There are many insightful researchers and theorist who come from other disciplines, but have much to offer knowledge managers. Richard Hackman is one of those. Hackman, a Harvard University Professor in Social and Organizational Psychology, is an acknowledged expert on teams and teamwork. He has studied airline cockpit teams, high tech teams, orchestras, chamber music groups, bank teams, and many others. In this post, I am reviewing his last book (he died in 2013) in which he turned his in-depth understanding to study teams in US intelligence agencies. His established reputation gave him access to that very difficult-to-breach classified environment. His study looks at how teams of intelligence analysts learn from each other in order to tackle the threats of international terrorist and other risks to our security.
I have followed his work for many years, but this latest book was of particularly interest to me because of the five years I spent working on knowledge management projects within one of those agencies, The Defense Intelligence Agency (DIA). There I saw first-hand the difficulties that teams of intelligence analysts face in sharing knowledge within their own agency, not to mention the enormous problems of sharing between agencies. Along with Adrian Wolfberg, I wrote and article, Speaking Truth to Power, about that work.
Hackman’s book is filled with practical understanding about teams. In this post I will focus on a major finding Hackman made from his studies, the 60-30-10 rule. A rule that will probably surprise you. This is what it says:
- 60 percent of the difference in how well a team eventually performs, depends on the pre-work the leader does before the team even meets for the first time!
- 30 percent depends on the quality of a team’s launch, leaving just
- 10 percent to depend on the leader’s hands-on, real-time coaching.
So, what is this pre-work that makes such a difference? Hackman names six issues a leader needs to address even before a team starts its work – six issues that make a difference in team effectiveness
1.The first issue is designing a “real team,” rather than a team in name only. Hackman has a specific meaning for a real team. It is one where team members “work interdependently to generate a product or service for which members have collective, rather than individual accountability. And the team has, at least moderate stability, which gives members time to learn how to work well together.”
Hackman finds that in many situations teams are put together when it would be better to have individuals working on a task. It is the leader’s job to be sure a team is needed, not just that it is the fashionable thing to do. In general, teams are required when 1) more resources are needed than any individual working alone has available, 2) a diversity of resources are needed in terms of varied knowledge, skill, experiences, and external relationships, and 3) the task is wider in scope than an individual could address. When these requirements are not present it is better not to assign the task to a team, because a team created without a real need will invariably run into difficulty. “The benefits of teamwork come only when capable people work together interdependently to achieve some collective purpose.”
2. The second issue is that the leader specify a compelling purpose or end toward the team is working; one that energizes team members and fully engages their talents and is “consequential for the achievement of some larger aspiration.” He explains that if the purpose is compelling, then the team’s motivation is internal and therefore less need for the leader to intervene to attempt to motivate the them. This diagram looks at the leader's role in terms identifying the ends, and means. The optimum is in the upper right corner, labeled “self-managing.” The leader has provided a purpose or direction for the team that is clear, palpable and incomplete. It needs to be incomplete because there necessarily must be room for the team to make sense of the goal for themselves. Hackman says, “sense-making is an essential part of coming to ‘own’ a piece of work, and an overly explicit statement of direction can preempt that process.” Thus, in the upper right box the end is specified but the means are left to the team. The lower right box, labeled “Wasted human resources” is a situation where the leader specifies both the end and the means. Hackman notes, “Expecting teams to follow a defined set of steps assumes that the solution can be completely known in advance and the path to that solution can be charted fully ahead of time,” whereas a complex goal requires that teams continually make adaptations, reflect on the outcome, and then make course corrections. In the upper left box neither the ends or means are specified resulting in "Fragmentation" as a team attempts to figure out what they should be doing. Hackman calls the lower left hand box, the “Worst” because a team is told to how to do something with no understanding of the why behind it.
3. Putting the right number and the right mix of people on the team is the third issue. The right mix includes 1) task expertise, 2) the ability to work collaboratively with others, while keeping the team 3) as small and 4) as diverse as possible. In terms of the right number, Hackman cautions, “Work teams in organizations generally are substantially larger than they really need to be and therefore encounter problems that they do not need to have.” “Work groups of 10 or more are certain to encounter free-riding or “social loafing” problems.” Elsewhere, Hackman has said the ideal number of team members is six, large enough to have needed diversity and small enough for a team to coordinate among members.
In terms of the second criteria, being able to work collaboratively with others, Hackman notes, “Even a single member who is a ‘team destroyer,’ someone whose behavior is egregiously disruptive and destructive, can cause a team to fail.” He cautions that the only way to know if a person has those unwanted characteristics, is to talk with past team mates of the individual, relying on how the individual interacts with a manager, is not a good way to make that assessment. Excessive homogeneity can also cripple a team because if everyone has the same perspective, the team risks performing poorly on non routine tasks that require original ideas or approaches. “Well-composed teams have a good mix of members, people who are neither so similar to one another they duplicate one another’s resources nor so different that they are unable to communicate or coordinate well.”
Hackman cautions against adding members when a team falls behind, because adding new members requires taking the time to bring them up to speed and to reconfigure roles to accommodate them, which almost always has the opposite effect. He quotes Fredrick Brooks’ law, formulated from his involvement in IBM’s OS/360 programming effort. Brooks' law says, “Adding manpower to a late software project makes it later.”
4. The fourth issue a leader needs to address is helping the teams specify clear norms of conduct for team behavior. “The greatest advantage of teamwork is that team members have diverse information and expertise that, if properly integrated, can produce something that no one member could possibly have come up with.” But good use of team members’ information and expertise does not happen automatically.
A great deal of research has shown that in teams 1) higher prestige member’s voices tend dominate, 2) team members are likely to give special weight to contributions by other members from their own discipline, while discounting contributions made by people from other disciplines, and 3) the risk of professional embarrassment causes members to be reluctant to acknowledge they do not know how to do some aspect of the work or to appear to be uncertain about something. All of these very understandable human interaction shortcomings and biases, prevent all the knowledge of a team from being considered and thus defeat the primary purpose of utilizing a team. Establishing team norms reduces the effects of these shortcomings and biases allowing the team to capture and use all potentially valuable ideas and perspectives.
“Norms of conduct specify what behaviors are, and are not, acceptable in a team. Having clear, well enforced norms greatly reduces the amount of time a team must spend actively managing member behavior. The best norms promote continuous scanning of the performance situation and the deployment of work strategies that are well tuned to the special features of the team’s task and situation.” Examples of such norms are, “Hold off submitting RFIs [requests for information] until we all agree about what we most need to find out.” or “Hear out the ideas of others without interruption or contradiction.”
It is important for the team to identify and agree upon such norms at the launch of the team, because, later, when members are fully occupied with urgent subtasks they are unlikely to be willing to take the time to develop norms.
5. Providing a supportive organizational context is the fifth issue a leader needs to address. That context includes 1) having the material resources needed to carry out the work of the team, 2) a reward system that provides recognition and positive consequences for excellent team performance, rather than individual performance, (3) an information system that provides the team with the data and the information-processing tools members need to plan and execute their work, and 4) how the organization’s educational system makes available to the team needed technical and/or educational assistance.
Hackman describes a situation from his study of airline crews to illustrate the four aspects of a supportive context. He depicts a situation in which a plane is coming in for a landing. The captain has pulled the lever to lower the wheels, when a light goes on indicating that the wheels have not locked. The captain turns over control of the plane to the first officer while she deals with the problem. First, the captain calls the airfield to terminate the landing and to request a vector where she can work on the problem clear of other traffic. The airfield quickly responds, providing a clear space. While the captain is doing that, the crew brings up a gear problem checklist on the aircraft computer and goes through it step by step. These steps, including resetting the circuit breaker for the indicators, do not extinguish the offending light. Visual inspection by the crew through the cockpit floor shows that the gear is down, but there is no way to know if it is locked. The captain calls the FAA inspector to ask if he has any suggestions, but he doesn’t. She radios the dispatch desk to report the problem and then company maintenance for consultation. The maintenance supervisor calls a technical specialist with the airline manufacturer and patches him into the conversation, but the additional checks and procedures still do not solve the problem. Eventually, everyone agrees the crew should attempt a landing. The captain tells the lead flight attendant to prepare the passengers and calls approach control to request a straight-in approach with fire equipment standing by just in case. She then flies the approach. The gear holds and the landing is routine. After the passengers have departed, the FAA inspector turns to the crew and says, “Nice flying guys, very professional.”
In this scenario, all four of the suggested supports are readily available. 1) There are “ample material resources for use in carrying out the team’s work, ranging from the availability of extra fuel on demand, to the presence of fire equipment at the arrival airport.” 2) The team, not just the captain, is recognized for excellent performance in a difficult situation and the recognition occurs in real time. 3) The captain is able to call on a number of information sources, including the FAA inspector, the maintenance supervisor, and the airline manufacturer, not to mention the readily available check lists. Finally, the team has been prepared through classroom training as well simulators to deal effectively with emergencies.
Hackman notes three ways to go wrong in giving recognition and reinforcement for task-performing teams. “The first is to ignore successes ('That’s what we expect from our teams so further comment is unnecessary') but to call out failures.” He notes that strategy encourages risk aversion. “A second way to go wrong is to identify the person whom one views as mainly responsible for the team’s success and single out that individual for special recognition.” The third way is go wrong is to provide specific, concrete objectives for the team’s work, reinforced by strong incentives contingent upon achieving those objectives. This way risks inviting unintended and unwanted secondary outcomes, that occur because the team figures out ways to game the system, for example, by overlooking troublesome data or violating expected standards of behavior.
6. The sixth issue for a leader to address is making competent team-focused coaching available to the team. Hackman suggests that teams are open to coaching during three time periods, 1) when they are just starting their work, 2) then half way through their work, and 3) when their work is finished or nearly finished. He quotes Gersick, a researcher who has tracked hundreds of teams whose periods of work ranged from just a few days to several months. She found that each team “developed a distinctive approach toward its task immediately…. and then stayed with that approach until precisely half-way between its first meeting and its project deadline." At that point, “all teams underwent a major transition altering member roles and behavior patterns, re-engaging with outside authority figures or clients, and exploring new strategies for proceeding with the work. Then, following that midpoint transition, teams entered a period of focused task execution that persisted until very near the project deadline, at which time a new set of issues having to do with the termination process arose and captured members’ attention.”
Using Gersick’s model, Hackman shows that a team is particularly open to coaching at the beginning, especially interventions that orient members to each other, define member roles and assisting them in coming to terms with task requirements. A leader’s coaching at this point can foster collective motivation. Hackman notes that this is best accomplished by a face-to-face meeting. By midpoint, teams have gained a sense of what has been working well and what not. They are ready to reflect on how things have been going. However, he notes that, “midpoints are not good times for teaching team members new skills.” If they discover they lack skills it is better to locate external people to lend a hand. The end calls for a systematic debrief to learn from their own experience. A debrief can help members learn how to better deploy member talents in future tasks, by addressing questions like, “What resources went unused?” or “What would we do differently another time?” Hackman acknowledges that the major challenge is to get the debriefing to happen, because If the team was successful they are more likely to be interested in celebrating and if it did less well, they are more likely to just want to move on to the next task. But when held, there are critical lessons relevant to the development of individuals and certainly to their understanding of team work.
In Collaborative Intelligence, there is a wealth of information about how to help teams become more effective, from the most celebrated researcher of teamwork. In this blog post I’ve reported on the six issues that Hackman says cause 60% of team effectiveness. The sixth issue has also said a bit about the 30% of the launch and the 10% of coaching. But there is more to say about both that I will leave to another post.