• When the subgroups function as discussion groups, the goal is information sharing. Executives compare notes and update one another on existing accomplishments. In this mode, teams do not make strategic decisions, and thus there is no great need for active leadership; the leader tends to simply “go around the room” and keep the conversation on track.
• In single-leader units, everyone understands that there is a single boss with authority over the subgroup for this task, and the other members all have clear, stable roles with individual accountability to the leader. This is a very useful operating model when speed and efficiency are called for.
• When innovative action, group insight, or breakthrough performance is needed, “real teams” are put into action. A real team is a small group of people with complementary skills, all of whom are committed to a common purpose. They share performance goals, adhere rigorously to a cohesive working approach, and operate with flexible leadership. Anyone on the team may step forward to articulate what the next challenge is and how to meet it; team members invest their time and reputation because of their common commitment.
Most executives have experienced discussion groups and single-leader units, but real teams are comparatively rare. Being on a real team can be challenging at first for some members, who may not be used to leadership roles that shift or to having mutual accountability for the team’s results.
A high-functioning subgroup in a top team of 20 senior executives might include three or four of those individuals, along with one or two specialists or high-potential people drawn in from the rest of the organization. Because it can operate in any form — as a discussion group, single-leader unit, or real team — this subgroup can solve a variety of difficult problems and make a number of decisions rapidly that might take months for the whole executive group to work through. (A dysfunctional subgroup, by contrast, tends to remain within one mode.) Developing this proficiency takes some time and effort. Top team members must learn to make disciplined choices about which subgroups to form, who the members should be, and which type of collaboration will be required.
One company that saw the benefits of disciplined subgroups was a US$1 billion provider of IT consulting services, with 10,000 employees spread across more than 70 offices around the world. In 2005, as part of a restructuring effort, the senior vice president of human resources conducted a network analysis of the top 250 executives and managers, mapping information flows and collaborations against the generation of revenue for the firm. The analysis revealed some unexpected discrepancies. First, many of the highest-revenue-producing account managers — who had critical expertise and key relationships with clients — operated with very few internal connections. As a result, their superior expertise was seldom brought to bear in client sales and project execution. Second, the senior leadership group, which had been together for many years, had grown contentious and unproductive. The CEO had hired a team-building facilitator, and the team adopted extensive consensus and feedback processes that improved the meeting atmosphere. But these new methods also bogged down the senior group; the process for reaching a decision often took longer than discussions about the decision itself. The CEO himself found the new consensus approach to be an impediment to progress. But he did not want to return to open hostility.
Third, there were problems with the senior leadership group members. A few habitually demanded that their colleagues support their positions; this habit had undermined the others’ trust in them. Some group members were so highly networked in the rest of the company — with dozens of people coming directly to them for information — that they had become significant bottlenecks; others couldn’t get information without going through them. This caused delays in decision making, projects, and sales efforts. Moreover, their actions solidified the influence of functional silos; only 5 percent of the company’s people managed 30 percent of the revenue-producing collaborations.