Here are two ways a leader can get it wrong in setting a team’s purpose. Wrong Way #1: “Something is going on in that region that doesn’t seem quite right, and I’d like you all to take a look at it. Let me know what you come up with.” Wrong Way #2: “I’d like you to monitor, around the clock, all the traffic that comes across your desk about ship activities in those ports. Every morning, give me a listing of all the previous day’s movements.”
What is wrong with these two pictures? The first one is something of a projective test, an inkblot. Members have to make assumptions about what the leader is most interested in, and what they infer may or may not be aligned with what he actually had in mind. Indeed, the leader himself may not have been entirely clear about just what was needed, perhaps because he had not thought it through carefully enough beforehand. As unhelpful as this statement of purpose is, it could have been even worse. For example, the leader might have told the team to go ahead and do “whatever makes sense to promote the national interest.” That would have been an inkblot without any ink, of no use whatever to members in figuring out what they were supposed to do or how they should do it. Vague direction like that may help explain the behavior of the “rogue” intelligence teams that one sometimes reads about. For many such teams, I suspect, the problem is not that members decided on their own to head off in an unfortunate direction but instead that the team’s purpose was underspecified by the leader who assigned the team its work.
The second “wrong way” is nearly the opposite of the first: the team’s purpose is clear, specific, and boring. In this case, there is no question about exactly what the manager who formed the team wants, so all the team has to do is to execute very specific instructions. The team, in effect, is being asked to babysit a data stream, knowing that at some point information technologies will be able to perform the same work less expensively and more reliably. Computers do not get bored, distracted, or fall asleep at the monitor. Members of teams that have mindless tasks to perform sometimes do.
Establishing a good team purpose involves charting a course between the two extremes just described, finding a way to frame and communicate the work that both points the team in the right direction and fully engages its members. Our research has shown that the best statements of purpose—whether for a team of senior managers, for a front-line team, or for any team in between—have three attributes.1 Team purposes should be clear, they should be challenging to accomplish, and they should be consequential for the achievement of some larger aspiration.
The purposes of the red teams in the PLG simulations (Chapter 1) and of the exfiltration team (Chapter 4) were clear, challenging, and consequential; those of the blue teams in the simulation and the emerging threats team in the previous chapter were not. How about the two teams described just above? The purpose of the “take a look at it” team was not at all clear but was quite challenging (it always is challenging to figure out what should be written on a blank slate). The purpose of the “ship movements” team, by contrast, was crystal clear but entirely devoid of challenge. And, for both these teams, the broader consequentiality of the work was not explained and therefore was unknown to team members. Teams are served best by leaders who make sure that team purposes stand high on all three attributes.
A clear purpose orients a team toward its objective and therefore is invaluable to members as they assess and choose among alternative strategies for proceeding with the work. There are numerous choices to be made in the course of work on almost any task, and decision making about such matters is facilitated by a clear and concrete statement of direction. It’s like planning a hike. Should we take this trail, or that one? If we are clear about our intended destination we are much more likely to make a good decision about the better route to take.
Purposes such as “serving our customers” or “staying on top of developments in our sector” are so vague and general that they provide little help to a team in developing a task-appropriate performance strategy. Although there is no excuse for failing to think through what one wants a team to accomplish, some intelligence situations are so ambiguous that no amount of thought could generate clarity of purpose. What should a leader do in such circumstances? One possibility is to convert direction-setting into a two-stage process. The team’s direction for the first stage is to scope out the situation—to attempt to clarify what, if anything, merits further attention. When the scoping work has been completed, the team and the leader would reconvene to decide whether a second stage is called for and, if so, to establish the team’s new objectives. In this scenario, clarity would come not all at once but gradually as the leader and team collaborate to make sense of an initially ambiguous situation.
Sometimes the problem with a team’s direction is not that it is too ambiguous but that it is over-specified. When a team’s purpose is spelled out in exhaustive detail, there is little room for members to add their own shades of meaning and thereby make the purpose their own. 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.2 Great team leaders therefore tend to use words about team purposes that are just a bit ambiguous, and they are likely to draw on stories, analogies, and metaphors to get the point across. Such linguistic devices, far more than any specific quantitative objective, encourage members to project their own interpretations onto what is being said and to develop their own images of the end states that are sought. Good direction for a work team is clear, it is palpable—but it also is incomplete.
The best team purposes are not only clear but also require members to stretch to accomplish them. They are neither so easy that they fail to motivate members nor so demanding that they are beyond the team’s reach. Research established many years ago that individual motivation is greatest when a person has about a 50-50 chance of succeeding on a task, and there is no reason to doubt that the same is true for work teams.3 Those who set a team’s direction, therefore, should strike a balance between too easy and too hard. Achieving that balance requires a good understanding of both the work to be done and the people who will do it. That cannot be done by someone who is unfamiliar with the team or who imagines that tasking a team involves little more than sending out a memo from one’s headquarters office. Taking the time to properly set a team’s direction shows respect for a team; to task a team casually, thoughtlessly, or by remote control is implicitly to belittle the team and the work it will be doing.
One of the great benefits of a challenging team purpose is that it frees leaders from the temptation to rely either on their personal charisma or on the promise of performance-contingent rewards to motivate the team. Both of these devices can have unintended negative effects on team behavior and performance—for example, substantial motivational decrements when the charismatic leader is absent, or the displacement of focus from the actual team purpose to doing whatever is needed to get the promised reward. Motivation that derives from a challenging purpose, by contrast, can become self-sustaining as members work together to achieve it.
A team’s purpose need not necessarily be one that members themselves would have chosen or that is inherently and obviously important, such as thwarting an imminent act of terrorism. Even tracking slowly evolving economic changes in a remote part of the world can be a compelling team task if the leader has framed the team’s purpose to highlight the considerable challenge involved in generating a high-quality product that is just what policymakers need. Creative leaders formulate team purposes that elicit and focus members’ motivation even for work about which members initially may have been uninterested or skeptical.
Work that has clear consequences for other people or for achieving a major organizational objective is much more likely to engage team members’ talents than a task of little significance in the broader scheme of things. When a team’s purpose is highly consequential, members want to contribute what they know, or know how to do, to help the team succeed. And they are disposed to properly weight other members’ contributions, relying more on their teammates’ actual expertise than on purpose-irrelevant attributes such as status, gender, or equality of workload in deciding whom to listen to most attentively or to rely upon most extensively. When it is the team’s big game, the one for the championship, we put forward our most competent members and they, with our support, give their all to accomplish the team’s objective.
The work of many teams in the intelligence community is of considerable consequence—how well they perform really does make a difference in things that members and their clients care a great deal about. Expert team leaders build on that reality and reinforce with their teams the importance of the work. And they never use what I call the “temple ploy.” This involves using rhetorical devices to make a team’s purpose seem more consequential than it is, as in the apocryphal story of the leader who motivated his group of brick carriers by characterizing their work as building a temple rather than carting loads of bricks—which, of course, is what they actually were doing day after day. In many cases, the proper act of leadership is to reframe the team’s work to actually make it more consequential, not merely to describe it in ways intended to make it seem consequential.
A compelling team purpose is clear (which orients the team), it is challenging to achieve (which energizes the team), and it is consequential (which engages the full range of members’ talents). Moreover, each of these attributes reinforces the benefits of the other two—as, for example, when a challenging purpose engages members’ talents in addition to heightening their motivation. It is the constellation of the three attributes that brings the greatest-benefits to a task performing team.
Try a little thought experiment for a team on which you currently serve or, perhaps, one that you are thinking about creating. How does that team’s purpose stand on the three attributes of a compelling purpose summarized in Figure 5-1? What might be done to strengthen the team’s standing on those attributes, and what would be the effect of doing so? And, finally, what do you predict would be the consequences of reducing the clarity, challenge, or consequentiality of that team’s purpose?
FIGURE 5-1 Checklist for Assessing Team Purposes
A team’s purpose should come first when one contemplates forming a team because so many other leadership decisions and actions depend on it—how the team is structured, the organizational supports that the team will need, and the type of hands-on coaching that will be most helpful. Indeed, leaders who create compelling purposes for their teams reduce considerably the amount of attention that they must give to monitoring and managing team processes in real time. Teams that have well-crafted purposes are much more likely to competently manage their own processes than are those whose purposes are either too vague and general or too specific and detailed.
There remains the question of the proper focus of a team’s direction. In formulating a team purpose, should leaders focus mainly on what is to be achieved, the hoped-for outcomes? Or should they emphasize the procedures the team is to use in carrying out the work, on the assumption that good processes will generate good outcomes?
There is a right answer to those questions, as is seen in Figure 5-2. The best statements of team purpose are those that clearly specify the ends a team is to achieve but that leave it to the team to decide about the means it uses in pursuing those ends (the upper right cell). Team leaders should be insistent and unapologetic about exercising their authority to specify end-states—and they should be just as insistent about not specifying all the procedural details.
Considerable research affirms the advantages of focusing on ends rather than means. In a simulation of forest firefighting, for example, teams whose leaders communicated their intentions about what should be accomplished performed better than those whose leaders issued orders for specific actions. This focus on intentions rather than actions brought several advantages—it signaled respect for the teams’ competence, evened out the cognitive workload between the leader and team members, and allowed teams to take greater advantage of the local knowledge of their members. Similarly, in a laboratory study of teams performing a creative task in a dynamic environment, those teams that had an outcome focus exhibited a greater ability to identify problems and to adapt their work processes to contextual changes than did process-focused teams.4 Indeed, as team members try out alternative ways of proceeding with the work, they are likely to enrich their understanding of what they are supposed to achieve. And, in the process, they may even come up with some clarifications, elaborations, or revisions that they would want to explore with the leader who created their team.
FIGURE 5-2 Specifying Means vs. Ends Adapted from Hackman (2002, Chap. 3)
Now look at the other three cells in Figure 5-2. Specifying both ends and means (the lower right cell) reduces the challenge to team members. It radically shrinks a team’s time horizon because members are focusing on immediate process requirements rather than long-term objectives. And, eventually, that can result in procedural compliance becoming an end in itself, displacing the team’s real purpose. Perhaps the greatest liability of requiring members to follow a given procedure, however, is that members’ collective knowledge, skill, and experience are underutilized. This was vividly illustrated in the U.S. military campaign in Afghanistan following 9/11. As William Arkin reported in the Los Angeles Times, Defense Secretary Donald Rumsfeld chose to personally decide about bombing attacks on all targets he deemed potentially sensitive. That proved unworkable, so authority was passed on to General Tommy Franks, who commanded U.S. operations in Afghanistan from his headquarters in Florida. That also did not work (both the Secretary and the General had larger issues to occupy them), so authority was further delegated, this time to a staff operations officer at headquarters, and that is where it stayed.5 Leaders of military units in the field, those who had local knowledge and who were responsible for real-time management of their forces, were reduced to executing operational decisions made in Florida.
Both of the remaining cells in Figure 5-2 are also highly problematic. Specifying neither ends nor means (the upper left cell) is the inkblot: teams risk fragmenting and falling into disarray as members struggle to figure out what they are supposed to accomplish. Rarely do teams in this cell exhibit competent work processes in pursuit of a common purpose. Teams of consultants and contractors who work with intelligence community clients sometimes wind up in this cell. When they have not been read into a relevant compartment, even consultants with high-level clearances may not be able to know the use that will be made of the work they are doing.
Finally, specifying means but not ends (the lower left cell) is the worst of all possible cases, the last resort of leaders who have no idea what their teams should be doing but who feel compelled nonetheless to exercise personal control over team behavior. So they write another memo or issue another directive and wind up disappointed that their team once again has turned in a subpar performance.
My observations of intelligence community teams have identified two disconnects between rhetoric and reality in the location of teams in the Figure 5-2 matrix. I have discussed that figure with many managers in the community and no one has yet argued that the upper right cell is a bad place for intelligence teams to be. Yet I also have noted two fairly strong tendencies to migrate away from that cell when teams actually are formed.
One tendency is to move to the left and specify neither a specific team purpose nor the procedures that are to be used to carry them out. This is the “the team will work things out” philosophy of leadership, which in reality is a philosophy of leadership abdication. It is especially common for teams of leaders or high-level professionals who are assumed, wrongly, not to need an explicit delineation of team purposes. “We’re seasoned professionals,” they say, “so we can figure it out. No need to treat us like novices.” And the person who is forming the team, most likely an even more senior leader, makes the mistake of going along.
It is true that there are some special leadership challenges in establishing a compelling purpose for a team of leaders or senior professionals. The issue is not their rank or seasoning—it is that such teams have more legitimate authority to manage their own affairs than do most other intelligence teams. Specifically, they typically are responsible not only for executing their work but also for shaping their own team purposes.6 But team members often do not recognize that. Instead, they assume, without explicitly checking with one another, that they have a common understanding of what they are supposed to be doing. When that assumption is incorrect, as it often is, dissension and wheelspinning develop.
One way to lessen the likelihood of purpose-related problems in managerial and professional teams is to establish, as a team’s first and most important task, the development of an agreed-upon statement of the team’s main purposes. Just that simple assignment can greatly diminish the likelihood that a professional team will wallow about, uncertain what it is supposed to accomplish, rather than move smartly ahead with the work. Senior teams need compelling purposes every bit as much as front-line teams. It’s just that they also may need a little help recognizing that and a little encouragement to take the time to agree about what they exist to accomplish.7
A second frequently observed tendency is to drift downward and wind up specifying both end states and the procedures a team is to use to achieve them. The use of structured techniques by analytic teams is a case in point. For teams that have a compelling purpose and the authority to make their own decisions about work processes (that is, teams in the upper right cell of Figure 5-2), the availability of a variety of analytic procedures is an extraordinarily valuable resource.8 But if a team is told which technique to use when, it no longer can tailor its performance strategy to the unique features of its analytic task—and the team takes a motivational hit in the bargain.
Ironically, the tendency to move “down” in the matrix is most commonly seen when team purposes are particularly urgent or consequential, such as dealing with an immediate crisis. Conventional wisdom is that self-managing teams may be fine during normal operations, but competent crisis management requires centralized operational control. The image, which you certainly have seen in motion pictures if not at work, is compelling: The officer in charge sits at the head of the table in the crisis center receiving reports from unit managers around the table and issuing orders about what each unit is to do next.
In fact, competent crisis management requires that teams in the field have not just clear purposes but also the latitude to deal in real time with rapidly changing local circumstances—the upper right cell of Figure 5-2. Senior leaders therefore must make sure that team purposes are clear and well understood (challenge and consequentiality rarely are a problem during crises), and then provide them with the support and resources they need to competently manage what is developing on the ground. Management of a crisis by remote control is just as problematic as was the attempt to manage the air war in Afghanistan from a headquarters office halfway around the world. Marine General James Mattis put it this way: “What are we creating today with our command-and-control systems? I don’t think we have turned off our radios in the last eight years. What kind of systems are we creating where we depend on this connection to headquarters?”9
Purpose comes first. But to obtain the considerable benefit of a compelling purpose, it must be well linked to a team’s actual work—what is being done and who is doing it. Otherwise, purpose becomes a mere abstraction, perhaps admirable but of little practical importance.
Whereas a team’s purpose is mainly about end states, its tasks are about the work itself—the materials the team deals with and what it does with them. If a team task is well designed, it can smooth the way toward achieving overall purposes. If it is not, the team may find itself occupied with activities that have little to do with what the team is supposed to accomplish.
In the era of “scientific management,” which dominated thinking about industrial work design in the middle of the last century, tasks were divided up into small pieces and made as simple as possible. That minimized the amount of employee training needed and allowed people to be swapped among work stations as readily as standardized parts were swapped in and out of production machinery. As elegant as this strategy for designing and staffing work was on paper, it turned out to generate numerous unanticipated and unwanted consequences—such as strained relationships between workers and management, quality problems, pervasive employee alienation, and unanticipated overhead costs in monitoring and coordinating workers’ activities. Eventually, scientific management was challenged by an alternative philosophy in which organization members were given semi-autonomous responsibility for completing whole, meaningful pieces of work, with customer or client feedback coming directly to the producer rather than to a manager or a quality control unit. Both the simplified and the enriched designs for work continue to be seen in contemporary organizations.10
To illustrate the difference between the two strategies for designing work, consider a simple production task such as manufacturing a toaster. In the scientific management approach, each individual would be assigned only one small piece of the work, such as attaching the line cord to the chassis, and would do that task repeatedly. Someone else would assemble the components that various workers produced, someone else would inspect the assembled unit, someone else would pack the toasters, and so on. In the “enriched” approach, one person (or, perhaps, a work team) would have full responsibility for producing the whole toaster as well as for inspecting and shipping completed products. The same differences are seen in intelligence work. Compare, for example, the task of merely documenting ship movements (mentioned earlier) versus the larger and more meaningful task of preparing a complete analysis of a potential adversary’s preparations for a possible military intervention. The latter task brings gains in motivation, in the utilization and development of workers’ capabilities, and in the quality of the work produced. It illustrates why, for intelligence work, it can be worth the trouble to break through a couple of compartments or bureaucratic boundaries to create large, meaningful tasks for which team members share full responsibility and accountability.
A key indicator of well-designed work is that it engenders what my colleague Greg Oldham and I call internal work motivation. Someone who is internally motivated feels great when he or she has done well, and feels bad when things have gone poorly. It is those internally generated feelings that fuel motivation, not extrinsic rewards or prods from a supervisor.11 Although most of the early research on task design focused on jobs performed by individual workers, the principles of good work design apply just as well to team tasks. Here, for example, are four items about collective internal motivation, taken from the Team Diagnostic Survey, an online instrument we developed to assess the design and dynamics of work teams.12 As you read these items, think about the extent to which they accurately describe a team on which you serve.
I feel a real sense of personal satisfaction when our team does well.
I feel bad and unhappy when our team has performed poorly.
When our team has done well, I have done well.
My own feelings are not affected one way or the other by how well our team performs. (reverse scored)
The better a team’s work is designed (that is, the extent to which members have collective responsibility for carrying out a whole, meaningful piece of work for which they receive direct feedback), the higher a team’s collective internal motivation. And one of the most powerful ways to strengthen the design of a team’s work is to establish direct relationships between the team and the customers for whom the work is being done. Note, however, that in intelligence work it can be challenging to maintain simultaneously both close customer relationships and the highest standards of professional integrity.13
Those analytic teams in the Hackman-O’Connor study (described earlier) whose work was well designed scored higher than did coacting groups in the same agencies; similar findings were obtained in an unpublished analysis of red teams (which were well designed) relative to blue teams (which sometimes were not) in PLG simulations. It is well worth the effort, therefore, to give careful thought to how intelligence tasks are designed—confirming not just that a team’s actual work is congruent with its overall purposes but also that the way it is structured gives the team every chance to develop collective internal motivation to perform well.
Internal motivation can become self-sustaining—the better I do, the more I learn and the more I seek even greater challenges, creating a positive spiral. Not everyone responds positively to well-designed work, however. For one thing, if a person is not sufficiently knowledgeable or skilled to competently carry out the work, then a downward motivational spiral can develop. Look back at the internal motivation items: I feel good when I do well, and I feel bad when I do poorly. So if I do not have what it takes to do well, I will feel bad much more often than I feel good—a recipe for disillusionment and withdrawal. Moreover, people who are high on what Oldham and I call “growth need strength” respond much more positively to opportunities to perform challenging work than do people whose needs have less to do with professional growth than with, say, personal security or harmonious collegial relationships. People in the latter categories simply do not get the “kick” from achieving a challenging goal that their more growth-oriented colleagues do.
The good news for the intelligence community is that the workforce consists mainly of highly competent professionals who do, in fact, seek opportunities for continued learning and growth—at least when they first arrive. The challenge for community managers is to make sure that organizational policies and practices, very much including how tasks are designed and managed, do not snuff out that spark. According to one senior official with whom I spoke, “We have to get the way we do business changed now, before all our new people become cynical and disillusioned and check out psychologically.” The official’s worry is well founded. People really do check out when they spend months or even years working under close supervision on simple or trivial tasks that seem to make no difference whatsoever for anything of real consequence. Creating teams that perform large, well-designed tasks can go a long way toward minimizing the chances of that happening.
Designing work for teams can seem like an unnatural act. When managers think about the best way to set up a piece of work, they almost always focus more on how to divide it up (that is, how to partition the overall task into separate pieces to be performed by separate individuals) than on ways to elicit and integrate the contributions of a diverse set of performers.14 That impulse is reinforced by tradition; by the presumed efficiency of simple, tightly defined jobs; and by human resource practices that favor well-specified individual tasks for which employees can be readily selected, trained, assessed, and rewarded. Moreover, the need for secrecy in some intelligence units can make it nearly impossible to structure team tasks that cross departmental lines or compartment boundaries.
For all of these reasons, a great deal of work in the intelligence community is designed in general accord with the precepts of scientific management. Yet when a task is important enough, intelligence community managers sometimes do abandon traditional approaches to work design—for example, by creating counterterrorism teams whose analytic and operational members cross organizational lines to work closely together, or by cutting through bureaucratic policies and practices to encourage collaboration among diverse national security professionals, such as in the search for Saddam Hussein. These teams were designed to take on highly consequential tasks in extraordinary times. But they suggest that well-designed work teams may be a good device for fostering collaboration and learning in normal times as well. Indeed, it may be that teamwork is generally preferable to traditional work systems whenever the work must be carried out in demanding or rapidly changing operational contexts.
Let me close with an account I wrote a few years ago in another context.15 It is about how David Mathiasen, who at the time was head of the Fiscal Analysis Branch of the Office of Management and the Budget (OMB), created a compelling purpose for his team in quite difficult circumstances. You will see striking parallels with the challenges commonly encountered in crafting the purposes of teams in the intelligence community.
David Mathiasen’s branch conducted economic analyses of the federal budget for the president’s budget director. Ronald Reagan had just defeated Jimmy Carter for the presidency and had appointed David Stockman as budget director. Soon thereafter, Stockman told Mathiasen that the agency would proceed immediately to dismantle the Carter budget and replace it with one that emphasized the priorities of the new president.
Mathiasen wondered how he could possibly engender sufficient commitment among members of his team to do what Stockman had asked. The team had worked terribly hard on the Carter budget, and now it was to be discarded. How could he get team members fired up to restart a task they had just completed—especially since their personal politics ranged from strong liberalism to committed conservatism? How much conflict would develop among them as they worked on a conservative federal budget?
Mathiasen’s solution relied mainly on how he framed the purpose of his team. He was not one to call a big meeting and make a charismatic speech. Instead, he went around from person to person on no special schedule, making sure that everybody understood what the mission of the fiscal analysis team really was. The essence of what he said on his rounds was this:
As corny as it may sound, what we are here for is to serve democracy. We don’t make policy, but we make sure that the people who do have absolutely the best information that they can have. Some of you applaud the priorities being set by Reagan and Stockman; others of you are certain that their proposals will lead the country into social and economic disaster.
As a citizen, I too have some opinions about what they are doing. But my personal views don’t matter in our work here and neither should yours. We are the only people on this planet who are in a position to provide the President and his Director with comprehensive and valid analyses of the likely effects of their policies. The PADs [politically appointed associate directors of the OMB] can’t do it—they don’t have the time or the expertise and, besides, they have to keep passing political litmus tests. The Director cannot do it himself, although this particular director, if we don’t do our job right, just might give it a try. And the Congressional Budget Office works for that other branch of government, they have different job to do.
So there’s nobody else, it falls on us. Those of you who love what Reagan is doing can take pleasure knowing that your analyses will give him the information he needs to implement his policies promptly and decisively. And those of you who detest what he is up to can take pleasure from the fact that, with complete and accurate data, he’ll probably do less damage than he would otherwise.
No matter what your personal politics, it all comes down to the same thing: Our democracy will work better if the President and his advisors have complete and trustworthy data. Frankly, I don’t know whether we can get it all done in the time that we have. It will be quite a stretch. But we’re all professionals, so let’s pitch in and show them what we can do.
Even staffers who unhesitatingly had pulled the “Carter” lever in the voting booth found themselves coming in evenings and weekends, when needed, to work with their teams to do their part in rebuilding the national budget. That is the power of a compelling and wellcommunicated team purpose.