Team of Teams: New Rules of Engagement for a Complex World

by Stanley McChrystal

We had worked out our solutions to the Task Force’s Prisoner’s Dilemma by trial and error, but we later learned that game theory scholars shared our conclusions.

Daniel Kahneman, the Nobel Prize–winning cognitive scientist, believes that the human mind has two different decision-making tracks: “system 1” operates automatically and quickly, while “system 2” is deliberate and effortful. We tend to use system 1 frequently and reflexively—for instance, gauging the emotions on someone’s face—and apply system 2 when weighing difficult decisions or attempting complex calculations. What we saw in the Task Force was that while cooperation began as a conscious system 2 decision (“they’ll help me later if I help them now; cooperation is in my interest”), a track record of productive collaborations led to reflexive, system 1 cooperation—in other words, real trust. Furthermore, this trust had a viral effect: once it passed a certain threshold, it became the norm.

Together, these two cornerstones—systemic understanding and strong lateral connectivity—grounded shared consciousness. Both diverged wildly from the MECE, reductionist doctrines we had spent most of our lives upholding, but, in this new setting, against this new threat, they worked. The two strains of shared consciousness also paralleled the ingredients that, at a lower level, had ensured the success of our small teams for decades: “seeing the system” is essentially a macro version of the “purpose” that gives our operators the context and commitment to persevere in volatile situations, and the interteam bonds we used to beat the Prisoner’s Dilemma are akin to the trust between team members. As we discussed in chapter 6, this—the realization of team traits at scale; the transformation into a team of teams—was exactly what we needed.

Mary Barra,

Forty-one-year-old Alfred P. Sloan, already an experienced executive, joined General Motors in 1918. GM founder William C. Durant had snapped up the United Motors Company, Sloan’s previous employer, in a spree of corporate acquisitions. It was an exciting time for American business, especially the nascent auto industry. Eight-year-old GM had already established itself as a market leader and was growing fast. But Durant’s binge mergers created problems. Though a visionary, he was unable to bind this sprawl of companies together in an orderly fashion.

Internal rivalries—the consequence of separate divisions and a competitive culture—inhibited communication. Each division maintained its own design and marketing operations and mistrusted other teams. This was one of the problems the silo system had been put in place to solve, and Sloan’s solution had worked well to a point; but like other command-and-control structures, it failed at the threshold of complexity.

Once reports of accidents began coming in, various divisions held meetings, but no meaningful action was taken. An internal report later concluded that “the engineers . . . did not know how their own vehicle had been designed. And GM did not have a process in place to make sure someone looking at the issue had a complete understanding of what the failure of the Ignition Switch meant for the customer.”

This was inextricably linked with the general culture of efficiency and internal competitiveness. Perhaps some employees would have tried harder to relay these issues up the chain of command, or perhaps senior leaders would have investigated the mysterious crashes more thoroughly, had the slogan “cost is everything” not dominated decision making. Like the “Faster, Better, Cheaper” approach that encouraged poor decision making at NASA in the 1970s, this drive to cut out fat inhibited systemic understanding. An engineer interviewed said that the emphasis on cost cutting “permeates the fabric of the whole culture,” leading to a privileging of timing over quality, and a resistance to raising issues. No team wanted to be the group that lagged in efficiency or took too long to fix a problem on account of being overly cautious. An avoidance of responsibility had become known as the “GM nod”—a staple of survival and job security at the company.

In 2005, Bill Ford saw the writing on the wall: “We can continue to cut costs and improve our efficiency, but we cannot win the hearts and minds of a new generation with efficiency alone.” The board brought in Alan Mulally as CEO. Mulally had been in charge of Boeing’s commercial airplanes division and had overseen the development of the 777, one of the safest, most advanced, most financially successful passenger planes ever created. He attributed the project’s success to a management approach called “working together” that involved forcing interaction between previously separate groups and cutting-edge technological platforms for ensuring constant, systemic transparency.

As Mulally put it, “Working together always works. It always works. Everybody has to be on the team. They have to be interdependent with one another.”

Mulally’s belief in the universal utility of rejecting silos and embracing interdependence is backed up by Sandy Pentland, an MIT professor who studies the effects of information flow on organizations and communities. Looking at very large data sets, Pentland has found that sharing information and creating strong horizontal relationships improves the effectiveness of everything from businesses to governments to cities. His research suggests that the collective intelligence of groups and communities has little to do with the intelligence of their individual members, and much more to do with the connections between them.* “The best ideas,” he writes, “come from careful and continuous social exploration . . . it is the idea flow within a community that builds the intelligence that makes it successful.”

It is necessary, we found, to forcibly dismantle the old system and replace it with an entirely new managerial architecture. Our new architecture was shared consciousness, and it consisted of two elements. The first was extreme, participatory transparency—the “systems management” of NASA that we mimicked with our O&I forums and our open physical space. This allowed all participants to have a holistic awareness equivalent to the contextual awareness of purpose we already knew at a team level. The second was the creation of strong internal connectivity across teams—something we achieved with our embedding and liaison programs. This mirrored the trust that enabled our small teams to function.

Alfred Sloan described his system as “decentralized operations with coordinated control.” We found that we benefited from the opposite. First, we needed coordinated operations, something that necessitated emergent, adaptive intelligence. Shared consciousness achieved this, but it was only the first half. As we would soon find, keeping pace with the speed of our environment and enemy would require something else as well: decentralized control.

RECAP Cooperation across silos would be necessary for success, and while systemic understanding was a valuable first step, we needed to build more trust if we were to achieve the fluid, teamlike cooperation that we needed across our force; we had to overcome the challenge of the Prisoner’s Dilemma. To this end, we used embedding and liaison programs to create strong lateral ties between our units, and with our partner organizations. Where systemic understanding mirrors the sense of “purpose” that bonds small teams, this mirrored the second ingredient to team formation: “trust.” Together, these two elements completed the establishment of shared consciousness, something that was vital to our success. As is evidenced by the failures of GM and successes of Ford, the same innovations are sorely needed by many organizations still using rigid silos in an interdependent world.

Being woken to make life-or-death decisions confirmed my role as a leader, and made me feel important and needed—something most managers yearn for. But it was not long before I began to question my value to the process. Unless I had been tracking the target the previous night, I would usually know only what the officers told me that morning. I could ask thoughtful questions, but I had no illusions that my judgment was markedly superior to that of the people with whom I worked. As much as I would like to think otherwise, I only rarely had some groundbreaking insight. Most of the time I would simply trust the recommendations made by those who came to get me, as they knew the most about the issue. My inclusion was a rubber stamp that slowed the process, and sometimes caused us to miss fleeting opportunities.

Shared consciousness helped us understand and react to the interdependence of the battlefield where we faced off against AQI. But interdependence was only half of the equation—the other half was speed, and that was still an issue. We had become vastly more thoughtful, integrated, and insightful, but the Task Force still was not fast enough.

Within the Task Force, thanks to radical information sharing, we had come a long way with regard to Drucker’s exhortation to “do the right thing” rather than “do things right”: people at every level of the organization had the information and connectivity to determine what the right thing was, in real time. But, held back by our internal processes, they lacked the ability to act on that determination. We had gotten halfway to transcending Krasnovian soccer and then stopped: we had built an outstanding team bound together by the oneness of trust and purpose and capable of devising, in real time, brilliant, emergent solutions to complex problems, but we still required every player to get written permission from the coach before passing the ball. Meanwhile, senior leaders from Iraq to Washington had in-boxes overflowing with requests to do things that they knew less about than the people on the ground and about which they were often unable to judge effectively.

Empowerment tends to be a tool of last resort.

I began to reconsider the nature of my role as a leader. The wait for my approval was not resulting in any better decisions, and our priority should be reaching the best possible decision that could be made in a time frame that allowed it to be relevant. I came to realize that, in normal cases, I did not add tremendous value, so I changed the process. I communicated across the command my thought process on decisions like airstrikes, and told them to make the call. Whoever made the decision, I was always ultimately responsible, and more often than not those below me reached the same conclusion I would have, but this way our team would be empowered to do what was needed.

The practice of relaying decisions up and down the chain of command is premised on the assumption that the organization has the time to do so, or, more accurately, that the cost of the delay is less than the cost of the errors produced by removing a supervisor. In 2004 this assumption no longer held.

We concluded that we would be better served by accepting the 70 percent solution today, rather than satisfying protocol and getting the 90 percent solution tomorrow (in the military you learn that you will never have time for the 100 percent solution).

The Ritz-Carlton hotel chain has spent a century building a reputation for quality, luxury, and reliability. Through recessions, depressions, corporate mergers, and world wars, the brand—originally a restaurant operated on high-end cruise ships, then a hotel that earned founder César Ritz the sobriquet “king of hoteliers and hotelier to kings”—has remained at the top of the food chain. Today, the company operates eighty-five hotels in thirty countries and regularly tops the Zagat lists for its hotels and dining. In particular, it has earned a reputation for offering outstanding service. It is César Ritz who is credited with the line, now a universal law in the hospitality industry, that “the customer is always right.” Nearly fifty thousand executives from other companies have traveled to the Ritz-Carlton Learning Institute and Ritz-Carlton Leadership Center to learn how they too can achieve such quality of service.

Since the 1980s, when companies began experimenting with “empowerment”—the buzzword that summarizes what we called “decentralization of decision-making authority”—myriad studies in the social sciences have concluded that this psychological difference of empowerment has a very real impact. Jay Conger and Rabindra Kanungo’s 1988 paper “The Empower Process: Integrating Theory and Practice” noted that empowerment improved employee satisfaction. Kenneth W. Thomas and Betty A. Velthouse identified the decentralization of authority as creating “intrinsic task motivation.” Studies have found this effect in domains ranging from nursing in China to five-star hotels in Turkey.

This is just what we wanted in the Task Force: we accepted that divergences from plan were inevitable—we wanted to improve our ability to respond to them. We needed to empower our teams to take action on their own.

Eventually a rule of thumb emerged: “If something supports our effort, as long as it is not immoral or illegal,” you could do it. Soon, I found that the question I most often asked my force was “What do you need?” We decentralized until it made us uncomfortable, and it was right there—on the brink of instability—that we found our sweet spot.

This took us by surprise and upended a lot of conventional assumptions about the superior wisdom of those at the top. Understanding the underlying causes of this unexpected outcome proved essential to sustaining and enhancing it.

Adam Nicolson’s Seize the Fire about Admiral Nelson.

The term “empowerment” gets thrown around a great deal in the management world, but the truth is that simply taking off constraints is a dangerous move. It should be done only if the recipients of newfound authority have the necessary sense of perspective to act on it wisely.

THE VISIBLE MAN MODEL OF DECISION MAKING He was just fifteen inches tall and made of plastic, but he left an enduring impression on me. The brainchild of Marcel Jovine, a former Italian soldier who came to America as a POW in World War II, the Visible Man was a plastic anatomical toy introduced by the Renwal Products Company in the fall of 1958. He cost $4.98 and his clear plastic body held a skeleton and organs that could be removed and replaced.

RECAP Traditionally, organizations have implemented as much control over subordinates as technology physically allowed. New technologies offer today’s leaders unprecedented opportunities to gather information and direct operations, but because of the speed necessary to remain competitive, centralization of power now comes at great cost. While shared consciousness had helped us overcome the interdependence of the environment, speed, the second ingredient of complexity, still posed a challenge. Effective adaptation to emerging threats and opportunities requires the disciplined practice of empowered execution. Individuals and teams closest to the problem, armed with unprecedented levels of insights from across the network, offer the best ability to decide and act decisively.

armed. It is the apogee of heroic leadership—omniscient, fearless, virile, and reassuring. It is also almost entirely unrealistic. While some leaders possess extraordinary gifts and project a charismatic presence, in a career alongside accomplished leaders, I never met a Marko Ramius—or anyone remotely close to the character author Tom Clancy created in The Hunt for Red October.

The doctrine of empowered execution may at first glance seem to suggest that leaders are no longer needed. That is certainly the connection made by many who have described networks such as AQI as “leaderless.” But this is wrong. Without Zarqawi, AQI would have been an entirely different organization. In fact, due to the leverage leaders can harness through technology and managerial practices like shared consciousness and empowered execution, senior leaders are now more important than ever, but the role is very different from that of the traditional heroic decision maker.

The role of the senior leader was no longer that of controlling puppet master, but rather that of an empathetic crafter of culture.

But of course we can’t. Author Dan Levitin explains: In 2011 Americans took in five times as much information every day as they did in 1986—the equivalent of 175 newspapers. During our leisure time, not counting work, each of us processes 34 gigabytes or 100,000 words every day. The world’s 21,274 television stations produce 85,000 hours of original programming every day, as we watch an average of 5 hours of television each day, the equivalent of 20 gigabytes of audio-video images.

One solution to information overload is to increase a leader’s access to information, fitting him with two smartphones, multiple computer screens, and weekend updates. But the leader’s access to information is not the problem. We can work harder, but how much can we actually take in? Attention studies have shown that most people can thoughtfully consider only one thing at a time, and that multitasking dramatically degrades our ability to accomplish tasks requiring cognitive concentration. Given these limitations, the idea that a “heroic leader” enabled with an über-network of connectivity can simultaneously control a thousand marionettes on as many stages is unrealistic.

Years later as Task Force commander, I began to view effective leadership in the new environment as more akin to gardening than chess. The move-by-move control that seemed natural to military operations proved less effective than nurturing the organization—its structure, processes, and culture—to enable the subordinate components to function with “smart autonomy.” It wasn’t total autonomy, because the efforts of every part of the team were tightly linked to a common concept for the fight, but it allowed those forces to be enabled with a constant flow of “shared consciousness” from across the force, and it freed them to execute actions in pursuit of the overall strategy as best they saw fit.

The gardener cannot actually “grow” tomatoes, squash, or beans—she can only foster an environment in which the plants do so.

First I needed to shift my focus from moving pieces on the board to shaping the ecosystem. Paradoxically, at exactly the time when I had the capability to make more decisions, my intuition told me I had to make fewer. At first it felt awkward to delegate decisions to subordinates that were technically possible for me to make. If I could make a decision, shouldn’t I? Wasn’t that my job? It could look and feel like I was shirking my responsibilities, a damning indictment for any leader. My role had changed, but leadership was still critical—perhaps more than ever.

Communicating priorities and cultural expectations to our team of teams spread across multiple continents was a challenge. Written guidance was essential, but memos competed with the flood of text that engulfed all of us every day. To post brief updates and observations, I used a secure Web-based portal accessible to everyone, carefully composing each memo to ensure that it reflected not only my thoughts, but also my “voice.” I tried to remember “less is more,” and stuck to a few key themes. Experience had taught me that nothing was heard until it had been said several times. Only when I heard my own words echoed or paraphrased back to me by subordinates as essential “truths” did I know they had been fully received.

As a leader, however, my most powerful instrument of communication was my own behavior.

There’s an art to asking questions. Briefings are valuable but normally communicate primarily what the subordinate leader wants you to know, and often the picture they provide is incomplete. Thoughtful questions can help fill in the blanks.

Once they recalculated, their answers were impressive. Most adjusted their approach to take a longer view of solving the problem. You might expect them to seek a quicker solution and an earlier ticket home. But they were experienced enough to know that real solutions demand the long view—simple fixes are illusory. Although I couldn’t change the troop rotation policy, as I left, I’d ask each soldier to execute his or her duties with that mind-set.

Simple honesty shows, and earns, respect.

I would tell my staff about the “dinosaur’s tail”: As a leader grows more senior, his bulk and tail become huge, but like the brontosaurus, his brain remains modestly small. When plans are changed and the huge beast turns, its tail often thoughtlessly knocks over people and things. That the destruction was unintentional doesn’t make it any better.

RECAP Although we intuitively know the world has changed, most leaders reflect a model and leader development process that are sorely out of date. We often demand unrealistic levels of knowledge in leaders and force them into ineffective attempts to micromanage. The temptation to lead as a chess master, controlling each move of the organization, must give way to an approach as a gardener, enabling rather than directing. A gardening approach to leadership is anything but passive. The leader acts as an “Eyes-On, Hands-Off” enabler who creates and maintains an ecosystem in which the organization operates.

The processes, relationships, and trust that underpinned that complex effort were by now things we almost took for granted. We used them every day and night as though they were the natural order of things. Inside, however, most of us knew just how much it had taken to bring that “natural order” into being.

that?” The answer was not some secret treasure trove of AQI data we stumbled across or a technological breakthrough in surveillance; it was the very edge that AQI had once held over us: a revolution in the mundane art of management.

An organization as regenerative and fluid as AQI would never possess a single point of failure, which was why it was important that we hit them relentlessly and accurately.

That's supposedly one reason why the internet was developed. No center, so you can't knock it out.

Tocqueville also wrote extensively on what he saw as America’s vulnerabilities. Although people then and now tend to consider the essential tenet of democracy to be the political empowerment of the people, this alone does not produce a successful democracy—the people can be effectively empowered only if they have enough context to make good decisions. Tocqueville emphasized this point, noting that “in the United States the instruction of the people powerfully contributes to the support of the democratic republic.”

An organization should empower its people, but only after it has done the heavy lifting of creating shared consciousness. This is much harder when you are trying to achieve something constructive: