Which team effectiveness models fit a leadership team?

What is a team effectiveness model?

A team effectiveness model names what makes a team effective and how those things connect. Team research is more than 50 years old and runs to thousands of studies (Kozlowski and Ilgen, 2006). Much of that research rests on one frame (Ilgen et al., 2005).

That frame is input-process-output. Inputs, such as who is on the team and what it must do, shape how it works, which shapes outcomes (Mathieu et al., 2008). Later reviews widened the middle of that chain and added time (Ilgen et al., 2005; Mathieu et al., 2008).

Ilgen and colleagues proposed an alternative they called IMOI: input, mediator, output, input (Ilgen et al., 2005). The middle holds mediators: what members do, feel and think together, not only the processes they follow (Ilgen et al., 2005). The final input marks a loop, because what a team achieves shapes its next cycle of work (Ilgen et al., 2005).

This frame helps you read every other model. Each one picks a part of the loop to work on. Some start from inputs, some from processes and some from outcomes.

Which research-based models fit teams at the top?

Hackman's five conditions

Hackman's model of team effectiveness starts from one claim: no leader can make a team perform well. What a leader can do is create conditions that raise the odds (Hackman, 2002; Coutu, 2009). He named five: a real team, a compelling direction, an enabling structure, a supportive organisational context and expert coaching (Hackman, 2002).

In a real team, membership is clear: everyone can say who is in and who is out. A compelling direction means members understand and accept the work they share. Coaching helps most when it is aimed at the whole team, at its start, its midpoint and its end (Coutu, 2009).

Wageman's six conditions for senior leadership teams

Wageman, Nunes, Burruss and Hackman applied this thinking to senior leadership teams. They drew on a study of more than 120 top teams (Wageman et al., 2008; Coutu, 2009). They describe six conditions. Three are essential, and three enable the team to make the most of them.

The same research found that membership at the top is often unclear. Almost every senior team thought its boundaries were clear. But when members described their team, under 10 per cent agreed about who belonged to it (Coutu, 2009).

What is the evidence for the conditions?

The Team Diagnostic Survey asks members about their team's design, the support it gets and its leadership (Wageman, Hackman and Lehman, 2005). The authors judged its psychometric properties satisfactory, using answers from 2,474 members of 321 teams in varied organisations.

A field study of self-managing teams compared two things leaders do (Wageman, 2001). Both shaped how well teams managed themselves and how satisfied members were. But only how leaders designed the team affected its task performance. Good coaching also helped well-designed teams more than poorly designed ones. These were not leadership teams, so carry the finding over with care.

Hackman and Wageman (2005) also set out a theory of team coaching. It focuses on what coaching does for a team, when it is most likely to work, and when it is unlikely to help. They offered it as a model to be tested, not as a finding (Hackman and Wageman, 2005).

Is your leadership team a real team?

Katzenbach called the top management team a myth. In his view, few groups at the top function as real teams, even in well-run companies (Katzenbach, 1997). With Smith, he had defined a team as a small group of people whose skills complement each other. They share a purpose, goals and an approach, and hold themselves mutually accountable (Katzenbach and Smith, 1993).

A working group is different. Its performance is what its members achieve as individuals, while a team also produces collective work products (Katzenbach and Smith, 1993).

Katzenbach did not treat the single-leader group as a failure. Such a group moves quickly and efficiently, provided its leader truly has the best grasp of the task (Katzenbach, 1997). His advice was to choose the discipline that fits the work in front of you.

He also explained why teaming at the top is hard. Purpose and goals are broad, members are chosen for position rather than skills, and time is short. Executives are also used to individual accountability, and most distrust mutual accountability (Katzenbach, 1997).

What do the popular models get right, and what do they miss?

The models below appear on most lists. Each is described first in its author's terms, then set against the evidence.

Tuckman's stages: forming, storming, norming, performing

Tuckman reviewed 50 articles on group development and proposed four stages: forming, storming, norming and performing (Tuckman, 1965). Each stage has an interpersonal side and a task side. A later review added a fifth stage, adjourning (Tuckman and Jensen, 1977).

Tuckman was frank about his base. Therapy groups were over-represented in the studies, and natural groups were under-represented (Tuckman, 1965). He called the model a conceptual statement, open to further test. Of 22 later studies, only one set out to test it directly (Tuckman and Jensen, 1977).

Gersick followed eight project teams from first meeting to deadline and found a different pattern (Gersick, 1988). Each team settled on an approach in its first meeting and kept it for half its time. At the midpoint of their calendars, the teams changed course in a concentrated burst.

She called this punctuated equilibrium. It came from an exploratory study, and she said it should apply only to groups working to a deadline (Gersick, 1988). Two of her lessons may still help a leadership team: first meetings set lasting patterns, and a midpoint is a natural time to review.

Lencioni's five dysfunctions

Lencioni stacks five dysfunctions in a pyramid (Lencioni, 2002). Absence of trust sits at the base, then fear of conflict, lack of commitment, avoidance of accountability and inattention to results. He presents the model in a fable about an executive team. What research says about each part is in five dysfunctions of a team.

GRPI: goals, roles, processes and interpersonal relationships

GRPI stands for goals, roles, processes and interpersonal relationships, in that order of priority. Plovnick, Fry and Rubin set out this order in 1974, building on Beckhard's four areas of team concern (Plovnick et al., 1974). Goals come first, then roles, then procedures, and relationships last.

Their argument was that many interpersonal problems are really symptoms of unclear goals, roles or procedures. They tested a programme built on this order in 18 teams, and called the results preliminary. The teams reported clearer goals, roles and procedures (Plovnick et al., 1974).

The Big Five in teamwork

Salas, Sims and Burke distilled teamwork research into five core components (Salas et al., 2005). They are team leadership, mutual performance monitoring, backup behaviour, adaptability and team orientation. They rest on coordinating mechanisms such as shared mental models, closed-loop communication and mutual trust.

The authors offered the model with propositions for future research to test. They expect the components to matter differently over a team's life and across tasks (Salas et al., 2005). The model is strong on how members coordinate, and says little about purpose or membership.

Google's five dynamics

Google's Project Aristotle studied 180 of its own teams, in engineering and in sales (re:Work, 2015). It listed five dynamics in order of importance: psychological safety, dependability, structure and clarity, meaning and impact. The way members worked together counted for more than who the members were. Effectiveness was judged by ratings from executives, leaders and members, and by sales results.

Google states its own limits. Some factors, such as team size, showed no significant effect at Google, which it says does not make them unimportant elsewhere (re:Work, 2015). It also says your organisation's findings may differ. For the top team, see psychological safety in leadership teams.

How do you judge whether a team is effective?

A model is only useful against a test of success. Hackman proposed three: the team's output, its ability to work together in future, and members' individual development (Haas and Mortensen, 2016).

In their own studies, Haas and Mortensen found three of Hackman's conditions still especially critical: compelling direction, strong structure and supportive context. Their fourth condition is a shared mindset. It guards against “us versus them” thinking and incomplete information (Haas and Mortensen, 2016).

A leadership team serves the whole organisation, so that is where its output is judged. Drath and colleagues offer one test of that: does leadership produce direction, alignment and commitment (Drath et al., 2008)? To measure your own team, start with how to assess a leadership team.

How do the team effectiveness models compare?

How do you choose a model for a leadership team?

Start from the setting. Most members lead a unit and answer for it, and give the team only part of their time (Katzenbach, 1997). Senior executives often focus on their own roles more than on what the team must do together (Wageman et al., 2008). Membership is often less clear than the team believes (Coutu, 2009).

That setting favours models that start with conditions. Behaviour models assume a real team with shared work, and at the top that is the first thing to check. For how conditions, processes and outcomes fit together, see leadership team effectiveness.

Haas and Mortensen recommend a quick rating every few months, and a deeper look when trouble shows (Haas and Mortensen, 2016). To write down purpose and membership, use a leadership team charter.

What can you do this month?

• Have every member write down, alone, who belongs to the leadership team. Put the lists side by side at your next meeting.
• Sort this quarter's agenda into two piles: work that needs a collective work product, and work one leader can own.
• Ask each member to rate the team on Wageman's six conditions, then discuss the lowest-rated condition together.
• Mark the midpoint of your current planning cycle, and book a short review of how the team is approaching its work.
• Close one meeting by asking which of Hackman's three tests the team passes, and which it does not.