It Worked For Me

by Colin Powell

  At levels above a small rifle company, there is a more sophisticated (tongue in cheek) way to handle these transitions. It’s called the “Three Envelopes Construct.” The outgoing leader gives the new leader three envelopes—labeled “Envelope 1,” “Envelope 2,” and “Envelope 3”—and tells him to open them in order if he runs into trouble. The new leader launches in a blaze of glory. But after a month or so, troubles start landing on him. He opens the first envelope, and the note inside says: “Blame me.” So he goes around complaining about the mess he inherited. Things settle down, but a couple of months later he is back in trouble. He opens the second envelope: “Reorganize.” He immediately starts a major study to determine the kind of reorganization that would improve the situation. For months, the reorganization study moves all the boxes and people around and creates a new paradigm. Everyone is distracted. The new paradigm looks exciting, but nothing is solved and everyone is confused.

  The now no longer new commander is in dire straits and beside himself with worry. In desperation he opens the third envelope. The note says: “Prepare three envelopes.”

  The Three Envelopes Construct does not work with elected politicians. They will blame their predecessors as long as they can. If things are going wrong, it is not their fault. If things are going well, it is only through their superb efforts to fix the mess they inherited. If their predecessor comes from their own party they may have to complain sotto voce.

  For normal mammalian human beings in a line position, assume your predecessor did a good job, and if he didn’t, be silent. Move onward and upward. You are in charge. Take charge. And always remember, “You now own the sheets.”

  CHAPTER TWENTY-SEVEN

  Mirror, Mirror, on the Wall

  I am pretty good at knowing and analyzing my strengths and weaknesses; but I keep the latter private. Though I never share these with anyone, my family and friends are quite willing to tell me what they are in detail. Self-examination is tough and worse when your friends and family join in. I am so glad that 360-degree evaluations came into vogue long after I stopped being evaluated. During the process, your ego is vulnerable, your self-respect challenged, your decisions questioned, and your fallibility made manifest. Still, such examination is essential to improving yourself, getting in better touch with the people in your life, facing your demons, and moving on. Looking deeply into a mirror and seeing an accurate reflection is therapeutic and healthy.

  If it is difficult for individuals, it is even more difficult for groups of individuals in an organization with superiors and subordinates, where candor can put members of the group at risk, or where your honesty may be seen as disloyal or get you condemned as failing to be a team member. An organization that is unable to create the environment for this kind of evaluation is an organization that is holding itself back. The challenge is to get beyond the personalities, the egos, and the tendency to be blind to unpleasant conditions and move forward with no feelings bruised beyond repair. This is a real test of leadership and confidence in the team and the bonds that hold a team together.

  Honest, brutal self-examination is especially difficult, but even more vital after a mess, a screw-up, or a failing performance. The Army faced such a crisis after the Vietnam War. There were no victory parades, and the nation got rid of the draft and distanced itself from the nascent all-volunteer force it had launched. We were in the midst of the countercultural revolution, racial and drug problems, and a shaken political system that would see the resignations in disgrace of a president and vice president. We had to reform an institution with deep cultural roots and a proud history—an institution that had recently failed to achieve its ultimate reason for existence, success in war. We set about rewriting our doctrine, reorganizing our units, and training all-volunteer recruits, many of whom were deficient in education or had behavioral issues. For me, it was the most demanding, exciting, and rewarding time in my career. We succeeded and rebuilt a first-class Army, as good as any that went before.

  One of the most powerful tools the Army used to achieve this success was a technique called the After-Action Review (AAR). The AAR concept was first tested and proved at the newly established National Training Center (NTC), at Fort Irwin, California, arguably the most innovative training facility ever created. Both the NTC and AARs are still going strong.

  The NTC consists of 600,000 acres of rolling desert, ideally suited for mechanized maneuver training and live firing in an utterly realistic environment. Units coming to the NTC to train face off against a highly skilled and trained enemy—called an Opposing Force (OpFor)—that is stationed at Fort Irwin. Both the good guys and the bad guys are wired, so their actions can be followed on computers at a centralized control center.

  Training against a simulated enemy is not new. Armies have been trying to make training realistically close to actual combat for a long, long time. What makes the NTC unique is the comprehensive AAR that follows the completion of every battle. At AARs leaders, observers, and evaluators sit in the control center and watch the battle replayed like a video game. Every vehicle moving across the battlefield can be identified; every movement of troops and vehicles, every action, every gunshot has been recorded and can be replayed in several ways. For instance, the actual battle can be superimposed over the commanders’ plans, comparing and contrasting reality with expectations. I have watched many an aspiring Patton put his original plan up on the screen and then watch his tanks and armored vehicles go wandering off in the wrong direction, firing at each other, as the OpFor rolls up his flank and defeats him. It reminds the young Patton of two military maxims: “No plan survives first contact with an enemy” and “Even the most brilliant of strategists must occasionally take into account the presence of an enemy.”

  All of this is then exhaustively analyzed. Nothing is held back, nothing is ignored. During the review, leaders, observers, and evaluators come together to present their own assessments of how they saw the battle unfold and why they made their decisions and took their actions.

  The purpose of the review is to autopsy the exercise, not to give a grade or to anoint the commander as a future Patton . . . or Custer. Learning and improvement are the sole focus, not the unit’s success or failure in the mission. It’s not a blame game.

  After the review, the subordinate leaders are then expected to go back to their units and share the AAR results down to the last soldier. Each subordinate unit conducts its own AAR.

  The AAR system works because it is a training process, not an evaluation process. That doesn’t mean feelings won’t be hurt or unfavorable impressions created. The needs of the mission must come first. Though AARs are not about assigning blame, poor performance over time will naturally be noticed. A commander who consistently does poorly, or worse, is probably not suited for the job, or for command at a higher level. Those who consistently do well get noticed.

  Because it works so well, the AAR system has been extended to all training throughout the Army. Watching AARs, I witnessed the birth of a new Army focused less on proving your worth by scoring points than on training our soldiers to be more effective. In my early days in the Army, evaluations were generally a matter of mechanically working through stylized teach-to-the-test checklists. Today, the system asks, “Where do we need more training? How do we make our troops better and more skilled?”

  The result of the new training system was demonstrated in Operation Desert Storm in 1991 and subsequent conflicts. After engaging in actual combat, officers and soldiers who had been through the NTC experience reported that it had replicated down to nitty-gritty details the demands of real combat. It gave them a decisive edge when they faced the Iraqi army.

  NFL teams go through a similar process after each week’s games. They review their own game films, they review films of their next opponent, and they constantly ask themselves, “What can we do to fix our mistakes and improve?”

  The AAR process is applicable to any organization that truly wants to know how it is doing, where
it needs to improve, and how it can get to the bottom of a problem or dispute. What have we done right? What have we done wrong? The sole goal is to improve our performance. It’s not about your ego or mine. If we are a team, we can level with each other in a spirit of “how do we do better?” We will not cover up mistakes, reorganize around them, or stare at the sky. It requires honest participation, a focus on learning, and a commitment not to use AARs as a means to assign grades. High-performing organizations understand the need for this kind of evaluation. I have also seen others whose leadership doesn’t have the guts to look into the mirror. All of us have seen in recent years too many pitiful examples of companies and organizations that live and succeed in the moment and refuse to see the reality of the fuse burning in the basement. Leaders should never bury a problem; you can be sure it will eventually rise from its grave and walk the earth again.

  I have tried to apply the AAR philosophy in all my post-Army assignments. During my days as Secretary of State, I was responsible for submitting an annual report to Congress about trends in terrorist incidents. The report was prepared by the CIA, reviewed by my staff, and sent to Congress in my name.

  One year, Congressman Henry Waxman of California attacked the report. He accused me of understating the terrorist problem and of cooking the books by reporting fewer worldwide terrorist incidents than he believed the data showed. My staff initially circled the wagons and defended our position—the traditional bureaucratic response. But I wanted to find out who was right. If Congressman Waxman was right, we had to make changes, and do that before we had to defend our position before an open congressional committee. If we were right, I was ready to take on Henry, a good friend as long as we weren’t across the table from each other at a televised hearing.

  At my staff meeting the next morning we conducted an AAR. I wasn’t happy with what I heard. Rather than starting at the beginning and analyzing exactly how the original report was generated, the staff just tossed up justifications for the report we had printed and distributed.

  I told them to look at Congressman Waxman as though he were our OpFor at the NTC; his negative evaluation of our report was equivalent to an OpFor victory in an early engagement. I thought we should listen to his criticism, concede that he might be right, and fix the problems he’d spotlighted so they would not end up in lurid display before his congressional committee. That would have been equivalent to losing the final battle.

  At another AAR the next morning, we brought in everyone involved in preparing the report, and continued to peel back the onion. But we also brought in my entire staff, so everyone could learn how AARs worked and could chime in with off-the-wall questions.

  As we dug deeper and deeper, we discovered significant errors in the CIA’s categorization and counting of terrorist incidents. These were errors, nothing more. They were not evidence of criminal, corrupt, or otherwise evil practices. The CIA’s errors were then compounded by my staff, who had to admit they hadn’t done an adequate job analyzing the draft report. The discussions were all conducted civilly and deliberately; no crucifixions were ordered.

  By the third morning’s AAR, everyone who knew anything about the issue was pitching in to make sure we had a clear view of exactly what had gone wrong. The AAR approach cut through all the Gordian knots and got to the core problems in short order. My staff and the CIA, working side by side, soon went to work redoing the analysis.

  I called Waxman to tell him that he was right and I was wrong and to assure him that my team was hard at work fixing the problem and preparing an amended report. Because he trusted us, he gave us the time we needed. We submitted an accurate revised report within a few weeks. Congressman Waxman publicly congratulated us, and there was no further congressional intervention. More important, we fixed the report-making system to avoid future problems.

  The problems I found were organizational and needed correction, and they were dealt with quietly and in a timely manner outside the AAR process. The goal of an AAR is to get everyone around a table to review the battle, learn what went wrong, learn what went right, and work out how to train to do better. Leadership and personnel problems revealed by AARs normally get fixed privately.

  Every organization needs to be introspective, transparent, and honest with itself. This only works if everyone is unified on the goals and purpose of the organization and there is trust within the team. High-performing, successful organizations build cultures of introspection and trust and never lose sight of their purpose.

  CHAPTER TWENTY-EIGHT

  Squirrels

  One morning early in 1988, shortly after I became President Reagan’s National Security Advisor, I went into the Oval Office to discuss a problem with the President. We were alone. He was sitting in his usual chair in front of the fireplace with a view of the Rose Garden through the beautiful glass-paned French doors. I was sitting on the end of the couch to his left.

  I don’t even remember what the problem was. But it involved a not-uncommon fight between the State and Defense departments, made more complicated by significant Commerce, Treasury Department, and congressional interests. I described the problem at some length and complexity to the President, underscoring that it had to be solved that day.

  To my discomfort, he kept looking past me through the French doors without paying much attention to my tale of woe. So I talked a little louder and added more detail. Just as I was running out of gas, the President raised up and interrupted me: “Colin, Colin, the squirrels just came and picked up the nuts I put out there for them this morning.” He then settled back into his chair and turned back to me. I decided the meeting was over, excused myself, and went back to my office down the hall in the northwest corner of the West Wing.

  I had the feeling that something important had just happened. I sat down, gazed out my windows across the north lawn and into Lafayette Park across Pennsylvania Avenue, and reflected on it. And it became clear.

  The President was teaching me: “Colin, I love you and I will sit here as long as you want me to, listening to your problem. Let me know when you bring me a problem I have to solve.” I smiled at this new insight. In my remaining months with him, I told him about all the problems we were working on, but never asked him to solve problems that he had hired me and the rest of his team to solve. Reagan believed in delegating responsibility and authority and he trusted those who worked for him to do the right thing. He put enormous trust in his staff. The President’s approach worked for him most of the time. But it could also get him in trouble, as the Iran-Contra debacle demonstrated.

  On another morning in 1988, I went into the Oval Office with another problem. U.S. naval forces in the Persian Gulf were chasing Iranian gunboats that had threatened them. Our ships were approaching the Iranian twelve-mile limit and Secretary of Defense Frank Carlucci wanted authority to break that limit in hot pursuit of the boats.

  President Reagan was sitting behind his desk, calmly signing photos, knowing that we were in action. He trusted our ability to manage the situation and keep him informed. He looked up as I approached and locked his eyes on me. He knew he was about to be handed a Commander in Chief problem to solve. I laid out the request, with all the upsides and downsides, potential consequences, press needs, and congressional briefing strategy. He took it all in and simply said, “Approved, do it.” I conveyed the answer to Frank, we chased the boats back to their bases, and the action was over.

  On many occasions during our time together, I brought Reagan presidential decisions that he would think through, question, analyze, and make. He was always available for Oval Office decisions. But he was happier if problems could be solved at a lower level.

  One of my most treasured mementos, and the only signed picture I have from Reagan, shows us sitting side by side in front of the Oval Office fireplace. We are leaning toward each other examining charts I am using to explain some issue. He later inscribed that photo, “Dear Colin, If you say so, I know it must be right.” Gulp.

  I have
always loved making things work well. From rebuilding worn-out Volvos to reshaping senior executive staffs into fine-tuned instruments, one of my deepest passions has been taking something that is not functioning as well as it should up to its highest level of performance. President Reagan taught me how to better achieve that goal by creating and maintaining mutual trust and accountability with my senior staff. They’re as essential to a smooth running organization as an electrical system or driveshaft is to a Volvo.

  In all my senior positions after serving under Reagan, I worked hard to create a Reaganesque level of mutual trust and accountability—trusting that my senior officials would be prepared, do the right thing, know what I wanted done, and be ready to be accountable for their actions.

  Maintaining mutual trust and accountability meant keeping my people close to me, with very short and direct lines of communication and authority and the fewest possible bureaucratic layers between us.

  My military training rested on the concept of the chain of command where everyone knows who is in charge and where only one person at a time can be in charge. From this training came my belief in working with direct-reporting subordinates without lots of assistants or other intervening layers helping me to run the staff.

  My later experience in government has been that staff numbers multiply and fancy titles proliferate in an inexorable kudzu manner unless the bosses regularly and viciously prune them.

  When Frank Carlucci was named National Security Advisor at the end of 1986 and asked me to be his deputy, I had only one condition. I wanted to be the only deputy: our predecessors had three. Frank readily agreed and the other two deputies were redesignated and given other duties.

  I kept this model when I replaced Frank as the National Security Advisor, and followed the same model when I became Chairman of the Joint Chiefs of Staff in October 1989. My immediate line subordinates were two- and three-star generals and admirals, experienced officers at the top of their profession, each averaging more than twenty years of service.