordination is often handled by individual team members. Despite the
use of these processes and technologies, software development teams turn in poor
performance resulting from costly rework, reduced quality, and lower customer
satisfaction. Taking a network approach to team structure and team dynamics provides a
perspective into the factors that help to make a team successful.
Consider a technology company where software teams develop custom
application and web projects for internal and external clients. We observed three teams
that used common software development processes and tools but that managed to also
deliver their projects on time, on budget, and with a high level of both client and team
satisfaction. Several network factors helped these teams succeed.
First, the people on these teams were aware of the knowledge and current work of a
large number of their teammates. Awareness is an important relational concept because it is
a necessary condition for communication and coordination. In large software development
teams composed of disparate expertise we find it common for people to be aware of the
knowledge and skills of 25% to 35% of others on their team. In contrast, the ONA from the
teams we studied revealed a much higher level of awareness; team members were aware of
the knowledge and skills of approximately 56% of their teammates and they were aware of
what approximately 50% of their teammates were working on (in teams ranging in size
from 23 to 83 members). This level of awareness meant that people had ready knowledge
of where to find the right expertise within the team and could more easily coordinate their
work than if they relied on written documentation, which is often out-of-date and
incomplete. Several factors contributed to the high level of awareness. First, several people
on each team had previously worked together on projects which allowed them to become
familiar with each other’s knowledge and skills. Second, these teams placed a lot of
emphasis on face-to-face meetings that brought in people from another site. Third, they
made extensive use of communication technologies, especially instant messaging, which
promoted both local and remote collaboration.
The ONA also revealed unique ways that structuring work into smaller, largely
autonomous, cohesive subgroups to execute specialized tasks while also maintaining
bridging relationships across subgroups ensured efficient execution and coordination/
integration. Teams often make the mistake of increasing the amount of communication
without thinking through who really needs to be involved in decisions, who needs to send
information, and who needs to receive it. Two of the software teams in this organization
used the ONA to achieve a good balance of communication by grouping themselves into
small subgroups; this allowed them to complete a lot of work in parallel and meet some
aggressive schedules. One of the teams, however, had a more centralized structure, in
which communication went through the project manager rather than through lateral
connections and within subgroups. When we presented the results to this project manager,
he was not surprised that his team was different because he had noticed several recent
communication problems. Seeing the results from the other teams led him to reorganize his
team around a more modular structure by reassigning some of the development roles.
A third factor setting these teams apart was that individuals had strong ties with
people outside the team that they used to get technical information and feedback to
improve the quality of their work. The individuals were especially likely to go outside the
team for specific technical information and help related to programming rather than for
general information. In some cases, people sought information from outside to avoid the
embarrassment of revealing ignorance to their teammates. But in other cases, they were
reaching outside the team to augment the internal knowledge or to canvas a wide range of
people for information. When it comes to staffing projects, many managers focus on
acquiring people with the best skills for the work, forgetting that knowledge is just as
often acquired from other people and that it is not necessary for those other people to be
part of the team for the knowledge to be useful.
As with other teams, both internal and external connectivity is critical to overall
effectiveness in execution teams. Internal relationships, especially awareness of others’
work, facilitate the informal coordination necessary for rapid and high-quality
development. Team members also mined their personal external relationships as a source
of valuable knowledge and information beyond what they could or were comfortable
getting from team members. We observed several management practices that facilitated
awareness and self-organization, but one stands out. Despite the challenges of being
distributed, the project managers of one of the teams made a point of arranging for the
developers to visit the client, who was in another country, at the client’s site to witness
the project requirements firsthand and to develop a stronger relationship and allegiance
with the client. To further strengthen that commitment, the project managers also
arranged for members of the client team to visit the developers at their site. These mutual
face-to-face meetings had a profound effect on clarifying the context of the requirements
and created a level of trust that resulted in much richer and more productive
communications between the development team and the client.
Today, teams are more important than ever in generating and delivering value for
an enterprise. But changes in the economy, organizational structures, work practices, and
technologies require rethinking traditional approaches to managing successful teams. In
this paper, we highlighted how a relational view taken by sales, innovation and execution
teams in a range of industries offers unique insights into team effectiveness that
traditional advice on teams would miss. Of course, the network lens is not just analytic;
it creates actionable insights that are practical and make a difference to the team and its
performance. Armed with new knowledge and insights about their teams, leaders can and
do take a range of actions that better position their teams to deliver on key goals. In sum,
the network perspective uniquely enables leaders to position for success via actions
targeting networks at the point of execution rather than excessive reliance on team
development, process, and roles that were more effective in times gone by.
The Changing Face of Teams
Team Levers Traditional View Network View
Are the right voices
Leader as ultimate decision
maker and direction setter.
Process and content roles in
team provide structure.
Decision making and direction
setting influence shifts based on
expertise. Leader and followers set
climate for shifts in responsibility.
Is the team
connected for the
task at hand?
Information and decision-
making networks either over-
connected or hierarchical.
Information and decision-making
networks are focused on an
archetype for success based on
point in process.
Has the team
Not heavily emphasized either
to others within an
organization or to experts
outside an organization.
Heavily emphasized and targeted
both within and outside
organization to bring the best
expertise to bear.
occurring in the
Principally just information
A focus on value added interactions
in terms of both performance and
team members’ engagement.
Communication focus on joint
commitment to goals,
benevolence-based trust and
group process and harmony.
Information focus on awareness of
expertise, timely accessibility,
competence-based trust, and
execution of commitments made to
factors are key to
Matrix reporting structures,
collaborative technologies, and
Positive organizational network
momentum, supported by consistent
meaningful exposure to others’
collaborative technologies, and
Organizational Context Supporting Positive Network Momentum
Examples that Support Positive Network Momentum
Formal structure • Decentralized decision rights
• Latitude for work to be performed outside formal
• Recognized broker or liaison roles
Work processes • Diverse team structures designed to fill network gaps
• Flexible workflow
• Use of collaborative technologies
Development activities • Systematic exposure of capabilities across organization
• Social and professional development activities facilitate
• Supports the development of external relationships
Culture • Supports ad hoc collaboration
• Recognizes external ideas and relationships
• Promotes development of personal expertise
Summary Issues Driving Team Performance
Team Issues driving team
performance Management actions based on
Sales • Identify opportunities that lead
to new sales
• Build strong client relationships
• Optimize sales of existing and
new product lines
• Identify key people who may be
getting overloaded and find
alternatives, such as emerging high
performers, who can relieve the key
person and develop their own skills
• Identify key influencers and target
with retention programs
• Build better lateral connections
within the team and between the
team and client
Innovation • Create new offerings and
• Streamline approval and
• Actively support key brokers who
provide access paths to other groups
and who are internal promoters of
ideas and reputation with internal 3rd
• Ensure links exist for broader buy in
and decision acceptance outside the
Execution • Timely coordination of
technical knowledge across
distributed sites to avoid costly
• Access to expertise inside and
outside the team
• Acquisition of clear and
• Provide team members with tools
that support the strategic use of
technology for collaboration across
remote sites. But make it a priority
to hold face to face meetings with
remote team members to build
awareness and commitment
• Leverage the personal networks of
team members as access points to
external knowledge and make
High- and Low-Performing Account Team Information Flow
“Please indicate the extent to which you turn to each person listed below for market or product expertise or
information to service this customer.” Responses of somewhat frequently to very frequently.
Primary Network Benefits from Relationships outside the Account
Please identify up to 15 people who are not within the specified core account team or key customer
contacts, but who are important in providing a high quality of service to the customer. Of these people,
please indicate the primary and secondary benefit you receive.
Primary and Secondary Benefit Received From People in Personal Network
Underperforming Team High-Performing Team
1. Industry or market trends that suggest opportunities with the customer
2. Specific opportunities that exist (or could exist) at this customer
3. Political awareness in terms of who is influential or things to avoid in conversation
4. Contacts to other relevant parties in the organization
5. Activities of organizations similar to the customer
6. Activities of competitors to the customer
Appendix: Conducting a Network Analysis
An Organizational Network Analysis (ONA) is conducted to collect data that reveal the
normally invisible relationships between people. These relationships bind people together
in ways that impact a team’s performance. A typical ONA can be divided into 4 phases.
Step 1: Determine the key strategic issues and identify the audience
Sponsorship. Before embarking on an ONA, it is important to identify a sponsor
preferably a high level executive responsible for the team. This person provides the
rationale for the assessment, linkages to other on-going initiatives and resources.
Translate key strategic objectives into specific network questions. An ONA
begins by identifying the key business objectives for the team. For example, a sales
team looking for ways to increase revenue might ask whether the right people are
connected to form solutions. Alternatively, an innovation team looking to develop
new products might ask whether people from different business units are
collaborating sufficiently. These key objectives are translated into specific network
questions that form the core of a survey administered to everyone in the team.
Select the audience. One of the key decisions in conducting an ONA is selecting the
people who are going to participate in the data collection. In the case of an ONA
done with a team, the decision focuses on how broadly the team is construed. Should
ONA only include the core members of the team or be broadened to include extended
team members? Should executives or even clients outside the team be included?
Answers to these questions depend on what information needs to be acquired to
address the key business challenges faced by the team.
Step 2: Construct survey and administer assessment
Compose survey questions. The key business problems are translated into questions
that are asked of everyone on the team about everyone else (bounded network
questions) or about people outside the team whose names are provided by the
respondents themselves (personal network name generator). Each of these questions
represents a different relationship that holds between people and it is important to
make these questions reveal actionable relationships. In seeking to understand
whether a sales team is structured to create cross-product solutions, the survey might
include the following questions:
These questions are accompanied, in the survey, with a roster of names against which
the respondent will provide answers to questions like the below.
How often do you talk with the following people regarding <solution>
Please indicate how likely you are to turn to each of the following people
before making a decision about <solution>
The pattern of answers to these questions reveals whether the right people are
communicating effectively with each other or if there are strategic gaps in
communication that can impact the ability of the team to create an integrated solution.
Personal network name generator
Personal network questions are especially useful for extracting information about
people who are outside the team or organization. In these questions, respondents
generate names of people in response to questions such as:
Please provide the names of up to 10 additional people who you regularly
communicate with to develop <solution>
Personal network name interpreter
The elicitation question is followed by additional questions about the relationships
and characteristics of each person listed. Common interpreter questions may include:
Please indicate where each of these people are located
Please indicate the hierarchical level of each person
Please indicate the primary benefit that you obtain from interactions with
Finally, the survey includes a range of relevant demographic and attribute questions
Please indicate your primary office location
Please indicate which business unit you are in
Please indicate how many years you have been in your current job
Taken together, the responses to the network and attribute questions provide
actionable insights into the patterns of communication and information sharing by
people in different locations and business units.
Administer survey. The completed survey may be administered in paper form, or,
increasingly, using a survey tool which lets users take the survey online. Unlike
traditional surveys, ONA surveys need a high response rate – we typically shoot for at
least a 75% response – to ensure accurate representation of the network.
Step 3: Analyze and interpret the data
Analysis. The response data from the survey are analyzed using a combination of
tools to generate the network diagrams and perform statistical analysesxxxv.
Quantitative Measures: Team Level
The statistical analysis is used to compute the extent to which everyone in the
group is closely connected and information is readily and easily transferred using
measures which include:
Density. Density measures the number of ties (connections) that exist divided
by the total number of ties possible if everyone on the team were connected to
Cohesion. Cohesion is a measure of the average number of ties that are
traversed from any person in the group to any other person.
Quantitative Measures: Individual level
Individual measures typically include degree centrality which identifies which
people are central to the team and betweenness which identifies those that sit on
the shortest information path between team members.
Degree Centrality. Degree centrality measures the number of ties into each
person (in-degree centrality) and the number of ties from each person out to
others in the team (out-degree centrality)
Betweenness Centrality. Betweenness measures how often each person lies
on the shortest path from any person to any other person in the team
Interpretation. Network analysts use these quantitative measures to identify
potential problems in the network. Is too much information going through one
person who could become a bottleneck preventing the efficient flow of
information or decisions? Is there a group of people who have lots of connections
with each other but few connections to anyone else? If these people also have
some attributes in common such as all belonging to the same geographic location
or the same business unit it could signal that there are barriers preventing people
from sharing what they know or supporting people to share within their group.
Step 4: Create recommendations for remedial action
The network analyst highlights the strengths and shortcomings in the team based
on team members’ interpretation of the data. But when it comes to taking action,
a plan is best developed with the participation and input from the team and the
sponsor as well as the network analyst. Remedial actions can range from ones
taken by any individual in the team – such as making a point to reach out to
people in other locations, actions taken by the team leader – such as re-aligning
roles and responsibilities to take advantage of network connectivity, to more
comprehensive changes in the overall incentive and reward structure.
i Specifically a number of influential practitioner books emerged in the early to mid-1990s after
the reengineering craze with advice on how best to manage and support teams in organizations.
Examples of this work include Katzenbach, J. & Smith, D. (1993). The Wisdom of Teams: Creating
the High-Performance Organization. New York: Harper Business, pp. 11-19 & 98-104. Mankin, D.,
Cohen, S., & Bikson, T. (1996). Teams and Technology: Fulfilling the Promise of the New Organization.
Boston: HBS Press. Mohrman, S., Cohen, S., & Mohrman, A. (1995). Designing Team-Based
Organizations: New Forms for Knowledge Work. San Francisco: Jossey-Bass. Parker, G. (1994). Cross
Functional Teams: Working with Allies, Enemies and Other Strangers. San Francisco: Jossey-Bass.
Wellins, R., Byham, W., & Dixon, G. (1994). Inside Teams: How 20 World-Class Organizations Are
Working through Teamwork. San Francisco: Jossey-Bass.
iiBell, M.A. (2004). Leading and Managing in the Virtual Matrix Organization. Gartner Research
iii McDonough, E., Kahn, K., and Barczak, G. (2001). An Investigation of the Use of Global,
Virtual, and Collocated New Product Development Teams, The Journal of Product Innovation
Management, Vol. 18, No.2, pp. 110–120.
iv Espinosa, J. A., Cummings, J. N., Pearce, B. M. &Wilson, J. M. (2002). Research on teams with
multiple boundaries. Proceedings of the 35th Hawaii International Conference on System Sciences.
vBell, M.A. (2004). Leading and Managing in the Virtual Matrix Organization. Gartner Research
vi Cross, R. & Parker, A. (2004). The Hidden Power of Social Networks. Harvard Business School
vii Eid, T. (2005). Market Trends: Content, Communications and Collaboration Technologies in the
High-Performance Workplace. Gartner Report.
viii Dyer, W. (1995). Team Building: Current Issues and New Alternatives. Reading, MA: Addison-
Wesley. Klimoski, R. & Jones, R. (1995). Staffing for Effective Group Decision-Making: Key Issues
in Matching People and Teams. In R. Guzzu, E. Salas and Associates. Team Effectiveness and
Decision-Making in Organizations. San Francisco: Jossey-Bass. Larson, C. & LaFasto, F. (1989).
TeamWork: What Must Go Right/What Can Go Wrong. San Francisco: Jossey-Bass.
ix Hackman, J. R. (1990). Groups that Work (and Those that Don’t): Creating Conditions for Effective
Teamwork. San Francisco: Jossey-Bass. Sundstrom, E. (1999). Supporting Work Team Effectiveness:
Best Management Practices for Fostering High Performance. San Francisco: Jossey-Bass. Goodman, P.
(1986). Designing Effective Work Groups. San Francisco: Jossey-Bass. Orsburn, J., Moran, L.,
Musselwhite, E. & Zenger, J. (1990). Self-Directed Work Teams: The New American Challenge. New
x Donnellon, A. (1996). Team Talk: Listening between the Lines to Improve Team Performance. Boston:
Harvard Business School Press. Dougherty, D. (1992). Interpretive Barriers to Successful Product
Innovation in Large Firms. Organization Science, 3(2), pp. 179-202. Edmondson, A. (1998).
Psychological Safety and Learning Behavior in Work Teams. Administrative Science Quarterly.
Senge, P. (1990). The Fifth Discipline: The Art and Practice of the Learning Organization. New York:
DoubleDay Currency pp. 198-202.
xi Mohrman, S., Cohen, S., & Mohrman, A. (1995). Designing Team-Based Organizations: New Forms
for Knowledge Work. Hackman, J.R. (1990). Groups that Work (and Those that Don’t): Creating
Conditions For Effective Teamwork. San Francisco: Jossey-Bass. Denison, D., Hart, S., & Kahn, J.
(1996). From Chimneys to Cross-Functional Teams: Developing and Validating a Diagnostic
Model. Academy of Management Journal, 39(4), pp. 1005-1023.
xii In our research we applied network analysis to 53 teams coming from the following industries:
(12 teams) consumer products; (8 teams) software development/technology; (11 teams)
consulting; (9 teams) manufacturing and (13 teams) pharmaceuticals/bio-tech/healthcare. In this
process, we employed a case-based logic in data collection by doing semi-structured interviews
guided by a pre-existing theoretical model (Per Yin, R.K. (1994). Case study research: Design and
methods (Rev.Ed.). Newbury Park, CA: Sage) that we held ‘loosely’ to allow for inductive theory
development (Per Glaser, B. & Strauss, A. (1967). The Discovery of Grounded Theory: Strategies for
Qualitative Research. Hawthorne, NY: Aldine de Gruyter and Lincoln, Y., & Guba, E. (1985).
Naturalistic Inquiry. Beverly Hills, CA: Sage). Our initial framework was informed by streams of
research on teams, social capital, social network analysis, transactive memory/distributed
cognition and communication studies. With each team we first applied a network survey that
obtained background information and then data on each respondent’s unique set of relationships
(Per Marsden, P.(1990). Network data and measurement. Annual Review of Sociology 16: 435-463).
In the surveys we followed a standard two-step name generator/interpreter methodology to
elicit and then characterize respondents’ relations both within and outside of each team (Per
Scott, J. (1990). Social network analysis. Thousand Oaks, CA: Sage Publications and Wasserman, S.,
& Faust, K. (1994). Social network analysis: Methods and applications. Cambridge, UK: Cambridge
University Press.) To ensure reliability, questions were specific and provided detail as to the
construct of interest while also assessing typical interactions (Per Freeman, L., Romney, K. &
Freeman, S. (1987). Cognitive structure and informant accuracy. American Anthropologist, 89: 310-
325), as research has demonstrated respondents to be poor at accurately recalling interactions
occurring in specific time intervals (Per Bernard, H.R., Killworth, P. & Sailer, L. 1982. Informant
accuracy in social network data V: An experimental attempt to predict actual communication
from recall data. Social Science Research 11: 30-66). Interviews were then conducted with team
leaders and/or executives to assess key interventions. These 30-90 minute interviews used in-
depth and semi-structured techniques to allow informant latitude in responses (Per Miles, M. &
Huberman, A. (1994). Qualitative Data Analysis 2nd Edition. Thousand Oaks, CA: Sage; Patton,
M.Q., (1987). How To Use Qualitative Methods in Evaluation. Newbury Park, CA: Sage and
Marshall, C. & Rossman, G. (1998). Designing Qualitative Research. Thousand Oaks, CA: Sage).
xiii Cross, R. & Parker, A. (2004). The Hidden Power of Social Networks. Cambridge, MA: Harvard
Business School Press. Krackhardt, D. & Hanson, J. R. (1993). Informal Networks: The Company
behind the Chart. Harvard Business Review, 71, pp. 104-111.
xiv Katzenbach, J. & Smith, D. (1993). The Wisdom of Teams: Creating the High-Performance
Organization. New York: Harper Business, pp. 11-19 & 98-104. Mankin, D., Cohen, S., & Bikson, T.
(1996). Teams and Technology: Fulfilling the Promise of the New Organization. Boston: Harvard
Business School Press; Mohrman, S., Cohen, S., & Mohrman, A. (1995). Designing Team-Based
Organizations: New Forms for Knowledge Work. San Francisco: Jossey-Bass pp. 63, 82-87, 181-185 &
xvBell, M. A. (2004). Leading and Managing in the Virtual Matrix Organization. Gartner Research
xvi For some of the more classic references to this research, please see Brass, D. (1984). Being in the
Right Place: A Structural Analysis of Individual Influence in an Organization. Administrative
Science Quarterly, 29, pp. 518-539. Burt, R. (1992). Structural Holes. Cambridge, MA: Harvard
University Press. Burt, R. (2000). The Network Structure of Social Capital. In B. Staw & R. Sutton
(Eds.), Research in Organizational Behavior. New York: JAI Press, pp. 345-423. Gargiulo, M. &
Benassi, M. (2000). Trapped in Your Own Net? Network Cohesion, Structural Holes, and the
Adaptation of Social Capital. Organization Science, 11(2), pp. 183-196. Mehra, A., Kilduff, M., &
Brass, D. (2001). The Social Networks of High and Low-Self Monitors: Implications for Workplace
Performance. Administrative Science Quarterly, 46, pp. 121-146. Podolny, J. & Baron, J. (1997).
Resources and Relationships: Social Networks and Mobility in the Workplace. American
Sociological Review, 62, pp. 673-693.
xvii Cummings, J. & Cross, R. (2003) Structural Properties of Work Groups and Their
Consequences for Performance. Social Networks, pp. 197-210.
xviii Lincoln, J. (1982). Intra- (and inter-) Organizational networks. Research in the Sociology of
Organizations, 1, pp. 1-38; Brass, D. (1984). Being in the Right Place: A Structural Analysis of
Individual Influence in an Organization. Administrative Science Quarterly, 29, pp. 518-539.
Stevenson, W. & Gilly, M. (1991). Information Processing and Problem Solving: The Migration of
Problems through Formal Positions and Network Ties. Academy of Management Journal, 34, pp.
918-928; Cross, R. & Cummings, J. (2004). Tie and Network Correlates of Individual Performance
in Knowledge Intensive Work. Academy of Management Journal.
xix McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a Feather: Homophily in Social
Networks. Annual Review of Sociology, 27, pp. 415–444.
xx Kilduff, M. (1992). The Friendship Network as a Decision-Making Resource: Dispositional
Moderators of Social Influences on Organizational Choice. Journal of Personality and Social
Psychology, 62, pp.168-180; Mehra, A., Kilduff, M., & Brass, D. (2001). The Social Networks of
High and Low-Self Monitors: Implications for Workplace Performance. Administrative Science
Quarterly, 46, pp. 121-146. Casciaro, T. & Sousa Lobo, M. (2005). Competent Jerks, Lovable Fools
and the Formation of Social Networks. Harvard Business Review.
xxi Ancona, D. G. (1990). Outward bound: Strategies for Team Survival in the Organization.
Academy of Management Journal, 33, pp. 334-365. Ancona, D. G. & Caldwell, D. F. (1988). Beyond
Task and Maintenance: Defining External Functions in Groups. Groups & Organization Studies,
13(4), pp. 468-494. Ancona, D. G., & Caldwell, D. F. (1992). Bridging the Boundary: External
Activity and Performance in Organizational Teams. Administrative Science Quarterly, 37(4), pp.
xxii Hansen, M. (1999). The Search-Transfer Problem: The Role of Weak Ties in Sharing
Knowledge across Organization Subunits. Administrative Science Quarterly, 44, pp. 82-111.
xxiii Burt, R. S. (2005). Brokerage and Closure: An Introduction to Social Capital. Oxford University
xxiv Cross, R., Baker, W., & Parker, A. (2003). What Creates Energy in Organizations? Sloan
Management Review, 44(4), pp. 51-57. Baker, W., Cross, R., & Wooten, M. (2003). Positive
Organizational Network Analysis and Energizing Relationships. In K. Cameron, J. Dutton, & R.
Quinn (Eds.). Positive Organizational Scholarship. Berrett-Koehler Publishers.
xxv Hollingshead, A. (1998). Retrieval Processes in Transactive Memory Systems. Journal of
Personality and Social Psychology,74(3), pp. 659-671. Hutchins, E. (1991). Organizing Work by
Adaptation. Organization Science, 2(1), pp. 14-29. Liang, D., Moreland, R., & Argote, L. (1995).
Group versus Individual Training and Group Performance: The Mediating Role of Transactive
Memory. Personality Social Psychology Bulletin, 21(4), pp. 384-393. Moreland, R., Argote, L., &
Krishnan, R. (1996). Socially Shared Cognition at Work: Transactive Memory and Group
Performance. In J. Nye & A. Brower (Eds.). What’s Social about Social Cognition. Thousand Oaks,
CA: Sage, pp. 57-85. Weick, K. & Roberts, K. (1993). Collective Mind in Organizations: Heedful
Interrelating on Flight Decks. Administrative Science Quarterly, 38, pp. 357-381.
xxvi Cross, R. & Prusak, L. (2003). The People that Make Organizations Stop—or Go. Harvard
Business Review, 80 (6), pp. 104-112. O’Reilly, C. & Roberts, K. (1974). Information Filtration in
Organizations: Three Experiments. Organizational Behavior and Human Decision Processes, 11, pp.
253-265. Penley, L. E. & Hawkins, B. (1985). Studying Interpersonal Communication in
Organizations: A Leadership Application. Academy of Management Journal, 28, pp. 309-326. Tsai,
W. & Ghoshal, S. (1998). Social Capital and Value Creation: The Role of Intrafirm Networks.
Academy of Management Journal, 41, pp. 464-476. Zand, D. E. (1972). Trust and Managerial Problem
Solving. Administrative Science Quarterly, 17, pp. 229-239.
xxvii Currall, S. & Judge, T. (1995). Measuring Trust between Organizational Boundary Role
Persons. Organizational Behavior and Human Decision Processes, 64, pp. 151-170. Zaheer, A.,
McEvily, B., & Perrone, V. (1998). Exploring the Effects of Interorganizational and Interpersonal
Trust on Performance. Organization Science, 9, pp. 141-159.
xxviii Argyris, C. (1982). Reasoning, Learning and Action. San Francisco: Jossey-Bass. Cross, R., Rice,
R. & Parker, A. (2001). Information Seeking in Social Context: Structural Influences and Receipt of
Informational Benefits. IEEE Transactions, 31(4), pp. 438-448. Levin, D., & Cross, R. (in press). The
Strength of Weak Ties You Can Trust: The Mediating Role of Trust in Effective Knowledge
Transfer. Management Science. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An
Integration Model of Organizational Trust. Academy of Management Review, 20, pp. 709-734.
xxix In their 1995 article, “An Integrative Model of Organizational Trust,” Academy of Management
Review, 20, pp. 709-734, Roger C. Mayer and his colleagues, James H. Davis and F. David
Schoorman, identify a third dimension of trustworthiness, integrity, defined as consistently
adhering to a set of principles that the trustor finds acceptable. Integrity is clearly important in
many situations. Parties to a market exchange, colleagues counting on each other to complete
certain tasks, or subordinates committing their efforts and career progression to a superior are
surely affected by the perceived integrity of others. Yet it is not clear that seeking a person out for
information or advice is contingent on that person following a particular set of principles
consistently. For example, malevolent integrity—a condition of low benevolence and high
integrity—might apply to situations that are purely competitive, such as two boxers trying to
hurt each other but still playing by the rules. However, it is unlikely that knowledge seekers would
make much distinction between someone who is out to harm them versus someone who is honest
and consistent about an intention to harm them.
xxx Prior research has shown that those who are seen as trustworthy sources of knowledge tend to
(1) act with discretion; (2) be consistent between word and deed; (3) ensure frequent and rich
communication; (4) engage in collaborative communication; and (5) ensure that decisions are fair
and transparent. Under organizational factors, we identified two ways to promote interpersonal
trust: (1) establish and ensure shared vision and language and (2) hold people accountable for
trust. Under relational factors, there is some overlap with the trustworthy behaviors mentioned
above, but we also identified two new behaviors: (1) create personal connections and (2) give
away something of value. Finally, under individual factors, a person’s own judgment of his or
her abilities (self-efficacy) also matters, a trust-promoting behavior identified in our interviews
we characterize as (10) disclose your expertise and limitations. Abrams, L., Cross, R., Lesser, E., &
Levin, D. (2003). Nurturing Trust in Knowledge Intensive Work. The Academy of Management
Executive, 17(4), pp. 1-13.
xxxi See for example Moon, M. A. & Armstrong, G. M. (1994). Selling Teams: A Conceptual
Framework and Research Agenda. Journal of Personal Selling & Sales Management, 14 (1),pp. 17-41,
or Homburg, C., Workman Jr., J.P, & Jensen, O. (2002). A Configurational Perspective on Key
Account Management, Journal of Marketing, 66, pp. 38-60.
xxxii Workman Jr. J.P, Homburg, C., & Jensen, O. (2003).Intraorganizational Determinants of Key
Account Management Effectiveness. Journal of the Academy of Marketing Science, 31, 1, pp. 3-21.
xxxiii Basalla, G. (1988). The Evolution of Technology. New York: Cambridge University Press. Bijker,
W. E. 1995). Of Bicycles, Brakelites, and Bulbs: Toward a Theory of Sociotechnical Change. Cambridge,
Mass.: MIT Press. Hughes, T. P. (1989). American Genesis: A Century of Invention and Technological
Enthusiasm, 1870-1890. New York: Viking. Kodama, F. (1991). Emerging Patterns of Innovation:
Sources of Japan's Technological Edge. Boston: Harvard Business School Press.
xxxiv Dawson, R (2005). Developing Knowledge-Based Client Relationships: Leadership in Professional
Services, Second Edition. Burlington, MA: Elsevier Butterworth-Heinemann.
xxxv The tools in common use by practitioners include UCINet (Borgatti, S.P., Everett, M.G. and
Freeman, L.C. (2002). Ucinet for Windows: Software for Social Network Analysis. Harvard, MA:
Analytic Technologies.), NetDraw, InFlow and Pajek. More information about these tools and the
process of conducting an ONA can be found in Cross, R. & Parker, A. (2004). The Hidden Power of
Social Networks. Cambridge, MA: Harvard Business School Press.