Team Structure Affected Performance More than Coaching

A team's organizational structure had far more influence over its measurable performance than did coaching by its manager in a study of self-managed Xerox customer equipment repair teams. This was true whether the teams were well coached or not. "Effective coaching helps well-designed teams more than poorly designed teams, and ineffective coaching undermines poorly designed teams more than well-designed teams," wrote business administration researcher Ruth Wageman of Dartmouth College. "Self-managed" means the teams reported to a manager who was not involved in their daily activities or regularly in their meetings.

Wageman set out to avoid some of the pitfalls of field research, such as depending on team managers' and members' opinions about how their teams did, by including multiple sources and hard performance data. For example, a team was included in the study only if three managers at different levels felt it was superb or ineffective by Wageman's three-part definitions. And final performance ratings were based on data such as response and repair times, parts expenses, machine reliability, and customer satisfaction. In addition, managers and members of each team were interviewed using pre-defined questionnaires and completed 108-item survey forms. The focus in each was how often certain behaviors occurred, not what people thought of those behaviors.

The study report in Organization Science said team structure ratings were based on how many of the following were true:

• "Group has clear membership and is stable over time."
• "Statement of purpose for the team is clear and is about ends and not means."
• "Number of members is adequate for the work and no larger" (4-7 members in this case).
• "Team is composed of members with substantial (diversity) of task-relevant skills…"
• "Team has a group task such that members must work together to accomplish it."
• "Team has objective performance targets that are 'stretch' goals."
• The team has openly stated norms that members are expected to experiment with procedures, seek out best practices, take action to solve problems, and discuss "what different members have to contribute to the work."
• "80% or more of available rewards are contingent on the team rather than individual performance."
• Task-relevant information such as machine data and customer feedback is provided to the team at least weekly.
• "Educational system provides training and technical consultation."
• "Group is provided with the basic material resources it needs to accomplish its work."

"Self-management" was defined as the degree to which the team members took responsibility as a group for their work results, actively monitored their performance, and took steps to improve. Coaching behaviors were drawn from the interviews (not predefined) and seemed to clump into six categories. Behaviors that supported the team's self-management were considered positive and included rewarding the team informally for working as a group and teaching it new problem-solving techniques. Negative behaviors included bypassing the group to direct individuals or deal with customers, running group meetings, and pointing out problems in the group (rather than helping it spot them). Other coaching behaviors had no effect on self-management. The study did not compare self-managed teams to those with leaders, and higher self-management behaviors did not directly relate to performance. Instead, team design impacted both self-management and performance.

Wageman also looked at worker satisfaction, and at quality of group processes by using ratings of statements like, "Every time someone tries to straighten out a work group member whose behavior is not acceptable, things seem to get worse rather than better." Findings included:

• Leader coaching did not directly affect quality of processes, instead helping or hurting by supporting or interfering with self-management.
• Self-management, in turn, impacted quality of process.
• The link between coaching and self-management was only part of the reason that good coaching raised worker satisfaction and bad coaching lowered it, meaning coaching also had some direct impact.
• Good team structure design raised quality of group process ratings, but not worker satisfaction.
• Neither positive nor negative coaching had a direct impact on performance, but positive coaching helped group processes and negative hurt worker satisfaction.

Though not discussed by the author, her data showed that level of self-management, monitoring of performance, managing of performance, and worker satisfaction were all correlated with performance. Perceptions of group process quality were not, however.

Wageman concludes that "decentralization and major structural change" are the two ways leaders can "make real differences in team effectiveness." She says a research focus on those areas "may be more fruitful in the long term than continuing to search for the best kinds of day-to-day styles for leaders to use in interaction with their teams."

Source: Wageman, R. (2001), "How Leaders Foster Self-Managing Team Effectiveness: Design Choices Versus Hands-on Coaching," Organization Science 12(5):559.