The Power of Habit – Why We Do What We Do in Life and Business

I.

On a blustery October day in 1987, a herd of prominent Wall Street investors and stock analysts gathered in the ballroom of a posh Manhattan hotel. They were there to meet the new CEO of the Aluminum Company of America-or Alcoa, as it was known-a corporation that, for nearly a century, had manufactured everything from the foil that wraps Hershey’s Kisses and the metal in Coca-Cola cans to the bolts that hold satellites together. [101]

Alcoa’s founder had invented the process for smelting aluminum a century earlier, and since then the company had become one of the largest on earth. Many of the people in the audience had invested millions of dollars in Alcoa stock and had enjoyed a steady return. In the past year, however, investor grumblings started. Alcoa’s management had made misstep after misstep, unwisely trying to expand into new product lines while competitors stole customers and profits away.

So there had been a palpable sense of relief when Alcoa’s board announced it was time for new leadership. That relief, though, turned to unease when the choice was announced: the new CEO would be a former government bureaucrat named Paul O’Neill. Many on Wall Street had never heard of him. When Alcoa scheduled this meet and greet at the Manhattan ballroom, every major investor asked for an invitation.

A few minutes before noon, O’Neill took the stage. He was fifty-one years old, trim, and dressed in gray pinstripes and a red power tie. His hair was white and his posture military straight. He bounced up the steps and smiled warmly. He looked dignified, solid, confident. Like a chief executive.

Then he opened his mouth.

“I want to talk to you about worker safety,” he said. “Every year, numerous Alcoa workers are injured so badly that they miss a day of work. Our safety record is better than the general American workforce, especially considering that our employees work with metals that are 1500 degrees and machines that can rip a man’s arm off. But it’s not good enough. I intend to make Alcoa the safest company in America. I intend to go for zero injuries.”

The audience was confused. These meetings usually followed a predictable script: A new CEO would start with an introduction, make a faux self-deprecating joke-something about how he slept his way through Harvard Business School-then promise to boost profits and lower costs. Next would come an excoriation of taxes, business regulations, and sometimes, with a fervor that suggested firsthand experience in divorce court, lawyers. Finally, the speech would end with a blizzard of buzzwords-“synergy,” “rightsizing,” and “co-opetition”-at which point everyone could return to their offices, reassured that capitalism was safe for another day.

O’Neill hadn’t said anything about profits. He didn’t mention taxes. There was no talk of “using alignment to achieve a win-win synergistic market advantage.” For all anyone in the audience knew, given his talk of worker safety, O’Neill might be pro-regulation. Or, worse, a Democrat. It was a terrifying prospect.

“Now, before I go any further,” O’Neill said, “I want to point out the safety exits in this room.” He gestured to the rear of the ballroom. “There’s a couple of doors in the back, and in the unlikely event of a fire or other emergency, you should calmly walk out, go down the stairs to the lobby, and leave the building.”

Silence. The only noise was the hum of traffic through the windows. Safety? Fire exits? Was this a joke? One investor in the audience knew that O’Neill had been in Washington, D.C., during the sixties. Guy must have done a lot of drugs, he thought.

Eventually, someone raised a hand and asked about inventories in the aerospace division. Another asked about the company’s capital ratios.

“I’m not certain you heard me,” O’Neill said. “If you want to understand how Alcoa is doing, you need to look at our workplace safety figures. If we bring our injury rates down, it won’t be because of cheerleading or the nonsense you sometimes hear from other CEOs. It will be because the individuals at this company have agreed to become part of something important: They’ve devoted themselves to creating a habit of excellence. Safety will be an indicator that we’re making progress in changing our habits across the entire institution. That’s how we should be judged.”

The investors in the room almost stampeded out the doors when the presentation ended. One jogged to the lobby, found a pay phone, and called his twenty largest clients.

“I said, ‘The board put a crazy hippie in charge and he’s going to kill the company,’ ” that investor told me. “I ordered them to sell their stock immediately, before everyone else in the room started calling their clients and telling them the same thing.

“It was literally the worst piece of advice I gave in my entire career.”

Within a year of O’Neill’s speech, Alcoa’s profits would hit a record high. By the time O’Neill retired in 2000, the company’s annual net income was five times larger than before he arrived, and its market capitalization had risen by $27 billion. Someone who invested a million dollars in Alcoa on the day O’Neill was hired would have earned another million dollars in dividends while he headed the company, and the value of their stock would be five times bigger when he left.

What’s more, all that growth occurred while Alcoa became one of the safest companies in the world. Before O’Neill’s arrival, almost every Alcoa plant had at least one accident per week. Once his safety plan was implemented, some facilities would go years without a single employee losing a workday due to an accident. The company’s worker injury rate fell to one-twentieth the U.S. average.

So how did O’Neill make one of the largest, stodgiest, and most potentially dangerous companies into a profit machine and a bastion of safety?

By attacking one habit and then watching the changes ripple through the organization.

“I knew I had to transform Alcoa,” O’Neill told me. “But you can’t order people to change. That’s not how the brain works. So I decided I was going to start by focusing on one thing. If I could start disrupting the habits around one thing, it would spread throughout the entire company.”

O’Neill believed that some habits have the power to start a chain reaction, changing other habits as they move through an organization. Some habits, in other words, matter more than others in remaking businesses and lives. These are “keystone habits,” and they can influence how people work, eat, play, live, spend, and communicate. Keystone habits start a process that, over time, transforms everything.

Keystone habits say that success doesn’t depend on getting every single thing right, but instead relies on identifying a few key priorities and fashioning them into powerful levers. This book’s first section explained how habits work, how they can be created and changed. However, where should a would-be habit master start? Understanding keystone habits holds the answer to that question: The habits that matter most are the ones that, when they start to shift, dislodge and remake other patterns.

Keystone habits explain how Michael Phelps became an Olympic champion and why some college students outperform their peers. They describe why some people, after years of trying, suddenly lose forty pounds while becoming more productive at work and still getting home in time for dinner with their kids. And keystone habits explain how Alcoa became one of the best performing stocks in the Dow Jones index, while also becoming one of the safest places on earth.

When Alcoa first approached O’Neill about becoming CEO, he wasn’t sure he wanted the job. He’d already earned plenty of money, and his wife liked Connecticut, where they lived. They didn’t know anything about Pittsburgh, where Alcoa was headquartered. But before turning down the offer, O’Neill asked for some time to think it over. To help himself make the decision, he started working on a list of what would be his biggest priorities if he accepted the post.

O’Neill had always been a big believer in lists. Lists were how he organized his life. In college at Fresno State-where he finished his courses in a bit over three years, while also working thirty hours a week-O’Neill had drafted a list of everything he hoped to accomplish during his lifetime, including, near the top, “Make a Difference.” After graduating in 1960, at a friend’s encouragement, O’Neill picked up an application for a federal internship and, along with three hundred thousand others, took the government employment exam. Three thousand people were chosen for interviews. Three hundred of them were offered jobs. O’Neill was one. [102]

He started as a middle manager at the Veterans Administration and was told to learn about computer systems. All the while, O’Neill kept writing his lists, recording why some projects were more successful than others, which contractors delivered on time and which didn’t. He was promoted each year. And as he rose through the VA’s ranks, he made a name for himself as someone whose lists always seemed to include a bullet point that got a problem solved.

By the mid-1960s, such skills were in high demand in Washington, D.C. Robert McNamara had recently remade the Pentagon by hiring a crop of young mathematicians, statisticians, and computer programmers. President Johnson wanted some whiz kids of his own. So O’Neill was recruited to what eventually became known as the Office of Management and Budget, one of D.C.’s most powerful agencies. Within a decade, at age thirty-eight, he was promoted to deputy director and was, suddenly, among the most influential people in town.

That’s when O’Neill’s education in organizational habits really started. One of his first assignments was to create an analytical framework for studying how the government was spending money on health care. He quickly figured out that the government’s efforts, which should have been guided by logical rules and deliberate priorities, were instead driven by bizarre institutional processes that, in many ways, operated like habits. Bureaucrats and politicians, rather than making decisions, were responding to cues with automatic routines in order to get rewards such as promotions or reelection. It was the habit loop-spread across thousands of people and billions of dollars.

For instance, after World War II, Congress had created a program to build community hospitals. A quarter century later, it was still chugging along, and so whenever lawmakers allocated new health-care funds, bureaucrats immediately started building. The towns where the new hospitals were located didn’t necessarily need more patient beds, but that didn’t matter. What mattered was erecting a big structure that a politician could point to while stumping for votes. [103]

Federal workers would “spend months debating blue or yellow curtains, figuring out if patient rooms should contain one or two televisions, designing nurses’ stations, real pointless stuff,” O’Neill told me. “Most of the time, no one ever asked if the town wanted a hospital. The bureaucrats had gotten into a habit of solving every medical problem by building something so that a congressman could say, ‘Here’s what I did!’ It didn’t make any sense, but everybody did the same thing again and again.”

Researchers have found institutional habits in almost every organization or company they’ve scrutinized. “Individuals have habits; groups have routines,” wrote the academic Geoffrey Hodgson, who spent a career examining organizational patterns. “Routines are the organizational analogue of habits.” [104]

To O’Neill, these kinds of habits seemed dangerous. “We were basically ceding decision making to a process that occurred without actually thinking,” O’Neill said. But at other agencies, where change was in the air, good organizational habits were creating success.

Some departments at NASA, for instance, were overhauling themselves by deliberately instituting organizational routines that encouraged engineers to take more risks. When unmanned rockets exploded on takeoff, department heads would applaud, so that everyone would know their division had tried and failed, but at least they had tried. Eventually, mission control filled with applause every time something expensive blew up. It became an organizational habit. [105] Or take the Environmental Protection Agency, which was created in 1970. The EPA’s first administrator, William Ruckelshaus, consciously engineered organizational habits that encouraged his regulators to be aggressive on enforcement. When lawyers asked for permission to file a lawsuit or enforcement action, it went through a process for approval. [106] The default was authorization to go ahead. The message was clear: At the EPA, aggression gets rewarded. By 1975, the EPA was issuing more than fifteen hundred new environmental rules a year. [107]

“Every time I looked at a different part of the government, I found these habits that seemed to explain why things were either succeeding or failing,” O’Neill told me. “The best agencies understood the importance of routines. The worst agencies were headed by people who never thought about it, and then wondered why no one followed their orders.”

In 1977, after sixteen years in Washington, D.C., O’Neill decided it was time to leave. He was working fifteen hours a day, seven days a week, and his wife was tired of raising four children on her own. O’Neill resigned and landed a job with International Paper, the world’s largest pulp and paper company. He eventually became its president.

By then, some of his old government friends were on Alcoa’s board. When the company needed a new chief executive, they thought of him, which is how he ended up writing a list of his priorities if he decided to take the job.

At the time, Alcoa was struggling. Critics said the company’s workers weren’t nimble enough and the quality of its products was poor. But at the top of O’Neill’s list he didn’t write “quality” or “efficiency” as his biggest priorities. At a company as big and as old as Alcoa, you can’t flip a switch and expect everyone to work harder or produce more. The previous CEO had tried to mandate improvements, and fifteen thousand employees had gone on strike. It got so bad they would bring dummies to the parking lots, dress them like managers, and burn them in effigy. “Alcoa was not a happy family,” one person from that period told me. “It was more like the Manson family, but with the addition of molten metal.”

O’Neill figured his top priority, if he took the job, would have to be something that everybody-unions and executives-could agree was important. He needed a focus that would bring people together, that would give him leverage to change how people worked and communicated.

“I went to basics,” he told me. “Everyone deserves to leave work as safely as they arrive, right? You shouldn’t be scared that feeding your family is going to kill you. That’s what I decided to focus on: changing everyone’s safety habits.”

At the top of O’Neill’s list he wrote down “SAFETY” and set an audacious goal: zero injuries. Not zero factory injuries. Zero injuries, period. That would be his commitment no matter how much it cost.

O’Neill decided to take the job.

“I’m really glad to be here,” O’Neill told a room full of workers at a smelting plant in Tennessee a few months after he was hired. Not everything had gone smoothly. Wall Street was still panicked. The unions were concerned. Some of Alcoa’s vice presidents were miffed at being passed over for the top job. And O’Neill kept talking about worker safety.

“I’m happy to negotiate with you about anything,” O’Neill said. He was on a tour of Alcoa’s American plants, after which he was going to visit the company’s facilities in thirty-one other countries. “But there’s one thing I’m never going to negotiate with you, and that’s safety. I don’t ever want you to say that we haven’t taken every step to make sure people don’t get hurt. If you want to argue with me about that, you’re going to lose.”

The brilliance of this approach was that no one, of course, wanted to argue with O’Neill about worker safety. Unions had been fighting for better safety rules for years. Managers didn’t want to argue about it, either, since injuries meant lost productivity and low morale.

What most people didn’t realize, however, was that O’Neill’s plan for getting to zero injuries entailed the most radical realignment in Alcoa’s history. The key to protecting Alcoa employees, O’Neill believed, was understanding why injuries happened in the first place. And to understand why injuries happened, you had to study how the manufacturing process was going wrong. To understand how things were going wrong, you had to bring in people who could educate workers about quality control and the most efficient work processes, so that it would be easier to do everything right, since correct work is also safer work.

In other words, to protect workers, Alcoa needed to become the best, most streamlined aluminum company on earth.

O’Neill’s safety plan, in effect, was modeled on the habit loop. He identified a simple cue: an employee injury. He instituted an automatic routine: Any time someone was injured, the unit president had to report it to O’Neill within twenty-four hours and present a plan for making sure the injury never happened again. [108] [109] And there was a reward: The only people who got promoted were those who embraced the system.

Unit presidents were busy people. To contact O’Neill within twenty-four hours of an injury, they needed to hear about an accident from their vice presidents as soon as it happened. So vice presidents needed to be in constant communication with floor managers. And floor managers needed to get workers to raise warnings as soon as they saw a problem and keep a list of suggestions nearby, so that when the vice president asked for a plan, there was an idea box already full of possibilities. To make all of that happen, each unit had to build new communication systems that made it easier for the lowliest worker to get an idea to the loftiest executive, as fast as possible. Almost everything about the company’s rigid hierarchy had to change to accommodate O’Neill’s safety program. He was building new corporate habits.

ALCOA’S INSTITUTIONAL HABIT LOOP

As Alcoa’s safety patterns shifted, other aspects of the company started changing with startling speed, as well. Rules that unions had spent decades opposing-such as measuring the productivity of individual workers-were suddenly embraced, because such measurements helped everyone figure out when part of the manufacturing process was getting out of whack, posing a safety risk. Policies that managers had long resisted-such as giving workers autonomy to shut down a production line when the pace became overwhelming-were now welcomed, because that was the best way to stop injuries before they occurred. The company shifted so much that some employees found safety habits spilling into other parts of their lives.

“Two or three years ago, I’m in my office, looking at the Ninth Street bridge out the window, and there’s some guys working who aren’t using correct safety procedures,” said Jeff Shockey, Alcoa’s current safety director. One of them was standing on top of the bridge’s guardrail, while the other held on to his belt. They weren’t using safety harnesses or ropes. “They worked for some company that has nothing to do with us, but without thinking about it, I got out of my chair, went down five flights of stairs, walked over the bridge and told these guys, hey, you’re risking your life, you have to use your harness and safety gear.” The men explained their supervisor had forgotten to bring the equipment. So Shockey called the local Occupational Safety and Health Administration office and turned the supervisor in.

“Another executive told me that one day, he stopped at a street excavation near his house because they didn’t have a trench box, and gave everyone a lecture on the importance of proper procedures. It was the weekend, and he stopped his car, with his kids in the back, to lecture city workers about trench safety. That isn’t natural, but that’s kind of the point. We do this stuff without thinking about it now.”

O’Neill never promised that his focus on worker safety would increase Alcoa’s profits. However, as his new routines moved through the organization, costs came down, quality went up, and productivity skyrocketed. If molten metal was injuring workers when it splashed, then the pouring system was redesigned, which led to fewer injuries. It also saved money because Alcoa lost less raw materials in spills. If a machine kept breaking down, it was replaced, which meant there was less risk of a broken gear snagging an employee’s arm. It also meant higher quality products because, as Alcoa discovered, equipment malfunctions were a chief cause of subpar aluminum.

Researchers have found similar dynamics in dozens of other settings, including individuals’ lives.

Take, for instance, studies from the past decade examining the impacts of exercise on daily routines. [110] When people start habitually exercising, even as infrequently as once a week, they start changing other, unrelated patterns in their lives, often unknowingly. Typically, people who exercise start eating better and becoming more productive at work. They smoke less and show more patience with colleagues and family. They use their credit cards less frequently and say they feel less stressed. It’s not completely clear why. But for many people, exercise is a keystone habit that triggers widespread change. “Exercise spills over,” said James Prochaska, a University of Rhode Island researcher. “There’s something about it that makes other good habits easier.”

Studies have documented that families who habitually eat dinner together seem to raise children with better homework skills, higher grades, greater emotional control, and more confidence. [111] Making your bed every morning is correlated with better productivity, a greater sense of well-being, and stronger skills at sticking with a budget. [112] It’s not that a family meal or a tidy bed causes better grades or less frivolous spending. But somehow those initial shifts start chain reactions that help other good habits take hold.

If you focus on changing or cultivating keystone habits, you can cause widespread shifts. However, identifying keystone habits is tricky. To find them, you have to know where to look. Detecting keystone habits means searching out certain characteristics. Keystone habits offer what is known within academic literature as “small wins.” They help other habits to flourish by creating new structures, and they establish cultures where change becomes contagious.

But as O’Neill and countless others have found, crossing the gap between understanding those principles and using them requires a bit of ingenuity.

II.

When Michael Phelps’s alarm clock went off at 6:30 A.M. on the morning of August 13, 2008, he crawled out of bed in the Olympic Village in Beijing and fell right into his routine.

He pulled on a pair of sweatpants and walked to breakfast. He had already won three gold medals earlier that week-giving him nine in his career-and had two races that day. By 7 A.M. [113], he was in the cafeteria, eating his regular race-day menu of eggs, oatmeal, and four energy shakes, the first of more than six thousand calories he would consume over the next sixteen hours.

Phelps’s first race-the 200-meter butterfly, his strongest event-was scheduled for ten o’clock. Two hours before the starting gun fired, he began his usual stretching regime, starting with his arms, then his back, then working down to his ankles, which were so flexible they could extend more than ninety degrees, farther than a ballerina’s en pointe. At eight-thirty, he slipped into the pool and began his first warm-up lap, 800 meters of mixed styles, followed by 600 meters of kicking, 400 meters pulling a buoy between his legs, 200 meters of stroke drills, and a series of 25-meter sprints to elevate his heart rate. The workout took precisely forty-five minutes.

At nine-fifteen, he exited the pool and started squeezing into his LZR Racer, a bodysuit so tight it required twenty minutes of tugging to put it on. Then he clamped headphones over his ears, cranked up the hip-hop mix he played before every race, and waited.

Phelps had started swimming when he was seven years old to burn off some of the energy that was driving his mom and teachers crazy. When a local swimming coach named Bob Bowman saw Phelps’s long torso, big hands, and relatively short legs (which offered less drag in the water), he knew Phelps could become a champion. But Phelps was emotional. He had trouble calming down before races. His parents were divorcing, and he had problems coping with the stress. Bowman purchased a book of relaxation exercises and asked Phelps’s mom to read them aloud every night. The book contained a script-“Tighten your right hand into a fist and release it. Imagine the tension melting away”-that tensed and relaxed each part of Phelps’s body before he fell asleep.

Bowman believed that for swimmers, the key to victory was creating the right routines. Phelps, Bowman knew, had a perfect physique for the pool. That said, everyone who eventually competes at the Olympics has perfect musculature. Bowman could also see that Phelps, even at a young age, had a capacity for obsessiveness that made him an ideal athlete. Then again, all elite performers are obsessives.

What Bowman could give Phelps, however-what would set him apart from other competitors-were habits that would make him the strongest mental swimmer in the pool. He didn’t need to control every aspect of Phelps’s life. All he needed to do was target a few specific habits that had nothing to do with swimming and everything to do with creating the right mind-set. He designed a series of behaviors that Phelps could use to become calm and focused before each race, to find those tiny advantages that, in a sport where victory can come in milliseconds, would make all the difference.

When Phelps was a teenager, for instance, at the end of each practice, Bowman would tell him to go home and “watch the videotape. Watch it before you go to sleep and when you wake up.”

The videotape wasn’t real. Rather, it was a mental visualization of the perfect race. Each night before falling asleep and each morning after waking up, Phelps would imagine himself jumping off the blocks and, in slow motion, swimming flawlessly. He would visualize his strokes, the walls of the pool, his turns, and the finish. He would imagine the wake behind his body, the water dripping off his lips as his mouth cleared the surface, what it would feel like to rip off his cap at the end. He would lie in bed with his eyes shut and watch the entire competition, the smallest details, again and again, until he knew each second by heart.

During practices, when Bowman ordered Phelps to swim at race speed, he would shout, “Put in the videotape!” and Phelps would push himself, as hard as he could. It almost felt anticlimactic as he cut through the water. He had done this so many times in his head that, by now, it felt rote. But it worked. He got faster and faster. Eventually, all Bowman had to do before a race was whisper, “Get the videotape ready,” and Phelps would settle down and crush the competition.

And once Bowman established a few core routines in Phelps’s life, all the other habits-his diet and practice schedules, the stretching and sleep routines-seemed to fall into place on their own. At the core of why those habits were so effective, why they acted as keystone habits, was something known within academic literature as a “small win.”

Small wins are exactly what they sound like, and are part of how keystone habits create widespread changes. A huge body of research has shown that small wins have enormous power, an influence disproportionate to the accomplishments of the victories themselves. “Small wins are a steady application of a small advantage,” one Cornell professor wrote in 1984. “Once a small win has been accomplished, forces are set in motion that favor another small win.” [114] Small wins fuel transformative changes by leveraging tiny advantages into patterns that convince people that bigger achievements are within reach. [115]

For example, when gay rights organizations started campaigning against homophobia in the late 1960s, their initial efforts yielded only a string of failures. They pushed to repeal laws used to prosecute gays and were roundly defeated in state legislatures. Teachers tried to create curriculums to counsel gay teens, and were fired for suggesting that homosexuality should be embraced. It seemed like the gay community’s larger goals-ending discrimination and police harassment, convincing the American Psychiatric Association to stop defining homosexuality as a mental disease-were out of reach. [116]

Then, in the early 1970s, the American Library Association’s Task Force on Gay Liberation decided to focus on one modest goal: convincing the Library of Congress to reclassify books about the gay liberation movement from HQ 71-471 (“Abnormal Sexual Relations, Including Sexual Crimes”) to another, less pejorative category. [117]

In 1972, after receiving a letter requesting the reclassification, the Library of Congress agreed to make the shift, reclassifying books into a newly created category, HQ 76.5 (“Homosexuality, Lesbianism-Gay Liberation Movement, Homophile Movement”). It was a minor tweak of an old institutional habit regarding how books were shelved, but the effect was electrifying. News of the new policy spread across the nation. Gay rights organizations, citing the victory, started fund-raising drives. Within a few years, openly gay politicians were running for political office in California, New York, Massachusetts, and Oregon, many of them citing the Library of Congress’s decision as inspiration. In 1973, the American Psychiatric Association, after years of internal debate, rewrote the definition of homosexuality so it was no longer a mental illness-paving the way for the passage of state laws that made it illegal to discriminate against people because of their sexual orientation.

And it all began with one small win.

“Small wins do not combine in a neat, linear, serial form, with each step being a demonstrable step closer to some predetermined goal,” wrote Karl Weick, a prominent organizational psychologist. “More common is the circumstance where small wins are scattered… like miniature experiments that test implicit theories about resistance and opportunity and uncover both resources and barriers that were invisible before the situation was stirred up.”

Which is precisely what happened with Michael Phelps. When Bob Bowman started working with Phelps and his mother on the keystone habits of visualization and relaxation, neither Bowman nor Phelps had any idea what they were doing. “We’d experiment, try different things until we found stuff that worked,” Bowman told me. “Eventually we figured out it was best to concentrate on these tiny moments of success and build them into mental triggers. We worked them into a routine. There’s a series of things we do before every race that are designed to give Michael a sense of building victory.

“If you were to ask Michael what’s going on in his head before competition, he would say he’s not really thinking about anything. He’s just following the program. But that’s not right. It’s more like his habits have taken over. When the race arrives, he’s more than halfway through his plan and he’s been victorious at every step. All the stretches went like he planned. The warm-up laps were just like he visualized. His headphones are playing exactly what he expected. The actual race is just another step in a pattern that started earlier that day and has been nothing but victories. Winning is a natural extension.”

Back in Beijing, it was 9:56 A.M.-four minutes before the race’s start-and Phelps stood behind his starting block, bouncing slightly on his toes. When the announcer said his name, Phelps stepped onto the block, as he always did before a race, and then stepped down, as he always did. He swung his arms three times, as he had before every race since he was twelve years old. He stepped up on the blocks again, got into his stance, and, when the gun sounded, leapt.

Phelps knew that something was wrong as soon as he hit the water. There was moisture inside his goggles. He couldn’t tell if they were leaking from the top or bottom, but as he broke the water’s surface and began swimming, he hoped the leak wouldn’t become too bad. [118]

By the second turn, however, everything was getting blurry. As he approached the third turn and final lap, the cups of his goggles were completely filled. Phelps couldn’t see anything. Not the line along the pool’s bottom, not the black T marking the approaching wall. He couldn’t see how many strokes were left. For most swimmers, losing your sight in the middle of an Olympic final would be cause for panic.

Phelps was calm.

Everything else that day had gone according to plan. The leaking goggles were a minor deviation, but one for which he was prepared. Bowman had once made Phelps swim in a Michigan pool in the dark, believing that he needed to be ready for any surprise. Some of the videotapes in Phelps’s mind had featured problems like this. He had mentally rehearsed how he would respond to a goggle failure. As he started his last lap, Phelps estimated how many strokes the final push would require-nineteen or twenty, maybe twenty-one-and started counting. He felt totally relaxed as he swam at full strength. Midway through the lap he began to increase his effort, a final eruption that had become one of his main techniques in overwhelming opponents. At eighteen strokes, he started anticipating the wall. He could hear the crowd roaring, but since he was blind, he had no idea if they were cheering for him or someone else. Nineteen strokes, then twenty. It felt like he needed one more. That’s what the videotape in his head said. He made a twenty-first, huge stroke, glided with his arm outstretched, and touched the wall. He had timed it perfectly. When he ripped off his goggles and looked up at the scoreboard, it said “WR”-world record-next to his name. He’d won another gold.

After the race, a reporter asked what it had felt like to swim blind.

“It felt like I imagined it would,” Phelps said. It was one additional victory in a lifetime full of small wins. [119]

Six months after Paul O’Neill became CEO of Alcoa, he got a telephone call in the middle of the night. A plant manager in Arizona was on the line, panicked, talking about how an extrusion press had stopped operating and one of the workers-a young man who had joined the company a few weeks earlier, eager for the job because it offered health care for his pregnant wife-had tried a repair. He had jumped over a yellow safety wall surrounding the press and walked across the pit. There was a piece of aluminum jammed into the hinge on a swinging six-foot arm. The young man pulled on the aluminum scrap, removing it. The machine was fixed. Behind him, the arm restarted its arc, swinging toward his head. When it hit, the arm crushed his skull. He was killed instantly. [120]

Fourteen hours later, O’Neill ordered all the plant’s executives-as well as Alcoa’s top officers in Pittsburgh-into an emergency meeting. For much of the day, they painstakingly re-created the accident with diagrams and by watching videotapes again and again. They identified dozens of errors that had contributed to the death, including two managers who had seen the man jump over the barrier but failed to stop him; a training program that hadn’t emphasized to the man that he wouldn’t be blamed for a breakdown; lack of instructions that he should find a manager before attempting a repair; and the absence of sensors to automatically shut down the machine when someone stepped into the pit.

“We killed this man,” a grim-faced O’Neill told the group. “It’s my failure of leadership. I caused his death. And it’s the failure of all of you in the chain of command.”

The executives in the room were taken aback. Sure, a tragic accident had occurred, but tragic accidents were part of life at Alcoa. It was a huge company with employees who handled red-hot metal and dangerous machines. “Paul had come in as an outsider, and there was a lot of skepticism when he talked about safety,” said Bill O’rourke, a top executive. “We figured it would last a few weeks, and then he would start focusing on something else. But that meeting really shook everyone up. He was serious about this stuff, serious enough that he would stay up nights worrying about some employee he’d never met. That’s when things started to change.”

Within a week of that meeting, all the safety railings at Alcoa’s plants were repainted bright yellow, and new policies were written up. Managers told employees not to be afraid to suggest proactive maintenance, and rules were clarified so that no one would attempt unsafe repairs. The newfound vigilance resulted in a short-term, noticeable decline in the injury rate. Alcoa experienced a small win.

Then O’Neill pounced.

“I want to congratulate everyone for bringing down the number of accidents, even just for two weeks,” he wrote in a memo that made its way through the entire company. “We shouldn’t celebrate because we’ve followed the rules, or brought down a number. We should celebrate because we are saving lives.”

Workers made copies of the note and taped it to their lockers. Someone painted a mural of O’Neill on one of the walls of a smelting plant with a quote from the memo inscribed underneath. Just as Michael Phelps’s routines had nothing to do with swimming and everything to do with his success, so O’Neill’s efforts began snowballing into changes that were unrelated to safety, but transformative nonetheless.

“I said to the hourly workers, ‘If your management doesn’t follow up on safety issues, then call me at home, here’s my number,’ ” O’Neill told me. “Workers started calling, but they didn’t want to talk about accidents. They wanted to talk about all these other great ideas.”

The Alcoa plant that manufactured aluminum siding for houses, for instance, had been struggling for years because executives would try to anticipate popular colors and inevitably guess wrong. They would pay consultants millions of dollars to choose shades of paint and six months later, the warehouse would be overflowing with “sunburst yellow” and out of suddenly in-demand “hunter green.” One day, a low-level employee made a suggestion that quickly worked its way to the general manager: If they grouped all the painting machines together, they could switch out the pigments faster and become more nimble in responding to shifts in customer demand. Within a year, profits on aluminum siding doubled.

The small wins that started with O’Neill’s focus on safety created a climate in which all kinds of new ideas bubbled up.

“It turns out this guy had been suggesting this painting idea for a decade, but hadn’t told anyone in management,” an Alcoa executive told me. “Then he figures, since we keep on asking for safety recommendations, why not tell them about this other idea? It was like he gave us the winning lottery numbers.”

III.

When a young Paul O’Neill was working for the government and creating a framework for analyzing federal spending on health care, one of the foremost issues concerning officials was infant mortality. The United States, at the time, was one of the wealthiest countries on earth. Yet it had a higher infant mortality rate than most of Europe and some parts of South America. Rural areas, in particular, saw a staggering number of babies die before their first birthdays. [121]

O’Neill was tasked with figuring out why. He asked other federal agencies to start analyzing infant mortality data, and each time someone came back with an answer, he’d ask another question, trying to get deeper, to understand the problem’s root causes. Whenever someone came into O’Neill’s office with some discovery, O’Neill would start interrogating them with new inquiries. He drove people crazy with his never-ending push to learn more, to understand what was really going on. (“I love Paul O’Neill, but you could not pay me enough to work for him again,” one official told me. “The man has never encountered an answer he can’t turn into another twenty hours of work.”)

Some research, for instance, suggested that the biggest cause of infant deaths was premature births. And the reason babies were born too early was that mothers suffered from malnourishment during pregnancy. So to lower infant mortality, improve mothers’ diets. Simple, right? But to stop malnourishment, women had to improve their diets before they became pregnant. Which meant the government had to start educating women about nutrition before they became sexually active. Which meant officials had to create nutrition curriculums inside high schools.

However, when O’Neill began asking about how to create those curriculums, he discovered that many high school teachers in rural areas didn’t know enough basic biology to teach nutrition. So the government had to remake how teachers were getting educated in college, and give them a stronger grounding in biology so they could eventually teach nutrition to teenage girls, so those teenagers would eat better before they started having sex, and, eventually, be sufficiently nourished when they had children.

Poor teacher training, the officials working with O’Neill finally figured out, was a root cause of high infant mortality. If you asked doctors or public health officials for a plan to fight infant deaths, none of them would have suggested changing how teachers are trained. They wouldn’t have known there was a link. However, by teaching college students about biology, you made it possible for them to eventually pass on that knowledge to teenagers, who started eating healthier, and years later give birth to stronger babies. Today, the U.S. [122] infant mortality rate is 68 percent lower than when O’Neill started the job.

O’Neill’s experiences with infant mortality illustrate the second way that keystone habits encourage change: by creating structures that help other habits to flourish. In the case of premature deaths, changing collegiate curriculums for teachers started a chain reaction that eventually trickled down to how girls were educated in rural areas, and whether they were sufficiently nourished when they became pregnant. And O’Neill’s habit of constantly pushing other bureaucrats to continue researching until they found a problem’s root causes overhauled how the government thought about problems like infant mortality.

The same thing can happen in people’s lives. For example, until about twenty years ago, conventional wisdom held that the best way for people to lose weight was to radically alter their lives. Doctors would give obese patients strict diets and tell them to join a gym, attend regular counseling sessions-sometimes as often as every day-and shift their daily routines by walking up stairs, for instance, instead of taking the elevator. Only by completely shaking up someone’s life, the thinking went, could their bad habits be reformed.

But when researchers studied the effectiveness of these methods over prolonged periods, they discovered they were failures. Patients would use the stairs for a few weeks, but by the end of the month, it was too much hassle. They began diets and joined gyms, but after the initial burst of enthusiasm wore off, they slid back into their old eating and TV-watching habits. [123] Piling on so much change at once made it impossible for any of it to stick.

Then, in 2009 a group of researchers funded by the National Institutes of Health published a study of a different approach to weight loss. [124] They had assembled a group of sixteen hundred obese people and asked them to concentrate on writing down everything they ate at least one day per week.

It was hard at first. The subjects forgot to carry their food journals, or would snack and not note it. Slowly, however, people started recording their meals once a week-and sometimes, more often. Many participants started keeping a daily food log. Eventually, it became a habit. Then, something unexpected happened. The participants started looking at their entries and finding patterns they didn’t know existed. Some noticed they always seemed to snack at about 10 A.M., so they began keeping an apple or banana on their desks for midmorning munchies. Others started using their journals to plan future menus, and when dinner rolled around, they ate the healthy meal they had written down, rather than junk food from the fridge.

The researchers hadn’t suggested any of these behaviors. They had simply asked everyone to write down what they ate once a week. But this keystone habit-food journaling-created a structure that helped other habits to flourish. Six months into the study, people who kept daily food records had lost twice as much weight as everyone else.

“After a while, the journal got inside my head,” one person told me. [125] “I started thinking about meals differently. It gave me a system for thinking about food without becoming depressed.”

Something similar happened at Alcoa after O’Neill took over. Just as food journals provided a structure for other habits to flourish, O’Neill’s safety habits created an atmosphere in which other behaviors emerged. Early on, O’Neill took the unusual step of ordering Alcoa’s offices around the world to link up in an electronic network. This was in the early 1980s, when large, international networks weren’t usually connected to people’s desktop computers. O’Neill justified his order by arguing that it was essential to create a real-time safety data system that managers could use to share suggestions. As a result, Alcoa developed one of the first genuinely worldwide corporate email systems.

O’Neill logged on every morning and sent messages to make sure everyone else was logged on as well. At first, people used the network primarily to discuss safety issues. Then, as email habits became more ingrained and comfortable, they started posting information on all kinds of other topics, such as local market conditions, sales quotas, and business problems. High-ranking executives were required to send in a report every Friday, which anyone in the company could read. A manager in Brazil used the network to send a colleague in New York data on changes in the price of steel. The New Yorker took that information and turned a quick profit for the company on Wall Street. Pretty soon, everyone was using the system to communicate about everything. “I would send in my accident report, and I knew everyone else read it, so I figured, why not send pricing information, or intelligence on other companies?” one manager told me. “It was like we had discovered a secret weapon. The competition couldn’t figure out how we were doing it.”

When the Web blossomed, Alcoa was perfectly positioned to take advantage. O’Neill’s keystone habit-worker safety-had created a platform that encouraged another practice-email-years ahead of competitors.

By 1996, Paul O’Neill had been at Alcoa for almost a decade. His leadership had been studied by the Harvard Business School and the Kennedy School of Government. He was regularly mentioned as a potential commerce secretary or secretary of defense. His employees and the unions gave him high marks. Under his watch, Alcoa’s stock price had risen more than 200 percent. He was, at last, a universally acknowledged success.

In May of that year, at a shareholder meeting in downtown Pittsburgh, a Benedictine nun stood up during the question-and-answer session and accused O’Neill of lying. Sister Mary Margaret represented a social advocacy group concerned about wages and conditions inside an Alcoa plant in Ciudad Acuña, Mexico. She said that while O’Neill extolled Alcoa’s safety measures, workers in Mexico were becoming sick because of dangerous fumes.

“It’s untrue,” O’Neill told the room. On his laptop, he pulled up the safety records from the Mexican plant. “See?” he said, showing the room its high scores on safety, environmental compliance, and employee satisfaction surveys. The executive in charge of the facility, Robert Barton, was one of Alcoa’s most senior managers. He had been with the company for decades and was responsible for some of their largest partnerships. The nun said that the audience shouldn’t trust O’Neill. She sat down.

After the meeting, O’Neill asked her to come to his office. The nun’s religious order owned fifty Alcoa shares, and for months they had been asking for a shareholder vote on a resolution to review the company’s Mexican operations. O’Neill asked Sister Mary if she had been to any of the plants herself. No, she told him. To be safe, O’Neill asked the company’s head of human resources and general counsel to fly to Mexico to see what was going on.

When the executives arrived, they poked through the Acuña plant’s records, and found reports of an incident that had never been sent to headquarters. A few months earlier, there had been a buildup of fumes within a building. It was a relatively minor event. The plant’s executive, Barton, had installed ventilators to remove the gases. The people who had become ill had fully recovered within a day or two.

But Barton had never reported the illnesses.

When the executives returned to Pittsburgh and presented their findings, O’Neill had a question.

“Did Bob Barton know that people had gotten sick?”

“We didn’t meet with him,” they answered. “But, yeah, it’s pretty clear he knew.”

Two days later, Barton was fired.

The exit shocked outsiders. Barton had been mentioned in articles as one of the company’s most valuable executives. His departure was a blow to important joint ventures.

Within Alcoa, however, no one was surprised. It was seen as an inevitable extension of the culture that O’Neill had built.

“Barton fired himself,” one of his colleagues told me. “There wasn’t even a choice there.”

This is the final way that keystone habits encourage widespread change: by creating cultures where new values become ingrained. Keystone habits make tough choices-such as firing a top executive-easier, because when that person violates the culture, it’s clear they have to go. Sometimes these cultures manifest themselves in special vocabularies, the use of which becomes, itself, a habit that defines an organization. At Alcoa, for instance, there were “Core Programs” and “Safety Philosophies,” phrases that acted like suitcases, containing whole conversations about priorities, goals, and ways of thinking.

“It might have been hard at another company to fire someone who had been there so long,” O’Neill told me. “It wasn’t hard for me. It was clear what our values dictated. He got fired because he didn’t report the incident, and so no one else had the opportunity to learn from it. Not sharing an opportunity to learn is a cardinal sin.”

Cultures grow out of the keystone habits in every organization, whether leaders are aware of them or not. For instance, when researchers studied an incoming class of cadets at West Point, they measured their grade point averages, physical aptitude, military abilities, and self-discipline. When they correlated those factors with whether students dropped out or graduated, however, they found that all of them mattered less than a factor researchers referred to as “grit,” which they defined as the tendency to work “strenuously toward challenges, maintaining effort and interest over years despite failure, adversity, and plateaus in progress.” [126] [127]

What’s most interesting about grit is how it emerges. It grows out of a culture that cadets create for themselves, and that culture often emerges because of keystone habits they adopt at West Point. “There’s so much about this school that’s hard,” one cadet told me. “They call the first summer ‘Beast Barracks,’ because they want to grind you down. Tons of people quit before the school year starts.

“But I found this group of guys in the first couple of days here, and we started this thing where, every morning, we get together to make sure everyone is feeling strong. I go to them if I’m feeling worried or down, and I know they’ll pump me back up. There’s only nine of us, and we call ourselves the musketeers. Without them, I don’t think I would have lasted a month here.”

Cadets who are successful at West Point arrive at the school armed with habits of mental and physical discipline. Those assets, however, only carry you so far. To succeed, they need a keystone habit that creates a culture-such as a daily gathering of like-minded friends-to help find the strength to overcome obstacles. Keystone habits transform us by creating cultures that make clear the values that, in the heat of a difficult decision or a moment of uncertainty, we might otherwise forget.

In 2000, O’Neill retired from Alcoa, and at the request of the newly elected president George W. Bush, became secretary of the treasury.¹ He left that post two years later, and today spends most of his time teaching hospitals how to focus on worker safety and keystone habits that can lower medical error rates, as well as serving on various corporate boards.

Companies and organizations across America, in the meantime, have embraced the idea of using keystone habits to remake workplaces. At IBM, for instance, Lou Gerstner rebuilt the firm by initially concentrating on one keystone habit: IBM’s research and selling routines. At the consulting firm McKinsey & Company, a culture of continuous improvement is created through a keystone habit of wide-ranging internal critiques that are at the core of every assignment. Within Goldman Sachs, a keystone habit of risk assessment undergirds every decision.

And at Alcoa, O’Neill’s legacy lives on. Even in his absence, the injury rate has continued to decline. In 2010, 82 percent of Alcoa locations didn’t lose one employee day due to injury, close to an all-time high. On average, workers are more likely to get injured at a software company, animating cartoons for movie studios, or doing taxes as an accountant than handling molten aluminum at Alcoa.

“When I was made a plant manager,” said Jeff Shockey, the Alcoa executive, “the first day I pulled into the parking lot I saw all these parking spaces near the front doors with people’s titles on them. The head guy for this or that. People who were important got the best parking spots. The first thing I did was tell a maintenance manager to paint over all the titles. I wanted whoever got to work earliest to get the best spot. Everyone understood the message: Every person matters. It was an extension of what Paul was doing around worker safety. It electrified the plant. Pretty soon, everyone was getting to work earlier each day.”

¹ O’Neill’s tenure at Treasury was not as successful as his career at Alcoa. Almost immediately after taking office he began focusing on a couple of key issues, including worker safety, job creation, executive accountability, and fighting African poverty, among other initiatives.

However, O’Neill’s politics did not line up with those of President Bush, and he launched an internal fight opposing Bush’s proposed tax cuts. He was asked to resign at the end of 2002. “What I thought was the right thing for economic policy was the opposite of what the White House wanted,” O’Neill told me. “That’s not good for a treasury secretary, so I got fired.”

I.

The first time Travis Leach saw his father overdose, he was nine years old. His family had just moved into a small apartment at the end of an alleyway, the latest in a seemingly endless series of relocations that had most recently caused them to abandon their previous home in the middle of the night, throwing everything they owned into black garbage bags after receiving an eviction notice. Too many people coming and going too late at night, the landlord said. Too much noise.

Sometimes, at his old house, Travis would come home from school and find the rooms neatly cleaned, leftovers meticulously wrapped in the fridge and packets of hot sauce and ketchup in Tupperware containers. He knew this meant his parents had temporarily abandoned heroin for crank and spent the day in a cleaning frenzy. Those usually ended badly. Travis felt safer when the house was messy and his parents were on the couch, their eyes half-lidded, watching cartoons. There is no chaos at the end of a heroin fog.

Travis’s father was a gentle man who loved to cook and, except for a stint in the navy, spent his entire life within a few miles of his parents in Lodi, California. Travis’s mother, by the time everyone moved into the alleyway apartment, was in prison for heroin possession and prostitution. His parents were, essentially, functional addicts and the family maintained a veneer of normalcy. They went camping every summer and on most Friday nights attended his sister and brother’s softball games. When Travis was four years old, he went to Disneyland with his dad and was photographed for the first time in his life, by a Disney employee. The family camera had been sold to a pawn shop years before.

On the morning of the overdose, Travis and his brother were playing in the living room on top of blankets they laid out on the floor each night for sleeping. Travis’s father was getting ready to make pancakes when he stepped into the bathroom. He was carrying the tube sock that contained his needle, spoon, lighter, and cotton swabs. A few moments later, he came out, opened the refrigerator to get the eggs, and crashed to the floor. When the kids ran around the corner, their father was convulsing, his face turning blue.

Travis’s siblings had seen an overdose before and knew the drill. His brother rolled him onto his side. His sister opened his mouth to make sure he wouldn’t choke on his tongue, and told Travis to run next door, ask to use the neighbor’s phone, and dial 911.

“My name is Travis, my dad is passed out, and we don’t know what happened. He’s not breathing,” Travis lied to the police operator. Even at nine years old, he knew why his father was unconscious. He didn’t want to say it in front of the neighbor. Three years earlier, one of his dad’s friends had died in their basement after shooting up. When the paramedics had taken the body away, neighbors gawked at Travis and his sister while they held the door open for the gurney. One of the neighbors had a cousin whose son was in his class, and soon everyone in school had known.

After hanging up the phone, Travis walked to the end of the alleyway and waited for the ambulance. His father was treated at the hospital that morning, charged at the police station in the afternoon, and home again by dinnertime. He made spaghetti. Travis turned ten a few weeks later.

When Travis was sixteen, he dropped out of high school. “I was tired of being called a faggot,” he said, “tired of people following me home and throwing things at me. Everything seemed really overwhelming. It was easier to quit and go somewhere else.” He moved two hours south, to Fresno, and got a job at a car wash. He was fired for insubordination. He got jobs at McDonald’s and Hollywood Video, but when customers were rude-“I wanted ranch dressing, you moron!”-he would lose control.

“Get out of my drive-through!” he shouted at one woman, throwing the chicken nuggets at her car before his manager pulled him inside.

Sometimes he’d get so upset that he would start crying in the middle of a shift. He was often late, or he’d take a day off for no reason. In the morning, he would yell at his reflection in the mirror, order himself to be better, to suck it up. But he couldn’t get along with people, and he wasn’t strong enough to weather the steady drip of criticisms and indignities. When the line at his register would get too long and the manager would shout at him, Travis’s hands would start shaking and he’d feel like he couldn’t catch his breath. He wondered if this is what his parents felt like, so defenseless against life, when they started using drugs.

One day, a regular customer at Hollywood Video who’d gotten to know Travis a little bit suggested he think about working at Starbucks. “We’re opening a new store on Fort Washington, and I’m going to be an assistant manager,” the man said. “You should apply.” A month later, Travis was a barista on the morning shift.

That was six years ago. Today, at twenty-five, Travis is the manager of two Starbucks where he oversees forty employees and is responsible for revenues exceeding $2 million per year. His salary is $44,000 and he has a 401(k) and no debt. He’s never late to work. He does not get upset on the job. When one of his employees started crying after a customer screamed at her, Travis took her aside.

“Your apron is a shield,” he told her. “Nothing anyone says will ever hurt you. You will always be as strong as you want to be.”

He picked up that lecture in one of his Starbucks training courses, an education program that began on his first day and continues throughout an employee’s career. The program is sufficiently structured that he can earn college credits by completing the modules. The training has, Travis says, changed his life. Starbucks has taught him how to live, how to focus, how to get to work on time, and how to master his emotions. Most crucially, it has taught him willpower.

“Starbucks is the most important thing that has ever happened to me,” he told me. “I owe everything to this company.”

For Travis and thousands of others, Starbucks-like a handful of other companies-has succeeded in teaching the kind of life skills that schools, families, and communities have failed to provide. With more than 137,000 current employees and more than one million alumni, Starbucks is now, in a sense, one of the nation’s largest educators. All of those employees, in their first year alone, spent at least fifty hours in Starbucks classrooms, and dozens more at home with Starbucks’ workbooks and talking to the Starbucks mentors assigned to them.

At the core of that education is an intense focus on an all-important habit: willpower. Dozens of studies show that willpower is the single most important keystone habit for individual success. [128] In a 2005 study, for instance, researchers from the University of Pennsylvania analyzed 164 eighth-grade students, measuring their IQs and other factors, including how much willpower the students demonstrated, as measured by tests of their self-discipline.

Students who exerted high levels of willpower were more likely to earn higher grades in their classes and gain admission into more selective schools. They had fewer absences and spent less time watching television and more hours on homework. “Highly self-disciplined adolescents outperformed their more impulsive peers on every academic-performance variable,” the researchers wrote. “Self-discipline predicted academic performance more robustly than did IQ. Self-discipline also predicted which students would improve their grades over the course of the school year, whereas IQ did not… Self-discipline has a bigger effect on academic performance than does intellectual talent.” [129]

And the best way to strengthen willpower and give students a leg up, studies indicate, is to make it into a habit. “Sometimes it looks like people with great self-control aren’t working hard-but that’s because they’ve made it automatic,” Angela Duckworth, one of the University of Pennsylvania researchers told me. “Their willpower occurs without them having to think about it.”

For Starbucks, willpower is more than an academic curiosity. When the company began plotting its massive growth strategy in the late 1990s, executives recognized that success required cultivating an environment that justified paying four dollars for a fancy cup of coffee. The company needed to train its employees to deliver a bit of joy alongside lattes and scones. So early on, Starbucks started researching how they could teach employees to regulate their emotions and marshal their self-discipline to deliver a burst of pep with every serving. Unless baristas are trained to put aside their personal problems, the emotions of some employees will inevitably spill into how they treat customers. However, if a worker knows how to remain focused and disciplined, even at the end of an eight-hour shift, they’ll deliver the higher class of fast food service that Starbucks customers expect.

The company spent millions of dollars developing curriculums to train employees on self-discipline. Executives wrote workbooks that, in effect, serve as guides to how to make willpower a habit in workers’ lives. [130] Those curriculums are, in part, why Starbucks has grown from a sleepy Seattle company into a behemoth with more than seventeen thousand stores and revenues of more than $10 billion a year.

So how does Starbucks do it? How do they take people like Travis-the son of drug addicts and a high school dropout who couldn’t muster enough self-control to hold down a job at McDonald’s-and teach him to oversee dozens of employees and tens of thousands of dollars in revenue each month? What, precisely, did Travis learn?

II.

Everyone who walked into the room where the experiment was being conducted at Case Western Reserve University agreed on one thing: The cookies smelled delicious. They had just come out of the oven and were piled in a bowl, oozing with chocolate chips. On the table next to the cookies was a bowl of radishes. All day long, hungry students walked in, sat in front of the two foods, and submitted, unknowingly, to a test of their willpower that would upend our understanding of how self-discipline works.

At the time, there was relatively little academic scrutiny into willpower. Psychologists considered such subjects to be aspects of something they called “self-regulation,” but it wasn’t a field that inspired great curiosity. There was one famous experiment, conducted in the 1960s, in which scientists at Stanford had tested the willpower of a group of four-year-olds. The kids were brought into a room and presented with a selection of treats, including marshmallows. They were offered a deal: They could eat one marshmallow right away, or, if they waited a few minutes, they could have two marshmallows. Then the researcher left the room. Some kids gave in to temptation and ate the marshmallow as soon as the adult left. About 30 percent managed to ignore their urges, and doubled their treats when the researcher came back fifteen minutes later. Scientists, who were watching everything from behind a two-way mirror, kept careful track of which kids had enough self-control to earn the second marshmallow.

Years later, they tracked down many of the study’s participants. By now, they were in high school. The researchers asked about their grades and SAT scores, ability to maintain friendships, and their capacity to “cope with important problems.” They discovered that the four-year-olds who could delay gratification the longest ended up with the best grades and with SAT scores 210 points higher, on average, than everyone else. They were more popular and did fewer drugs. If you knew how to avoid the temptation of a marshmallow as a preschooler, it seemed, you also knew how to get yourself to class on time and finish your homework once you got older, as well as how to make friends and resist peer pressure. It was as if the marshmallow-ignoring kids had self-regulatory skills that gave them an advantage throughout their lives. [131]

Scientists began conducting related experiments, trying to figure out how to help kids increase their self-regulatory skills. They learned that teaching them simple tricks-such as distracting themselves by drawing a picture, or imagining a frame around the marshmallow, so it seemed more like a photo and less like a real temptation-helped them learn self-control. By the 1980s, a theory emerged that became generally accepted: Willpower is a learnable skill, something that can be taught the same way kids learn to do math and say “thank you.” But funding for these inquiries was scarce. The topic of willpower wasn’t in vogue. Many of the Stanford scientists moved on to other areas of research.

WHEN KIDS LEARN HABITS FOR DELAYING THEIR CRAVINGS…

THOSE HABITS SPILL OVER TO OTHER PARTS OF LIFE

However, when a group of psychology PhD candidates at Case Western-including one named Mark Muraven-discovered those studies in the mid-nineties, they started asking questions the previous research didn’t seem to answer. To Muraven, this model of willpower-as-skill wasn’t a satisfying explanation. A skill, after all, is something that remains constant from day to day. If you have the skill to make an omelet on Wednesday, you’ll still know how to make it on Friday.

In Muraven’s experience, though, it felt like he forgot how to exert willpower all the time. Some evenings he would come home from work and have no problem going for a jog. Other days, he couldn’t do anything besides lie on the couch and watch television. It was as if his brain-or, at least, that part of his brain responsible for making him exercise-had forgotten how to summon the willpower to push him out the door. Some days, he ate healthily. Other days, when he was tired, he raided the vending machines and stuffed himself with candy and chips.

If willpower is a skill, Muraven wondered, then why doesn’t it remain constant from day to day? He suspected there was more to willpower than the earlier experiments had revealed. But how do you test that in a laboratory?

Muraven’s solution was the lab containing one bowl of freshly baked cookies and one bowl of radishes. The room was essentially a closet with a two-way mirror, outfitted with a table, a wooden chair, a hand bell, and a toaster oven. Sixty-seven undergraduates were recruited and told to skip a meal. One by one, the undergrads sat in front of the two bowls.

“The point of this experiment is to test taste perceptions,” a researcher told each student, which was untrue. The point was to force students-but only some students-to exert their willpower. To that end, half the undergraduates were instructed to eat the cookies and ignore the radishes; the other half were told to eat the radishes and ignore the cookies. Muraven’s theory was that ignoring cookies is hard-it takes willpower. Ignoring radishes, on the other hand, hardly requires any effort at all.

“Remember,” the researcher said, “eat only the food that has been assigned to you.” Then she left the room.

Once the students were alone, they started munching. The cookie eaters were in heaven. The radish eaters were in agony. They were miserable forcing themselves to ignore the warm cookies. Through the two-way mirror, the researchers watched one of the radish eaters pick up a cookie, smell it longingly, and then put it back in the bowl. Another grabbed a few cookies, put them down, and then licked melted chocolate off his fingers.

After five minutes, the researcher reentered the room. By Muraven’s estimation, the radish eaters’ willpower had been thoroughly taxed by eating the bitter vegetable and ignoring the treats; the cookie eaters had hardly used any of their self-discipline.

“We need to wait about fifteen minutes for the sensory memory of the food you ate to fade,” the researcher told each participant. To pass the time, she asked them to complete a puzzle. It looked fairly simple: trace a geometric pattern without lifting your pencil from the page or going over the same line twice. If you want to quit, the researcher said, ring the bell. She implied the puzzle wouldn’t take long.

In truth, the puzzle was impossible to solve.

This puzzle wasn’t a way to pass time; it was the most important part of the experiment. It took enormous willpower to keep working on the puzzle, particularly when each attempt failed. The scientists wondered, would the students who had already expended their willpower by ignoring the cookies give up on the puzzle faster? In other words, was willpower a finite resource?

From behind their two-way mirror, the researchers watched. The cookie eaters, with their unused reservoirs of self-discipline, started working on the puzzle. In general, they looked relaxed. One of them tried a straightforward approach, hit a roadblock, and then started again. And again. And again. Some worked for over half an hour before the researcher told them to stop. On average, the cookie eaters spent almost nineteen minutes apiece trying to solve the puzzle before they rang the bell.

The radish eaters, with their depleted willpower, acted completely different. They muttered as they worked. They got frustrated. One complained that the whole experiment was a waste of time. Some of them put their heads on the table and closed their eyes. One snapped at the researcher when she came back in. On average, the radish eaters worked for only about eight minutes, 60 percent less time than the cookie eaters, before quitting. When the researcher asked afterward how they felt, one of the radish eaters said he was “sick of this dumb experiment.”

“By making people use a little bit of their willpower to ignore cookies, we had put them into a state where they were willing to quit much faster,” Muraven told me. “There’s been more than two hundred studies on this idea since then, and they’ve all found the same thing. Willpower isn’t just a skill. It’s a muscle, like the muscles in your arms or legs, and it gets tired as it works harder, so there’s less power left over for other things.”

Researchers have built on this finding to explain all sorts of phenomena. Some have suggested it helps clarify why otherwise successful people succumb to extramarital affairs (which are most likely to start late at night after a long day of using willpower at work) or why good physicians make dumb mistakes (which most often occur after a doctor has finished a long, complicated task that requires intense focus). [132] “If you want to do something that requires willpower-like going for a run after work-you have to conserve your willpower muscle during the day,” Muraven told me. “If you use it up too early on tedious tasks like writing emails or filling out complicated and boring expense forms, all the strength will be gone by the time you get home.” [133]

But how far does this analogy extend? Will exercising willpower muscles make them stronger the same way using dumbbells strengthen biceps?

In 2006, two Australian researchers-Megan Oaten and Ken Cheng-tried to answer that question by creating a willpower workout. They enrolled two dozen people between the ages of eighteen and fifty in a physical exercise program and, over two months, put them through an increasing number of weight lifting, resistance training, and aerobic routines. [134] Week after week, people forced themselves to exercise more frequently, using more and more willpower each time they hit the gym.

After two months, the researchers scrutinized the rest of the participants’ lives to see if increased willpower at the gym resulted in greater willpower at home. Before the experiment began, most of the subjects were self-professed couch potatoes. Now, of course, they were in better physical shape. But they were also healthier in other parts of their lives, as well. The more time they spent at the gym, the fewer cigarettes they smoked and the less alcohol, caffeine, and junk food they consumed. They were spending more hours on homework and fewer watching TV. They were less depressed.

Maybe, Oaten and Cheng wondered, those results had nothing to do with willpower. What if exercise just makes people happier and less hungry for fast food?

So they designed another experiment. [135] This time, they signed up twenty-nine people for a four-month money management program. They set savings goals and asked participants to deny themselves luxuries, such as meals at restaurants or movies. Participants were asked to keep detailed logs of everything they bought, which was annoying at first, but eventually people worked up the self-discipline to jot down every purchase.

People’s finances improved as they progressed through the program. More surprising, they also smoked fewer cigarettes and drank less alcohol and caffeine-on average, two fewer cups of coffee, two fewer beers, and, among smokers, fifteen fewer cigarettes each day. [136] They ate less junk food and were more productive at work and school. It was like the exercise study: As people strengthened their willpower muscles in one part of their lives-in the gym, or a money management program-that strength spilled over into what they ate or how hard they worked. Once willpower became stronger, it touched everything.

Oaten and Cheng did one more experiment. They enrolled forty-five students in an academic improvement program that focused on creating study habits. [137] Predictably, participants’ learning skills improved. And the students also smoked less, drank less, watched less television, exercised more, and ate healthier, even though all those things were never mentioned in the academic program. Again, as their willpower muscles strengthened, good habits seemed to spill over into other parts of their lives.

“When you learn to force yourself to go to the gym or start your homework or eat a salad instead of a hamburger, part of what’s happening is that you’re changing how you think,” said Todd Heatherton, a researcher at Dartmouth who has worked on willpower studies. [138] “People get better at regulating their impulses. They learn how to distract themselves from temptations. And once you’ve gotten into that willpower groove, your brain is practiced at helping you focus on a goal.”

There are now hundreds of researchers, at nearly every major university, studying willpower. Public and charter schools in Philadelphia, Seattle, New York, and elsewhere have started incorporating willpower-strengthening lessons into curriculums. At KIPP, or the “Knowledge Is Power Program”-a collection of charter schools serving low-income students across the nation-teaching self-control is part of the schools’ philosophy. (A KIPP school in Philadelphia gave students shirts proclaiming “Don’t Eat the Marshmallow.”) Many of these schools have dramatically raised students’ test scores. [139]

“That’s why signing kids up for piano lessons or sports is so important. It has nothing to do with creating a good musician or a five-year-old soccer star,” said Heatherton. “When you learn to force yourself to practice for an hour or run fifteen laps, you start building self-regulatory strength. A five-year-old who can follow the ball for ten minutes becomes a sixth grader who can start his homework on time.” [140]

As research on willpower has become a hot topic in scientific journals and newspaper articles, it has started to trickle into corporate America. Firms such as Starbucks-and the Gap, Walmart, restaurants, or any other business that relies on entry-level workers-all face a common problem: No matter how much their employees want to do a great job, many will fail because they lack self-discipline. They show up late. They snap at rude customers. They get distracted or drawn into workplace dramas. They quit for no reason.

“For a lot of employees, Starbucks is their first professional experience,” said Christine Deputy, who helped oversee the company’s training programs for more than a decade. “If your parents or teachers have been telling you what to do your entire life, and suddenly customers are yelling and your boss is too busy to give you guidance, it can be really overwhelming. A lot of people can’t make the transition. So we try to figure out how to give our employees the self-discipline they didn’t learn in high school.”

But when companies like Starbucks tried to apply the willpower lessons from the radish-and-cookie studies to the workplace, they encountered difficulties. They sponsored weight-loss classes and offered employees free gym memberships, hoping the benefits would spill over to how they served coffee. [141] Attendance was spotty. It was hard to sit through a class or hit the gym after a full day at work, employees complained. “If someone has trouble with self-discipline at work, they’re probably also going to have trouble attending a program designed to strengthen their self-discipline after work,” Muraven said.

But Starbucks was determined to solve this problem. By 2007, during the height of its expansion, the company was opening seven new stores every day and hiring as many as fifteen hundred employees each week. [142] Training them to excel at customer service-to show up on time and not get angry at patrons and serve everyone with a smile while remembering customers’ orders and, if possible, their names-was essential. People expect an expensive latte delivered with a bit of sparkle. “We’re not in the coffee business serving people,” Howard Behar, the former president of Starbucks, told me. “We’re in the people business serving coffee. Our entire business model is based on fantastic customer service. Without that, we’re toast.”

The solution, Starbucks discovered, was turning self-discipline into an organizational habit.

III.

In 1992, a British psychologist walked into two of Scotland’s busiest orthopedic hospitals and recruited five-dozen patients for an experiment she hoped would explain how to boost the willpower of people exceptionally resistant to change. [143]

The patients, on average, were sixty-eight years old. Most of them earned less than $10,000 a year and didn’t have more than a high school degree. All of them had recently undergone hip or knee replacement surgeries, but because they were relatively poor and uneducated, many had waited years for their operations. They were retirees, elderly mechanics, and store clerks. They were in life’s final chapters, and most had no desire to pick up a new book.

Recovering from a hip or knee surgery is incredibly arduous. The operation involves severing joint muscles and sawing through bones. While recovering, the smallest movements-shifting in bed or flexing a joint-can be excruciating. However, it is essential that patients begin exercising almost as soon as they wake from surgery. They must begin moving their legs and hips before the muscles and skin have healed, or scar tissue will clog the joint, destroying its flexibility. In addition, if patients don’t start exercising, they risk developing blood clots. But the agony is so extreme that it’s not unusual for people to skip out on rehab sessions. Patients, particularly elderly ones, often refuse to comply with doctors’ orders.

The Scottish study’s participants were the types of people most likely to fail at rehabilitation. The scientist conducting the experiment wanted to see if it was possible to help them harness their willpower. She gave each patient a booklet after their surgeries that detailed their rehab schedule, and in the back were thirteen additional pages-one for each week-with blank spaces and instructions: “My goals for this week are ____________________? Write down exactly what you are going to do. For example, if you are going to go for a walk this week, write down where and when you are going to walk.” She asked patients to fill in each of those pages with specific plans. Then she compared the recoveries of those who wrote out goals with those of patients who had received the same booklets, but didn’t write anything.

It seems absurd to think that giving people a few pieces of blank paper might make a difference in how quickly they recover from surgery. But when the researcher visited the patients three months later, she found a striking difference between the two groups. The patients who had written plans in their booklets had started walking almost twice as fast as the ones who had not. They had started getting in and out of their chairs, unassisted, almost three times as fast. They were putting on their shoes, doing the laundry, and making themselves meals quicker than the patients who hadn’t scribbled out goals ahead of time.

The psychologist wanted to understand why. She examined the booklets, and discovered that most of the blank pages had been filled in with specific, detailed plans about the most mundane aspects of recovery. One patient, for example, had written, “I will walk to the bus stop tomorrow to meet my wife from work,” and then noted what time he would leave, the route he would walk, what he would wear, which coat he would bring if it was raining, and what pills he would take if the pain became too much. Another patient, in a similar study, wrote a series of very specific schedules regarding the exercises he would do each time he went to the bathroom. A third wrote a minute-by-minute itinerary for walking around the block.

As the psychologist scrutinized the booklets, she saw that many of the plans had something in common: They focused on how patients would handle a specific moment of anticipated pain. The man who exercised on the way to the bathroom, for instance, knew that each time he stood up from the couch, the ache was excruciating. So he wrote out a plan for dealing with it: Automatically take the first step, right away, so he wouldn’t be tempted to sit down again. The patient who met his wife at the bus stop dreaded the afternoons, because that stroll was the longest and most painful each day. So he detailed every obstacle he might confront, and came up with a solution ahead of time.

Put another way, the patients’ plans were built around inflection points when they knew their pain-and thus the temptation to quit-would be strongest. The patients were telling themselves how they were going to make it over the hump.

Each of them, intuitively, employed the same rules that Claude Hopkins had used to sell Pepsodent. They identified simple cues and obvious rewards. The man who met his wife at the bus stop, for instance, identified an easy cue-It’s 3:30, she’s on her way home!-and he clearly defined his reward-Honey, I’m here! When the temptation to give up halfway through the walk appeared, the patient could ignore it because he had crafted self-discipline into a habit.

PATIENTS DESIGNED WILLPOWER HABITS TO HELP THEM OVERCOME PAINFUL INFLECTION POINTS

There’s no reason why the other patients-the ones who didn’t write out recovery plans-couldn’t have behaved the same way. All the patients had been exposed to the same admonitions and warnings at the hospital. They all knew exercise was essential for their recovery. They all spent weeks in rehab.

But the patients who didn’t write out any plans were at a significant disadvantage, because they never thought ahead about how to deal with painful inflection points. They never deliberately designed willpower habits. Even if they intended to walk around the block, their resolve abandoned them when they confronted the agony of the first few steps.

When Starbucks’s attempts at boosting workers’ willpower through gym memberships and diet workshops faltered, executives decided they needed to take a new approach. They started by looking more closely at what was actually happening inside their stores. They saw that, like the Scottish patients, their workers were failing when they ran up against inflection points. What they needed were institutional habits that made it easier to muster their self-discipline.

Executives determined that, in some ways, they had been thinking about willpower all wrong. Employees with willpower lapses, it turned out, had no difficulty doing their jobs most of the time. On the average day, a willpower-challenged worker was no different from anyone else. But sometimes, particularly when faced with unexpected stresses or uncertainties, those employees would snap and their self-control would evaporate. A customer might begin yelling, for instance, and a normally calm employee would lose her composure. An impatient crowd might overwhelm a barista, and suddenly he was on the edge of tears. [144]

What employees really needed were clear instructions about how to deal with inflection points-something similar to the Scottish patients’ booklets: a routine for employees to follow when their willpower muscles went limp. [145] So the company developed new training materials that spelled out routines for employees to use when they hit rough patches. The manuals taught workers how to respond to specific cues, such as a screaming customer or a long line at a cash register. Managers drilled employees, role-playing with them until the responses became automatic. The company identified specific rewards-a grateful customer, praise from a manager-that employees could look to as evidence of a job well done.

Starbucks taught their employees how to handle moments of adversity by giving them willpower habit loops.

When Travis started at Starbucks, for instance, his manager introduced him to the habits right away. “One of the hardest things about this job is dealing with an angry customer,” Travis’s manager told him. “When someone comes up and starts yelling at you because they got the wrong drink, what’s your first reaction?”

“I don’t know,” Travis said. “I guess I feel kind of scared. Or angry.”

“That’s natural,” his manager said. “But our job is to provide the best customer service, even when the pressure’s on.” The manager flipped open the Starbucks manual, and showed Travis a page that was largely blank. At the top, it read, “When a customer is unhappy, my plan is to… ”

“This workbook is for you to imagine unpleasant situations, and write out a plan for responding,” the manager said. “One of the systems we use is called the LATTE method. We Listen to the customer, Acknowledge their complaint, Take action by solving the problem, Thank them, and then Explain why the problem occurred. [146]

THE LATTE HABIT LOOP

“Why don’t you take a few minutes, and write out a plan for dealing with an angry customer. Use the LATTE method. Then we can role-play a little bit.”

Starbucks has dozens of routines that employees are taught to use during stressful inflection points. There’s the What What Why system of giving criticism and the Connect, Discover, and Respond system for taking orders when things become hectic. There are learned habits to help baristas tell the difference between patrons who just want their coffee (“A hurried customer speaks with a sense of urgency and may seem impatient or look at their watch”) and those who need a bit more coddling (“A regular customer knows other baristas by name and normally orders the same beverage each day”). Throughout the training manuals are dozens of blank pages where employees can write out plans that anticipate how they will surmount inflection points. Then they practice those plans, again and again, until they become automatic. [147]

This is how willpower becomes a habit: by choosing a certain behavior ahead of time, and then following that routine when an inflection point arrives. When the Scottish patients filled out their booklets, or Travis studied the LATTE method, they decided ahead of time how to react to a cue-a painful muscle or an angry customer. When the cue arrived, the routine occurred.

Starbucks isn’t the only company to use such training methods. For instance, at Deloitte Consulting, the largest tax and financial services company in the world, employees are trained in a curriculum named “Moments That Matter,” which focuses on dealing with inflection points such as when a client complains about fees, when a colleague is fired, or when a Deloitte consultant has made a mistake. For each of those moments, there are preprogrammed routines-Get Curious, Say What No One Else Will, Apply the 5/5/5 Rule-that guide employees in how they should respond. At the Container Store, employees receive more than 185 hours of training in their first year alone. They are taught to recognize inflection points such as an angry coworker or an overwhelmed customer, and habits, such as routines for calming shoppers or defusing a confrontation. When a customer comes in who seems overwhelmed, for example, an employee immediately asks them to visualize the space in their home they are hoping to organize, and describe how they’ll feel when everything is in its place. “We’ve had customers come up to us and say, ‘This is better than a visit to my shrink,’ ” the company’s CEO told a reporter. [148]

IV.

Howard Schultz, the man who built Starbucks into a colossus, isn’t so different from Travis in some ways. [149] He grew up in a public housing project in Brooklyn, sharing a two-bedroom apartment with his parents and two siblings. When he was seven years old, Schultz’s father broke his ankle and lost his job driving a diaper truck. That was all it took to throw the family into crisis. His father, after his ankle healed, began cycling through a series of lower-paying jobs. “My dad never found his way,” Schultz told me. “I saw his self-esteem get battered. I felt like there was so much more he could have accomplished.”

Schultz’s school was a wild, overcrowded place with asphalt playgrounds and kids playing football, basketball, softball, punch ball, slap ball, and any other game they could devise. If your team lost, it could take an hour to get another turn. So Schultz made sure his team always won, no matter the cost. He would come home with bloody scrapes on his elbows and knees, which his mother would gently rinse with a wet cloth. “You don’t quit,” she told him.

His competitiveness earned him a college football scholarship (he broke his jaw and never played a game), a communications degree, and eventually a job as a Xerox salesman in New York City. He’d wake up every morning, go to a new midtown office building, take the elevator to the top floor, and go door-to-door, politely inquiring if anyone was interested in toner or copy machines. Then he’d ride the elevator down one floor and start all over again.

By the early 1980s, Schultz was working for a plastics manufacturer when he noticed that a little-known retailer in Seattle was ordering an inordinate number of coffee drip cones. Schultz flew out and fell in love with the company. Two years later, when he heard that Starbucks, then just six stores, was for sale, he asked everyone he knew for money and bought it.

That was 1987. Within three years, there were eighty-four stores; within six years, more than a thousand. Today, there are seventeen thousand stores in more than fifty countries.

Why did Schultz turn out so different from all the other kids on that playground? Some of his old classmates are today cops and firemen in Brooklyn. Others are in prison. Schultz is worth more than $1 billion. He’s been heralded as one of the greatest CEOs of the twentieth century. Where did he find the determination-the willpower-to climb from a housing project to a private jet?

“I don’t really know,” he told me. “My mom always said, ‘You’re going to be the first person to go to college, you’re going to be a professional, you’re going to make us all proud.’ She would ask these little questions, ‘How are you going to study tonight? What are you going to do tomorrow? How do you know you’re ready for your test?’ It trained me to set goals.

“I’ve been really lucky,” he said. “And I really, genuinely believe that if you tell people that they have what it takes to succeed, they’ll prove you right.”

Schultz’s focus on employee training and customer service made Starbucks into one of the most successful companies in the world. For years, he was personally involved in almost every aspect of how the company was run. In 2000, exhausted, he handed over day-to-day operations to other executives, at which point, Starbucks began to stumble. Within a few years, customers were complaining about the quality of the drinks and customer service. Executives, focused on a frantic expansion, often ignored the complaints. Employees grew unhappy. Surveys indicated people were starting to equate Starbucks with tepid coffee and empty smiles.

So Schultz stepped back into the chief executive position in 2008. Among his priorities was restructuring the company’s training program to renew its focus on a variety of issues, including bolstering employees’-or “partners,” in Starbucks’ lingo-willpower and self-confidence. “We had to start earning customer and partner trust again,” Schultz told me.

At about the same time, a new wave of studies was appearing that looked at the science of willpower in a slightly different way. Researchers had noticed that some people, like Travis, were able to create willpower habits relatively easily. Others, however, struggled, no matter how much training and support they received. What was causing the difference?

Mark Muraven, who was by then a professor at the University of Albany, set up a new experiment. [150] He put undergraduates in a room that contained a plate of warm, fresh cookies and asked them to ignore the treats. Half the participants were treated kindly. “We ask that you please don’t eat the cookies. Is that okay?” a researcher said. She then discussed the purpose of the experiment, explaining that it was to measure their ability to resist temptations. She thanked them for contributing their time. “If you have any suggestions or thoughts about how we can improve this experiment, please let me know. We want you to help us make this experience as good as possible.”

The other half of the participants weren’t coddled the same way. They were simply given orders.

“You must not eat the cookies,” the researcher told them. She didn’t explain the experiment’s goals, compliment them, or show any interest in their feedback. She told them to follow the instructions. “We’ll start now,” she said.

The students from both groups had to ignore the warm cookies for five minutes after the researcher left the room. None gave in to temptation.

Then the researcher returned. She asked each student to look at a computer monitor. It was programmed to flash numbers on the screen, one at a time, for five hundred milliseconds apiece. The participants were asked to hit the space bar every time they saw a “6” followed by a “4.” This has become a standard way to measure willpower-paying attention to a boring sequence of flashing numbers requires a focus akin to working on an impossible puzzle.

Students who had been treated kindly did well on the computer test. Whenever a “6” flashed and a “4” followed, they pounced on the space bar. They were able to maintain their focus for the entire twelve minutes. Despite ignoring the cookies, they had willpower to spare.

Students who had been treated rudely, on the other hand, did terribly. They kept forgetting to hit the space bar. They said they were tired and couldn’t focus. Their willpower muscle, researchers determined, had been fatigued by the brusque instructions.

When Muraven started exploring why students who had been treated kindly had more willpower he found that the key difference was the sense of control they had over their experience. “We’ve found this again and again,” Muraven told me. “When people are asked to do something that takes self-control, if they think they are doing it for personal reasons-if they feel like it’s a choice or something they enjoy because it helps someone else-it’s much less taxing. If they feel like they have no autonomy, if they’re just following orders, their willpower muscles get tired much faster. In both cases, people ignored the cookies. But when the students were treated like cogs, rather than people, it took a lot more willpower.”

For companies and organizations, this insight has enormous implications. Simply giving employees a sense of agency-a feeling that they are in control, that they have genuine decision-making authority-can radically increase how much energy and focus they bring to their jobs. One 2010 study at a manufacturing plant in Ohio, for instance, scrutinized assembly-line workers who were empowered to make small decisions about their schedules and work environment. [151] They designed their own uniforms and had authority over shifts. Nothing else changed. All the manufacturing processes and pay scales stayed the same. Within two months, productivity at the plant increased by 20 percent. Workers were taking shorter breaks. They were making fewer mistakes. Giving employees a sense of control improved how much self-discipline they brought to their jobs.

The same lessons hold true at Starbucks. Today, the company is focused on giving employees a greater sense of authority. They have asked workers to redesign how espresso machines and cash registers are laid out, to decide for themselves how customers should be greeted and where merchandise should be displayed. It’s not unusual for a store manager to spend hours discussing with his employees where a blender should be located.

“We’ve started asking partners to use their intellect and creativity, rather than telling them ‘take the coffee out of the box, put the cup here, follow this rule,’ ” said Kris Engskov, a vice president at Starbucks. “People want to be in control of their lives.”

Turnover has gone down. Customer satisfaction is up. Since Schultz’s return, Starbucks has boosted revenues by more than $1.2 billion per year.

V.

When Travis was sixteen, before he dropped out of school and started working for Starbucks, his mother told him a story. They were driving together, and Travis asked why he didn’t have more siblings. His mother had always tried to be completely honest with her children, and so she told him that she had become pregnant two years before Travis was born but had gotten an abortion. They already had two children at that point, she explained, and were addicted to drugs. They didn’t think they could support another baby. Then, a year later, she became pregnant with Travis. She thought about having another abortion, but it was too much to bear. It was easier to let nature take its course. Travis was born.

“She told me that she had made a lot of mistakes, but that having me was one of the best things that ever happened to her,” Travis said. “When your parents are addicts, you grow up knowing you can’t always trust them for everything you need. But I’ve been really lucky to find bosses who gave me what was missing. If my mom had been as lucky as me, I think things would have turned out different for her.”

A few years after that conversation, Travis’s father called to say that an infection had entered his mother’s bloodstream through one of the places on her arm she used to shoot up. Travis immediately drove to the hospital in Lodi, but she was unconscious by the time he arrived. She died a half hour later, when they removed her life support.

A week later, Travis’s father was in the hospital with pneumonia. His lung had collapsed. Travis drove to Lodi again, but it was 8:02 P.M. when he got to the emergency room. A nurse brusquely told him he’d have to come back tomorrow; visiting hours were over.

Travis has thought a lot about that moment since then. He hadn’t started working at Starbucks yet. He hadn’t learned how to control his emotions. He didn’t have the habits that, since then, he’s spent years practicing. When he thinks about his life now, how far he is from a world where overdoses occur and stolen cars show up in driveways and a nurse seems like an insurmountable obstacle, he wonders how it’s possible to travel such a long distance in such a short time.

“If he had died a year later, everything would have been different,” Travis told me. By then, he would have known how to calmly plead with the nurse. He would have known to acknowledge her authority, and then ask politely for one small exception. He could have gotten inside the hospital. Instead, he gave up and walked away. “I said, ‘All I want to do is talk to him once,’ and she was like, ‘He’s not even awake, it’s after visiting hours, come back tomorrow.’ I didn’t know what to say. I felt so small.”

Travis’s father died that night.

On the anniversary of his death, every year, Travis wakes up early, takes an extra-long shower, plans out his day in careful detail, and then drives to work. He always arrives on time.

I.

The patient was already unconscious when he was wheeled into the operating room at Rhode Island Hospital. His jaw was slack, his eyes closed, and the top of an intubation tube peeked above his lips. As a nurse hooked him up to a machine that would force air into his lungs during surgery, one of his arms slipped off the gurney, the skin mottled with liver spots.

The man was eighty-six years old and, three days earlier, had fallen at home. Afterward, he had trouble staying awake and answering questions, and so eventually his wife called an ambulance. [152] In the emergency room, a doctor asked him what happened, but the man kept nodding off in the middle of his sentences. A scan of his head revealed why: The fall had slammed his brain against his skull, causing what’s known as a subdural hematoma. Blood was pooling within the left portion of his cranium, pushing against the delicate folds of tissue inside his skull. The fluid had been building for almost seventy-two hours, and those parts of the brain that controlled his breathing and heart were beginning to falter. Unless the blood was drained, the man would die. [153]

At the time, Rhode Island Hospital was one of the nation’s leading medical institutions, the main teaching hospital for Brown University and the only Level I trauma center in southeastern New England. Inside the tall brick and glass building, physicians had pioneered cutting-edge medical techniques, including the use of ultrasound waves to destroy tumors inside a patient’s body. In 2002, the National Coalition on Health Care rated the hospital’s intensive care unit as one of the finest in the country. [154]

But by the time the elderly patient arrived, Rhode Island Hospital also had another reputation: a place riven by internal tensions. There were deep, simmering enmities between nurses and physicians. In 2000, the nurses’ union had voted to strike after complaining that they were being forced to work dangerously long hours. More than three hundred of them stood outside the hospital with signs reading “Stop Slavery” and “They can’t take away our pride.” [155]

“This place can be awful,” one nurse recalled telling a reporter. “The doctors can make you feel like you’re worthless, like you’re disposable. Like you should be thankful to pick up after them.”

Administrators eventually agreed to limit nurses’ mandatory overtime, but tensions continued to rise. [156] A few years later, a surgeon was preparing for a routine abdominal operation when a nurse called for a “time-out.” Such pauses are standard procedure at most hospitals, a way for doctors and staff to make sure mistakes are avoided. [157] The nursing staff at Rhode Island Hospital was insistent on time-outs, particularly since a surgeon had accidentally removed the tonsils of a girl who was supposed to have eye surgery. Time-outs were supposed to catch such errors before they occurred.

At the abdominal surgery, when the OR nurse asked the team to gather around the patient for a time-out and to discuss their plan, the doctor headed for the doors.

“Why don’t you lead this?” the surgeon told the nurse. “I’m going to step outside for a call. Knock when you’re ready.”

“You’re supposed to be here for this, Doctor,” she replied.

“You can handle it,” the surgeon said, as he walked toward the door.

“Doctor, I don’t feel this is appropriate.”

The doctor stopped and looked at her. “If I want your damn opinion, I’ll ask for it,” he said. “Don’t ever question my authority again. If you can’t do your job, get the hell out of my OR.”

The nurse led the time-out, retrieved the doctor a few minutes later, and the procedure occurred without complication. She never contradicted a physician again, and never said anything when other safety policies were ignored.

“Some doctors were fine, and some were monsters,” one nurse who worked at Rhode Island Hospital in the mid-2000s told me. “We called it the glass factory, because it felt like everything could crash down at any minute.”

To deal with these tensions, the staff had developed informal rules-habits unique to the institution-that helped avert the most obvious conflicts. Nurses, for instance, always double-checked the orders of error-prone physicians and quietly made sure that correct doses were entered; they took extra time to write clearly on patients’ charts, lest a hasty surgeon make the wrong cut. One nurse told me they developed a system of color codes to warn one another. “We put doctors’ names in different colors on the whiteboards,” she said. “Blue meant ‘nice,’ red meant ‘jerk,’ and black meant, ‘whatever you do, don’t contradict them or they’ll take your head off.’ ”

Rhode Island Hospital was a place filled with a corrosive culture. Unlike at Alcoa, where carefully designed keystone habits surrounding worker safety had created larger and larger successes, inside Rhode Island Hospital, habits emerged on the fly among nurses seeking to offset physician arrogance. The hospital’s routines weren’t carefully thought out. Rather, they appeared by accident and spread through whispered warnings, until toxic patterns emerged. This can happen within any organization where habits aren’t deliberately planned. Just as choosing the right keystone habits can create amazing change, the wrong ones can create disasters.

And when the habits within Rhode Island Hospital imploded, they caused terrible mistakes.

When the emergency room staff saw the brain scans of the eighty-six-year-old man with the subdural hematoma, they immediately paged the neurosurgeon on duty. He was in the middle of a routine spinal surgery, but when he got the page, he stepped away from the operating table and looked at images of the elderly man’s head on a computer screen. The surgeon told his assistant-a nurse practitioner-to go to the emergency room and get the man’s wife to sign a consent form approving surgery. He finished his spinal procedure. A half hour later, the elderly man was wheeled into the same operating theater. [158]

Nurses were rushing around. The unconscious elderly man was placed on the table. A nurse picked up his consent form and medical chart.

“Doctor,” the nurse said, looking at the patient’s chart. “The consent form doesn’t say where the hematoma is.” The nurse leafed through the paperwork. There was no clear indication of which side of his head they were supposed to operate on. [159]

Every hospital relies upon paperwork to guide surgeries. Before any cut is made, a patient or family member is supposed to sign a document approving each procedure and verifying the details. In a chaotic environment, where as many as a dozen doctors and nurses may handle a patient between the ER and the recovery suite, consent forms are the instructions that keep track of what is supposed to occur. No one is supposed to go into surgery without a signed and detailed consent.

“I saw the scans before,” the surgeon said. “It was the right side of the head. If we don’t do this quickly, he’s gonna die.”

“Maybe we should pull up the films again,” the nurse said, moving toward a computer terminal. For security reasons, the hospital’s computers locked after fifteen minutes of idling. It would take at least a minute for the nurse to log in and load the patient’s brain scans onto the screen.

“We don’t have time,” the surgeon said. “They told me he’s crashing. We’ve got to relieve the pressure.”

“What if we find the family?” the nurse asked.

“If that’s what you want, then call the fucking ER and find the family! In the meantime, I’m going to save his life.” The surgeon grabbed the paperwork, scribbled “right” on the consent form, and initialed it.

“There,” he said. “We have to operate immediately.” [160]

The nurse had worked at Rhode Island Hospital for a year. He understood the hospital’s culture. This surgeon’s name, the nurse knew, was often scribbled in black on the large whiteboard in the hallway, signaling that nurses should beware. The unwritten rules in this scenario were clear: The surgeon always wins.

The nurse put down the chart and stood aside as the doctor positioned the elderly man’s head in a cradle that provided access to the right side of his skull and shaved and applied antiseptic to his head. The plan was to open the skull and suction out the blood pooling on top of his brain. The surgeon sliced away a flap of scalp, exposed the skull, and put a drill against the white bone. He began pushing until the bit broke through with a soft pop. He made two more holes and used a saw to cut out a triangular piece of the man’s skull. Underneath was the dura, the translucent sheath surrounding the brain.

“Oh my God,” someone said.

There was no hematoma. They were operating on the wrong side of the head.

“We need him turned!” the surgeon yelled. [161]

The triangle of bone was replaced and reattached with small metal plates and screws, and the patient’s scalp sewed up. His head was shifted to the other side and then, once again, shaved, cleansed, cut, and drilled until a triangle of skull could be removed. This time, the hematoma was immediately visible, a dark bulge that spilled like thick syrup when the dura was pierced. The surgeon vacuumed the blood and the pressure inside the old man’s skull fell immediately. The surgery, which should have taken about an hour, had run almost twice as long.

Afterward, the patient was taken to the intensive care unit, but he never regained full consciousness. Two weeks later, he died.

A subsequent investigation said it was impossible to determine the precise cause of death, but the patient’s family argued that the trauma of the medical error had overwhelmed his already fragile body, that the stress of removing two pieces of skull, the additional time in surgery, and the delay in evacuating the hematoma had pushed him over the edge. If not for the mistake, they claimed, he might still be alive. The hospital paid a settlement and the surgeon was barred from ever working at Rhode Island Hospital again. [162]

Such an accident, some nurses later claimed, was inevitable. Rhode Island Hospital’s institutional habits were so dysfunctional, it was only a matter of time until a grievous mistake occurred.¹ It’s not just hospitals that breed dangerous patterns, of course. Destructive organizational habits can be found within hundreds of industries and at thousands of firms. And almost always, they are the products of thoughtlessness, of leaders who avoid thinking about the culture and so let it develop without guidance. There are no organizations without institutional habits. There are only places where they are deliberately designed, and places where they are created without forethought, so they often grow from rivalries or fear.

But sometimes, even destructive habits can be transformed by leaders who know how to seize the right opportunities. Sometimes, in the heat of a crisis, the right habits emerge.

II.

When An Evolutionary Theory of Economic Change was first published in 1982, very few people outside of academia noticed. The book’s bland cover and daunting first sentence-“In this volume we develop an evolutionary theory of the capabilities and behavior of business firms operating in a market environment, and construct and analyze a number of models consistent with that theory”-almost seemed designed to ward off readers. [163] The authors, Yale professors Richard Nelson and Sidney Winter, were best known for a series of intensely analytic papers exploring Schumpeterian theory that even most PhD candidates didn’t pretend to understand. [164]

Within the world of business strategy and organizational theory, however, the book went off like a bombshell. [165] It was soon hailed as one of the most important texts of the century. Economics professors started talking about it to their colleagues at business schools, who started talking to CEOs at conferences, and soon executives were quoting Nelson and Winter inside corporations as different as General Electric, Pfizer, and Starwood Hotels.

Nelson and Winter had spent more than a decade examining how companies work, trudging through swamps of data before arriving at their central conclusion: “Much of firm behavior,” they wrote, is best “understood as a reflection of general habits and strategic orientations coming from the firm’s past,” rather than “the result of a detailed survey of the remote twigs of the decision tree.” [166]

Or, put in language that people use outside of theoretical economics, it may seem like most organizations make rational choices based on deliberate decision making, but that’s not really how companies operate at all. Instead, firms are guided by long-held organizational habits, patterns that often emerge from thousands of employees’ independent decisions. [167] And these habits have more profound impacts than anyone previously understood.

For instance, it might seem like the chief executive of a clothing company made the decision last year to feature a red cardigan on the catalog’s cover by carefully reviewing sales and marketing data. But, in fact, what really happened was that his vice president constantly trolls websites devoted to Japanese fashion trends (where red was hip last spring), and the firm’s marketers routinely ask their friends which colors are “in,” and the company’s executives, back from their annual trip to the Paris runway shows, reported hearing that designers at rival firms were using new magenta pigments. All these small inputs, the result of uncoordinated patterns among executives gossiping about competitors and talking to their friends, got mixed into the company’s more formal research and development routines until a consensus emerged: Red will be popular this year. No one made a solitary, deliberate decision. Rather, dozens of habits, processes, and behaviors converged until it seemed like red was the inevitable choice.

These organizational habits-or “routines,” as Nelson and Winter called them-are enormously important, because without them, most companies would never get any work done. [168] Routines provide the hundreds of unwritten rules that companies need to operate. [169] [170] They allow workers to experiment with new ideas without having to ask for permission at every step. They provide a kind of “organizational memory,” so that managers don’t have to reinvent the sales process every six months or panic each time a VP quits. [171] Routines reduce uncertainty-a study of recovery efforts after earthquakes in Mexico and Los Angeles, for instance, found that the habits of relief workers (which they carried from disaster to disaster, and which included things such as establishing communication networks by hiring children to carry messages between neighborhoods) were absolutely critical, “because without them, policy formulation and implementation would be lost in a jungle of detail.” [172]

But among the most important benefits of routines is that they create truces between potentially warring groups or individuals within an organization. [173]

Most economists are accustomed to treating companies as idyllic places where everyone is devoted to a common goal: making as much money as possible. Nelson and Winter pointed out that, in the real world, that’s not how things work at all. Companies aren’t big happy families where everyone plays together nicely. Rather, most workplaces are made up of fiefdoms where executives compete for power and credit, often in hidden skirmishes that make their own performances appear superior and their rivals’ seem worse. Divisions compete for resources and sabotage each other to steal glory. Bosses pit their subordinates against one another so that no one can mount a coup.

Companies aren’t families. They’re battlefields in a civil war.

Yet despite this capacity for internecine warfare, most companies roll along relatively peacefully, year after year, because they have routines-habits-that create truces that allow everyone to set aside their rivalries long enough to get a day’s work done.

Organizational habits offer a basic promise: If you follow the established patterns and abide by the truce, then rivalries won’t destroy the company, the profits will roll in, and, eventually, everyone will get rich. A salesperson, for example, knows she can boost her bonus by giving favored customers hefty discounts in exchange for larger orders. But she also knows that if every salesperson gives away hefty discounts, the firm will go bankrupt and there won’t be any bonuses to hand out. So a routine emerges: The salespeople all get together every January and agree to limit how many discounts they offer in order to protect the company’s profits, and at the end of the year everyone gets a raise.

Or take a young executive gunning for vice president who, with one quiet phone call to a major customer, could kill a sale and sabotage a colleague’s division, taking him out of the running for the promotion. The problem with sabotage is that even if it’s good for you, it’s usually bad for the firm. So at most companies, an unspoken compact emerges: It’s okay to be ambitious, but if you play too rough, your peers will unite against you. On the other hand, if you focus on boosting your own department, rather than undermining your rival, you’ll probably get taken care of over time. [174]

ROUTINES CREATE TRUCES THAT ALLOW WORK TO GET DONE

Routines and truces offer a type of rough organizational justice, and because of them, Nelson and Winter wrote, conflict within companies usually “follows largely predictable paths and stays within predictable bounds that are consistent with the ongoing routine… The usual amount of work gets done, reprimands and compliments are delivered with the usual frequency… Nobody is trying to steer the organizational ship into a sharp turn in the hope of throwing a rival overboard.” [175]

Most of the time, routines and truces work perfectly. Rivalries still exist, of course, but because of institutional habits, they’re kept within bounds and the business thrives.

However, sometimes even a truce proves insufficient. Sometimes, as Rhode Island Hospital discovered, an unstable peace can be as destructive as any civil war.

Somewhere in your office, buried in a desk drawer, there’s probably a handbook you received on your first day of work. It contains expense forms and rules about vacations, insurance options, and the company’s organizational chart. It has brightly colored graphs describing different health care plans, a list of relevant phone numbers, and instructions on how to access your email or enroll in the 401(k).

Now, imagine what you would tell a new colleague who asked for advice about how to succeed at your firm. Your recommendations probably wouldn’t contain anything you’d find in the company’s handbook. Instead, the tips you would pass along-who is trustworthy; which secretaries have more clout than their bosses; how to manipulate the bureaucracy to get something done-are the habits you rely on every day to survive. If you could somehow diagram all your work habits-and the informal power structures, relationships, alliances, and conflicts they represent-and then overlay your diagram with diagrams prepared by your colleagues, it would create a map of your firm’s secret hierarchy, a guide to who knows how to make things happen and who never seems to get ahead of the ball.

Nelson and Winter’s routines-and the truces they make possible-are critical to every kind of business. One study from Utrecht University in the Netherlands, for instance, looked at routines within the world of high fashion. To survive, every fashion designer has to possess some basic skills: creativity and a flair for haute couture as a start. But that’s not enough to succeed. [176] What makes the difference between success or failure are a designer’s routines-whether they have a system for getting Italian broadcloth before wholesalers’ stocks sell out, a process for finding the best zipper and button seamstresses, a routine for shipping a dress to a store in ten days, rather than three weeks. Fashion is such a complicated business that, without the right processes, a new company will get bogged down with logistics, and once that happens, creativity ceases to matter.

And which new designers are most likely to have the right habits? The ones who have formed the right truces and found the right alliances. [177] Truces are so important that new fashion labels usually succeed only if they are headed by people who left other fashion companies on good terms.

Some might think Nelson and Winter were writing a book on dry economic theory. But what they really produced was a guide to surviving in corporate America.

What’s more, Nelson and Winter’s theories also explain why things went so wrong at Rhode Island Hospital. The hospital had routines that created an uneasy peace between nurses and doctors-the whiteboards, for instance, and the warnings nurses whispered to one another were habits that established a baseline truce. These delicate pacts allowed the organization to function most of the time. But truces are only durable when they create real justice. If a truce is unbalanced-if the peace isn’t real-then the routines often fail when they are needed most.

The critical issue at Rhode Island Hospital was that the nurses were the only ones giving up power to strike a truce. It was the nurses who double-checked patients’ medications and made extra efforts to write clearly on charts; the nurses who absorbed abuse from stressed-out doctors; the nurses who helped separate kind physicians from the despots, so the rest of the staff knew who tolerated operating-room suggestions and who would explode if you opened your mouth. The doctors often didn’t bother to learn the nurses’ names. “The doctors were in charge, and we were underlings,” one nurse told me. “We tucked our tails and survived.”

The truces at Rhode Island Hospital were one-sided. So at those crucial moments-when, for instance, a surgeon was about to make a hasty incision and a nurse tried to intervene-the routines that could have prevented the accident crumbled, and the wrong side of an eighty-six-year-old man’s head was opened up.

Some might suggest that the solution is more equitable truces. That if the hospital’s leadership did a better job of allocating authority, a healthier balance of power might emerge and nurses and doctors would be forced into a mutual respect.

That’s a good start. Unfortunately, it isn’t enough. Creating successful organizations isn’t just a matter of balancing authority. For an organization to work, leaders must cultivate habits that both create a real and balanced peace and, paradoxically, make it absolutely clear who’s in charge.

III.

Philip Brickell, a forty-three-year-old employee of the London Underground, was inside the cavernous main hall of the King’s Cross subway station on a November evening in 1987 when a commuter stopped him as he was collecting tickets and said there was a burning tissue at the bottom of a nearby escalator. [178] [179]

King’s Cross was one of the largest, grandest, and most heavily trafficked of London’s subway stops, a labyrinth of deep escalators, passageways, and tunnels, some of which were almost a century old. The station’s escalators, in particular, were famous for their size and age. Some stretched as many as five stories into the ground and were built of wooden slats and rubber handrails, the same materials used to construct them decades earlier. More than a quarter million passengers passed through King’s Cross every day on six different train lines. During evening rush hour, the station’s ticketing hall was a sea of people hurrying beneath a ceiling repainted so many times that no one could recall its original hue.

The burning tissue, the passenger said, was at the bottom of one of the station’s longest escalators, servicing the Piccadilly line. Brickell immediately left his position, rode the escalator down to the platform, found the smoldering wad of tissue, and, with a rolled-up magazine, beat out the fire. Then he returned to his post.

Brickell didn’t investigate further. He didn’t try to figure out why the tissue was burning or if it might have flown off of a larger fire somewhere else within the station. He didn’t mention the incident to another employee or call the fire department. A separate department handled fire safety, and Brickell, in keeping with the strict divisions that ruled the Underground, knew better than to step on anyone’s toes. Besides, even if he had investigated the possibility of a fire, he wouldn’t have known what to do with any information he learned. The tightly prescribed chain of command at the Underground prohibited him from contacting another department without a superior’s direct authorization. And the Underground’s routines-handed down from employee to employee-told him that he should never, under any circumstances, refer out loud to anything inside a station as a “fire,” lest commuters become panicked. It wasn’t how things were done.

The Underground was governed by a sort of theoretical rule book that no one had ever seen or read-and that didn’t, in fact, exist except in the unwritten rules that shaped every employee’s life. For decades, the Underground had been run by the “Four Barons”-the chiefs of civil, signal, electrical, and mechanical engineering-and within each of their departments, there were bosses and subbosses who all jealously guarded their authority. The trains ran on time because all nineteen thousand Underground employees cooperated in a delicate system that passed passengers and trains among dozens-sometimes hundreds-of hands all day long. But that cooperation depended upon a balance of power between each of the four departments and all their lieutenants that, itself, relied upon thousands of habits that employees adhered to. These habits created a truce among the Four Barons and their deputies. And from that truce arose policies that told Brickell: Looking for fires isn’t your job. Don’t overstep your bounds.

“Even at the highest level, one director was unlikely to trespass on the territory of another,” an investigator would later note. “Thus, the engineering director did not concern himself with whether the operating staff were properly trained in fire safety and evacuation procedures because he considered those matters to be the province of the Operations Directorate.”

So Brickell didn’t say anything about the burning tissue. In other circumstances, it might have been an unimportant detail. In this case, the tissue was a stray warning-a bit of fuel that had escaped from a larger, hidden blaze-that would show how perilous even perfectly balanced truces can become if they aren’t designed just right. [180]

Fifteen minutes after Brickell returned to his booth, another passenger noticed a wisp of smoke as he rode up the Piccadilly escalator; he mentioned it to an Underground employee. The King’s Cross safety inspector, Christopher Hayes, was eventually roused to investigate. A third passenger, seeing smoke and a glow from underneath the escalator’s stairs, hit an emergency stop button and began shouting at passengers to exit the escalator. A policeman saw a slight smoky haze inside the escalator’s long tunnel, and, halfway down, flames beginning to dart above the steps.

Yet the safety inspector, Hayes, didn’t call the London Fire Brigade. He hadn’t seen any smoke himself, and another of the Underground’s unwritten rules was that the fire department should never be contacted unless absolutely necessary. The policeman who had noticed the haze, however, figured he should contact headquarters. His radio didn’t work underground, so he walked up a long staircase into the outdoors and called his superiors, who eventually passed word to the fire department. At 7:36 p.m.-twenty-two minutes after Brickell was alerted to the flaming tissue-the fire brigade received a call: “Small fire at King’s Cross.” Commuters were pushing past the policeman as he stood outside, speaking on his radio. They were rushing into the station, down into the tunnels, focused on getting home for dinner.

Within minutes, many of them would be dead.

At 7:36 P.M., an Underground worker roped off entry to the Piccadilly escalator and another started diverting people to a different stairway. New trains were arriving every few minutes. The platforms where passengers exited subway cars were crowded. A bottleneck started building at the bottom of an open staircase.

Hayes, the safety inspector, went into a passageway that led to the Piccadilly escalator’s machine room. In the dark, there was a set of controls for a sprinkler system specifically designed to fight fires on escalators. It had been installed years earlier, after a fire in another station had led to a series of dire reports about the risks of a sudden blaze. More than two dozen studies and reprimands had said that the Underground was unprepared for fires, and that staff needed to be trained in how to use sprinklers and fire extinguishers, which were positioned on every train platform. Two years earlier the deputy assistant chief of the London Fire Brigade had written to the operations director for railways, complaining about subway workers’ safety habits.

“I am gravely concerned,” the letter read. “I cannot urge too strongly that… clear instructions be given that on any suspicion of fire, the Fire Brigade be called without delay. This could save lives.”

However, Hayes, the safety inspector, never saw that letter because it was sent to a separate division from the one he worked within, and the Underground’s policies were never rewritten to reflect the warning. No one inside King’s Cross understood how to use the escalator sprinkler system or was authorized to use the extinguishers, because another department controlled them. Hayes completely forgot the sprinkler system existed. The truces ruling the Underground made sure everyone knew their place, but they left no room for learning about anything outside what you were assigned to know. Hayes ran past the sprinkler controls without so much as a glance.

When he reached the machine room, he was nearly overcome by heat. The fire was already too big to fight. He ran back to the main hall. There was a line of people standing at the ticket machines and hundreds of people milling about the room, walking to platforms or leaving the station. Hayes found a policeman.

“We’ve got to stop the trains and get everyone out of here,” he told him. “The fire is out of control. It’s going everywhere.”

At 7:42 P.M.-almost a half hour after the burning tissue-the first fireman arrived at King’s Cross. As he entered the ticketing hall he saw dense black smoke starting to snake along the ceiling. The escalator’s rubber handrails had begun to burn. As the acrid smell of burning rubber spread, commuters in the ticketing hall began to recognize that something was wrong. They moved toward the exits as firemen waded through the crowd, fighting against the tide.

Below, the fire was spreading. The entire escalator was now aflame, producing a superheated gas that rose to the top of the shaft enclosing the escalator, where it was trapped against the tunnel’s ceiling, which was covered with about twenty layers of old paint. A few years earlier, the Underground’s director of operations had suggested that all this paint might pose a fire hazard. Perhaps, he said, the old layers should be removed before a new one is applied?

Painting protocols were not in his purview, however. Paint responsibility resided with the maintenance department, whose chief politely thanked his colleague for the recommendation, and then noted that if he wanted to interfere with other departments, the favor would be swiftly returned.

The director of operations withdrew his recommendation.

As the superheated gases pooled along the ceiling of the escalator shaft, all those old layers of paint began absorbing the warmth. As each new train arrived, it pushed a fresh gust of oxygen into the station, feeding the fire like a bellows.

At 7:43 P.M., a train arrived and a salesman named Mark Silver exited. He knew immediately that something was wrong. The air was hazy, the platform packed with people. Smoke wafted around where he was standing, curling around the train cars as they sat on the tracks. He turned to reenter the train, but the doors had closed. He hammered on the windows, but there was an unofficial policy to avoid tardiness: Once the doors were sealed, they did not open again. Up and down the platform, Silver and other passengers screamed at the driver to open the doors. The signal light changed to green, and the train pulled away. One woman jumped on the tracks, running after the train as it moved into the tunnel. “Let me in!” she screamed.

Silver walked down the platform, to where a policeman was directing everyone away from the Piccadilly escalator and to another stairway. There were crowds of panicked people waiting to get upstairs. They could all smell the smoke, and everyone was packed together. It felt hot-either from the fire or the crush of people, Silver wasn’t sure. He finally got to the bottom of an escalator that had been turned off. As he climbed toward the ticketing hall, he could feel his legs burning from heat coming through a fifteen-foot wall separating him from the Piccadilly shaft. “I looked up and saw the walls and ceiling sizzling,” he later said.

At 7:45 P.M., an arriving train forced a large gust of air into the station. As the oxygen fed the fire, the blaze in the Piccadilly escalator roared. The superheated gases along the ceiling of the shaft, fueled by fire below and sizzling paint above, reached a combustion temperature, known as a “flashover point.” At that moment, everything inside the shaft-the paint, the wooden escalator stairs, and any other available fuel-ignited in a fiery blast. The force of the sudden incineration acted the explosion of gunpowder at the base of a rifle barrel. It began pushing the fire upward through the long shaft, absorbing more heat and velocity as the blaze expanded until it shot out of the tunnel and into the ticketing hall in a wall of flames that set metal, tile, and flesh on fire. The temperature inside the hall shot up 150 degrees in half a second. A policeman riding one of the side escalators later told investigators that he saw “a jet of flame that shot up and then collected into a kind of ball.” There were nearly fifty people inside the hall at the time.

Aboveground, on the street, a passerby felt heat explode from one of the subway’s exits, saw a passenger stagger out, and ran to help. “I got hold of his right hand with my right hand but as our hands touched I could feel his was red hot and some of the skin came off in my hand,” the rescuer said. A policeman who was entering the ticketing hall as the explosion occurred later told reporters, from a hospital bed, that “a fireball hit me in the face and knocked me off my feet. My hands caught fire. They were just melting.”

He was one of the last people to exit the hall alive.

Shortly after the explosion, dozens of fire trucks arrived. But because the fire department’s rules instructed them to connect their hoses to street-level hydrants, rather than those installed by the Underground inside the station, and because none of the subway employees had blueprints showing the station’s layout-all the plans were in an office that was locked, and none of the ticketing agents or the station manager had keys-it took hours to extinguish the flames.

When the blaze was finally put out at 1:46 A.M.-six hours after the burning tissue was noticed-the toll stood at thirty-one dead and dozens injured.

“Why did they send me straight into the fire?” a twenty-year-old music teacher asked the next day from a hospital bed. “I could see them burning. I could hear them screaming. Why didn’t someone take charge?” [181]

To answer those questions, consider a few of the truces the London Underground relied upon to function:

Ticketing clerks were warned that their jurisdiction was strictly limited to selling tickets, so if they saw a burning tissue, they didn’t warn anyone for fear of overstepping their bounds.

Station employees weren’t trained how to use the sprinkler system or extinguishers, because that equipment was overseen by a different division.

The station’s safety inspector never saw a letter from the London Fire Brigade warning about fire risks because it was sent to the operations director, and information like that wasn’t shared across divisions.

Employees were instructed only to contact the fire brigade as a last resort, so as not to panic commuters unnecessarily.

The fire brigade insisted on using its own street-level hydrants, ignoring pipes in the ticketing hall that could have delivered water, because they had been ordered not to use equipment installed by other agencies.

In some ways, each of these informal rules, on its own, makes a certain amount of sense. For instance, the habits that kept ticketing clerks focused on selling tickets instead of doing anything else-including keeping an eye out for warning signs of fire-existed because, years earlier, the Underground had problems with understaffed kiosks. Clerks kept leaving their posts to pick up trash or point tourists toward their trains, and as a result, long lines would form. So clerks were ordered to stay in their booths, sell tickets, and not worry about anything else. It worked. Lines disappeared. If clerks saw something amiss outside their kiosks-beyond their scope of responsibility-they minded their own business.

And the fire brigade’s habit of insisting on their own equipment? That was a result of an incident, a decade earlier, when a fire had raged in another station as firemen wasted precious minutes trying to hook up their hoses to unfamiliar pipes. Afterward, everyone decided it was best to stick with what they knew.

None of these routines, in other words, were arbitrary. Each was designed for a reason. The Underground was so vast and complicated that it could operate smoothly only if truces smoothed over potential obstacles. Unlike at Rhode Island Hospital, each truce created a genuine balance of power. No department had the upper hand.

Yet thirty-one people died.

The London Underground’s routines and truces all seemed logical until a fire erupted. At which point, an awful truth emerged: No one person, department, or baron had ultimate responsibility for passengers’ safety. [182]

Sometimes, one priority-or one department or one person or one goal-needs to overshadow everything else, though it might be unpopular or threaten the balance of power that keeps trains running on time. Sometimes, a truce can create dangers that outweigh any peace.

There’s a paradox in this observation, of course. How can an organization implement habits that balance authority and, at the same time, choose a person or goal that rises above everyone else? How do nurses and doctors share authority while still making it clear who is in charge? How does a subway system avoid becoming bogged down in turf battles while making sure safety is still a priority, even if that means lines of authority must be redrawn?

The answer lies in seizing the same advantage that Tony Dungy encountered when he took over the woeful Bucs and Paul O’Neill discovered when he became CEO of flailing Alcoa. It’s the same opportunity Howard Schultz exploited when he returned to a flagging Starbucks in 2007. All those leaders seized the possibilities created by a crisis. During turmoil, organizational habits become malleable enough to both assign responsibility and create a more equitable balance of power. Crises are so valuable, in fact, that sometimes it’s worth stirring up a sense of looming catastrophe rather than letting it die down.

IV.

Four months after the elderly man with the botched skull surgery died at Rhode Island Hospital, another surgeon at the hospital committed a similar error, operating on the wrong section of another patient’s head. The state’s health department reprimanded the facility and fined it $50,000. Eighteen months later, a surgeon operated on the wrong part of a child’s mouth during a cleft palate surgery. Five months after that, a surgeon operated on a patient’s wrong finger. Ten months after that, a drill bit was left inside a man’s head. For these transgressions, the hospital was fined another $450,000. [183]

Rhode Island Hospital is not the only medical institution where such accidents happen, of course, but they were unlucky enough to become the poster child for such mistakes. Local newspapers printed detailed stories of each incident. Television stations set up camp outside the hospital. The national media joined in, too. “The problem’s not going away,” a vice president of the national hospital accreditation organization told an Associated Press reporter. [184] Rhode Island Hospital, the state’s medical authorities declared to reporters, was a facility in chaos.

“It felt like working in a war zone,” a nurse told me. “There were TV reporters ambushing doctors as they walked to their cars. One little boy asked me to make sure the doctor wouldn’t accidentally cut off his arm during surgery. It felt like everything was out of control.” [185]

As critics and the media piled on, a sense of crisis emerged within the hospital. [186] Some administrators started worrying that the facility would lose its accreditation. Others became defensive, attacking the television stations for singling them out. “I found a button that said ‘Scapegoat’ that I was going to wear to work,” one doctor told me. “My wife said that was a bad idea.”

Then an administrator, Dr. Mary Reich Cooper, who had become chief quality officer a few weeks before the eighty-six-year-old man’s death, spoke up. In meetings with the hospital’s administrators and staff, Cooper said that they were looking at the situation all wrong.

All this criticism wasn’t a bad thing, she said. In fact, the hospital had been given an opportunity that few organizations ever received.

“I saw this as an opening,” Dr. Cooper told me. “There’s a long history of hospitals trying to attack these problems and failing. Sometimes people need a jolt, and all the bad publicity was a serious jolt. It gave us a chance to reexamine everything.”

Rhode Island Hospital shut down all elective surgery units for an entire day-a huge expense-and put the entire staff through an intensive training program that emphasized teamwork and stressed the importance of empowering nurses and medical staff. The chief of neurosurgery resigned and a new leader was selected. The hospital invited the Center for Transforming Healthcare-a coalition of leading medical institutions-to help redesign its surgical safeguards. Administrators installed video cameras in operating rooms to make sure time-outs occurred and checklists were mandated for every surgery. [187] A computerized system allowed any hospital employee to anonymously report problems that endangered patient health. [188]

Some of those initiatives had been proposed at Rhode Island Hospital in previous years, but they had always been struck down. Doctors and nurses didn’t want people recording their surgeries or other hospitals telling them how to do their jobs.

But once a sense of crisis gripped Rhode Island Hospital, everyone became more open to change. [189]

Other hospitals have made similar shifts in the wake of mistakes and have brought down error rates that just years earlier had seemed immune to improvement. [190] Like Rhode Island Hospital, these institutions have found that reform is usually possible only once a sense of crisis takes hold. For instance, one of Harvard University’s teaching hospitals, Beth Israel Deaconess Medical Center, went through a spate of errors and internal battles in the late 1990s that spilled into newspaper articles and ugly shouting matches between nurses and administrators at public meetings. There was talk among some state officials of forcing the hospital to close departments until they could prove the mistakes would stop. Then the hospital, under attack, coalesced around solutions to change its culture. Part of the answer was “safety rounds,” in which, every three months, a senior physician discussed a particular surgery or diagnosis and described, in painstaking detail, a mistake or near miss to an audience of hundreds of her or his peers.

“It’s excruciating to admit a mistake publicly,” said Dr. Donald Moorman, until recently Beth Israel Deaconess’s associate surgeon in chief. “Twenty years ago, doctors wouldn’t do it. But a real sense of panic has spread through hospitals now, and even the best surgeons are willing to talk about how close they came to a big error. The culture of medicine is changing.”

Good leaders seize crises to remake organizational habits. NASA administrators, for instance, tried for years to improve the agency’s safety habits, but those efforts were unsuccessful until the space shuttle Challenger exploded in 1986. In the wake of that tragedy, the organization was able to overhaul how it enforced quality standards. [191] Airline pilots, too, spent years trying to convince plane manufacturers and air traffic controllers to redesign how cockpits were laid out and traffic controllers communicated. Then, a runway error on the Spanish island of Tenerife in 1977 killed 583 people and, within five years, cockpit design, runway procedures, and air traffic controller communication routines were overhauled. [192]

In fact, crises are such valuable opportunities that a wise leader often prolongs a sense of emergency on purpose. That’s exactly what occurred after the King’s Cross station fire. Five days after the blaze, the British secretary of state appointed a special investigator, Desmond Fennell, to study the incident. Fennell began by interviewing the Underground’s leadership, and quickly discovered that everyone had known-for years-that fire safety was a serious problem, and yet nothing had changed. Some administrators had proposed new hierarchies that would have clarified responsibility for fire prevention. Others had proposed giving station managers more power so that they could bridge departmental divides. None of those reforms had been implemented.

When Fennell began suggesting changes of his own, he saw the same kinds of roadblocks-department chiefs refusing to take responsibility or undercutting him with whispered threats to their subordinates-start to emerge.

So he decided to turn his inquiry into a media circus.

He called for public hearings that lasted ninety-one days and revealed an organization that had ignored multiple warnings of risks. He implied to newspaper reporters that commuters were in grave danger whenever they rode the subway. He cross-examined dozens of witnesses who described an organization where turf battles mattered more than commuter safety. His final report, released almost a year after the fire, was a scathing, 250-page indictment of the Underground portraying an organization crippled by bureaucratic ineptitude. “Having set out as an Investigation into the events of one night,” Fennell wrote, the report’s “scope was necessarily enlarged into the examination of a system.” He concluded with pages and pages of stinging criticisms and recommendations that, essentially, suggested much of the organization was either incompetent or corrupt.

The response was instantaneous and overwhelming. Commuters picketed the Underground’s offices. The organization’s leadership was fired. A slew of new laws were passed and the culture of the Underground was overhauled. Today, every station has a manager whose primary responsibility is passenger safety, and every employee has an obligation to communicate at the smallest hint of risk. All the trains still run on time. But the Underground’s habits and truces have adjusted just enough to make it clear who has ultimate responsibility for fire prevention, and everyone is empowered to act, regardless of whose toes they might step on.

The same kinds of shifts are possible at any company where institutional habits-through thoughtlessness or neglect-have created toxic truces. A company with dysfunctional habits can’t turn around simply because a leader orders it. Rather, wise executives seek out moments of crisis-or create the perception of crisis-and cultivate the sense that something must change, until everyone is finally ready to overhaul the patterns they live with each day.

“You never want a serious crisis to go to waste,” Rahm Emanuel told a conference of chief executives in the wake of the 2008 global financial meltdown, soon after he was appointed as President Obama’s chief of staff. “This crisis provides the opportunity for us to do things that you could not do before.” Soon afterward, the Obama administration convinced a once-reluctant Congress to pass the president’s $787 billion stimulus plan. Congress also passed Obama’s health care reform law, reworked consumer protection laws, and approved dozens of other statutes, from expanding children’s health insurance to giving women new opportunities to sue over wage discrimination. It was one of the biggest policy overhauls since the Great Society and the New Deal, and it happened because, in the aftermath of a financial catastrophe, lawmakers saw opportunity.

Something similar happened at Rhode Island Hospital in the wake of the eighty-six-year-old man’s death and the other surgical errors. Since the hospital’s new safety procedures were fully implemented in 2009, no wrong-site errors have occurred. The hospital recently earned a Beacon Award, the most prestigious recognition of critical care nursing, and honors from the American College of Surgeons for the quality of cancer care.

More important, say the nurses and doctors who work there, Rhode Island Hospital feels like a completely different place.

In 2010, a young nurse named Allison Ward walked into an operating room to assist on a routine surgery. She had started working in the OR a year earlier. She was the youngest and least experienced person in the room. Before the surgery began, the entire surgical team gathered over the unconscious patient for a time-out. The surgeon read from a checklist, posted on the wall, which detailed every step of the operation.

“Okay, final step,” he said before he picked up his scalpel. “Does anyone have any concerns before we start?”

The doctor had performed hundreds of these surgeries. He had an office full of degrees and awards.

“Doctor,” the twenty-seven-year-old Ward said, “I want to remind everyone that we have to pause before the first and second procedures. You didn’t mention that, and I just want to make sure we remember.”

It was the type of comment that, a few years ago, might have earned her a rebuke. Or ended her career.

“Thanks for adding that,” the surgeon said. “I’ll remember to mention it next time.

“Okay,” he said, “let’s start.”

“I know this hospital has gone through some hard periods,” Ward later told me. “But it’s really cooperative now. Our training, all the role models-the whole culture of the hospital is focused on teamwork. I feel like I can say anything. It’s an amazing place to work.”

¹ The reporting in this chapter is based upon interviews with multiple people working at Rhode Island Hospital and involved in this incident some of whom provided different accounts of events. For details on responses from hospital representatives and the surgeon involved, please see the notes.

I.

Andrew Pole had just started working as a data expert for Target when a few colleagues from the marketing department stopped by his desk one day and asked the kind of question Pole had been born to answer:

“Can your computers figure out which customers are pregnant, even if they don’t want us to know?”

Pole was a statistician. His entire life revolved around using data to understand people. He had grown up in a small North Dakota town, and while his friends were attending 4-H or building model rockets, Pole was playing with computers. After college, he got a graduate degree in statistics and then another in economics, and while most of his classmates in the econ program at the University of Missouri were headed to insurance companies or government bureaucracies, Pole was on a different track. He’d become obsessed with the ways economists were using pattern analysis to explain human behavior. Pole, in fact, had tried his hand at a few informal experiments himself. He once threw a party and polled everyone on their favorite jokes, and then attempted to create a mathematical model for the perfect one-liner. He had sought to calculate the exact amount of beer he needed to drink in order to work up the confidence to talk to women at parties, but not so much that he would make a fool of himself. (That particular study never seemed to come out right.)

But those experiments were child’s play, he knew, to how corporate America was using data to scrutinize people’s lives. Pole wanted in. So when he graduated and heard that Hallmark, the greeting card company, was looking to hire statisticians in Kansas City, he submitted an application and was soon spending his days scouring sales data to determine if pictures of pandas or elephants sold more birthday cards, and if “What Happens at Grandma’s Stays at Grandma’s” is funnier in red or blue ink. It was heaven.

Six years later, in 2002, when Pole learned that Target was looking for number crunchers, he made the jump. Target, he knew, was a whole other magnitude when it came to data collection. Every year, millions of shoppers walked into Target’s 1,147 stores and handed over terabytes of information about themselves. Most had no idea they were doing it. They used their customer loyalty cards, redeemed coupons they had received in the mail, or used a credit card, unaware that Target could then link their purchases to an individualized demographic profile.

To a statistician, this data was a magic window for peering into customers’ preferences. Target sold everything from groceries to clothing, electronics and lawn furniture, and by closely tracking people’s buying habits, the company’s analysts could predict what was occurring within their homes. Someone’s buying new towels, sheets, silverware, pans, and frozen dinners? They probably just bought a new house-or are getting a divorce. A cart loaded up with bug spray, kids’ underwear, a flashlight, lots of batteries, Real Simple, and a bottle of Chardonnay? Summer camp is around the corner and Mom can hardly wait.

Working at Target offered Pole a chance to study the most complicated of creatures-the American shopper-in its natural habitat. His job was to build mathematical models that could crawl through data and determine which households contained kids and which were dedicated bachelors; which shoppers loved the outdoors and who was more interested in ice cream and romance novels. Pole’s mandate was to become a mathematical mind reader, deciphering shoppers’ habits in order to convince them to spend more.

Then, one afternoon, a few of Pole’s colleagues from the marketing department stopped by his desk. They were trying to figure out which of Target’s customers were pregnant based on their buying patterns, they said. Pregnant women and new parents, after all, are the holy grail of retail. There is almost no more profitable, product-hungry, price-insensitive group in existence. It’s not just diapers and wipes. People with infants are so tired that they’ll buy everything they need-juice and toilet paper, socks and magazines-wherever they purchase their bottles and formula. What’s more, if a new parent starts shopping at Target, they’ll keep coming back for years.

Figuring out who was pregnant, in other words, could make Target millions of dollars.

Pole was intrigued. What better challenge for a statistical fortune-teller than not only getting inside shoppers’ minds, but their bedrooms?

By the time the project was done, Pole would learn some important lessons about the dangers of preying on people’s most intimate habits. He would learn, for example, that hiding what you know is sometimes as important as knowing it, and that not all women are enthusiastic about a computer program scrutinizing their reproductive plans.

Not everyone, it turns out, thinks mathematical mind reading is cool.

“I guess outsiders could say this is a little bit like Big Brother,” Pole told me. “That makes some people uncomfortable.”

Once upon a time, a company like Target would never have hired a guy like Andrew Pole. As little as twenty years ago retailers didn’t do this kind of intensely data-driven analysis. Instead, Target, as well as grocery stores, shopping malls, greeting card sellers, clothing retailers, and other firms, tried to peer inside consumers’ heads the old-fashioned way: by hiring psychologists who peddled vaguely scientific tactics they claimed could make customers spend more.

Some of those methods are still in use today. If you walk into a Walmart, Home Depot, or your local shopping center and look closely, you’ll see retailing tricks that have been around for decades, each designed to exploit your shopping subconscious.

Take, for instance, how you buy food.

Chances are, the first things you see upon entering your grocery store are fruits and vegetables arranged in attractive, bountiful piles. If you think about it, positioning produce at the front of a store doesn’t make much sense, because fruits and vegetables bruise easily at the bottom of a shopping cart; logically, they should be situated by the registers, so they come at the end of a trip. But as marketers and psychologists figured out long ago, if we start our shopping sprees by loading up on healthy stuff, we’re much more likely to buy Doritos, Oreos, and frozen pizza when we encounter them later on. The burst of subconscious virtuousness that comes from first buying butternut squash makes it easier to put a pint of ice cream in the cart later.

Or take the way most of us turn to the right after entering a store. (Did you know you turn right? It’s almost certain you do. There are thousands of hours of videotapes showing shoppers turning right once they clear the front doors.) As a result of this tendency, retailers fill the right side of the store with the most profitable products they’re hoping you’ll buy right off the bat. Or consider cereal and soups: When they’re shelved out of alphabetical order and seemingly at random, our instinct is to linger a bit longer and look at a wider selection. So you’ll rarely find Raisin Bran next to Rice Chex. Instead, you’ll have to search the shelves for the cereal you want, and maybe get tempted to grab an extra box of another brand. [193]

The problem with these tactics, however, is that they treat each shopper exactly the same. They’re fairly primitive, one-size-fits-all solutions for triggering buying habits.

In the past two decades, however, as the retail marketplace has become more and more competitive, chains such as Target began to understand they couldn’t rely on the same old bag of tricks. The only way to increase profits was to figure out each individual shopper’s habits and to market to people one by one, with personalized pitches designed to appeal to customers’ unique buying preferences.

In part, this realization came from a growing awareness of how powerfully habits influence almost every shopping decision. A series of experiments convinced marketers that if they managed to understand a particular shopper’s habits, they could get them to buy almost anything. [194] One study tape-recorded consumers as they walked through grocery stores. Researchers wanted to know how people made buying decisions. In particular, they looked for shoppers who had come with shopping lists-who, theoretically, had decided ahead of time what they wanted to get.

What they discovered was that despite those lists, more than 50 percent of purchasing decisions occurred at the moment a customer saw a product on the shelf, because, despite shoppers’ best intentions, their habits were stronger than their written intentions. “Let’s see,” one shopper muttered to himself as he walked through a store. “Here are the chips. I will skip them. Wait a minute. Oh! The Lay’s potato chips are on sale!” He put a bag in his cart. [195] Some shoppers bought the same brands, month after month, even if they admitted they didn’t like the product very much (“I’m not crazy about Folgers, but it’s what I buy, you know? What else is there?” one woman said as she stood in front of a shelf containing dozens of other coffee brands). Shoppers bought roughly the same amount of food each time they went shopping, even if they had pledged to cut back.

“Consumers sometimes act like creatures of habit, automatically repeating past behavior with little regard to current goals,” two psychologists at the University of Southern California wrote in 2009. [196]

The surprising aspect of these studies, however, was that even though everyone relied on habits to guide their purchases, each person’s habits were different. The guy who liked potato chips bought a bag every time, but the Folgers woman never went down the potato chip aisle. There were people who bought milk whenever they shopped-even if they had plenty at home-and there were people who always purchased desserts when they said they were trying to lose weight. But the milk buyers and the dessert addicts didn’t usually overlap.

The habits were unique to each person.

Target wanted to take advantage of those individual quirks. But when millions of people walk through your doors every day, how do you keep track of their preferences and shopping patterns?

You collect data. Enormous, almost inconceivably large amounts of data.

Starting a little over a decade ago, Target began building a vast data warehouse that assigned every shopper an identification code-known internally as the “Guest ID number”-that kept tabs on how each person shopped. When a customer used a Target-issued credit card, handed over a frequent-buyer tag at the register, redeemed a coupon that was mailed to their house, filled out a survey, mailed in a refund, phoned the customer help line, opened an email from Target, visited Target.com, or purchased anything online, the company’s computers took note. A record of each purchase was linked to that shopper’s Guest ID number along with information on everything else they’d ever bought.

Also linked to that Guest ID number was demographic information that Target collected or purchased from other firms, including the shopper’s age, whether they were married and had kids, which part of town they lived in, how long it took them to drive to the store, an estimate of how much money they earned, if they’d moved recently, which websites they visited, the credit cards they carried in their wallet, and their home and mobile phone numbers. Target can purchase data that indicates a shopper’s ethnicity, their job history, what magazines they read, if they have ever declared bankruptcy, the year they bought (or lost) their house, where they went to college or graduate school, and whether they prefer certain brands of coffee, toilet paper, cereal, or applesauce.

There are data peddlers such as InfiniGraph that “listen” to shoppers’ online conversations on message boards and Internet forums, and track which products people mention favorably. A firm named Rapleaf sells information on shoppers’ political leanings, reading habits, charitable giving, the number of cars they own, and whether they prefer religious news or deals on cigarettes. [197] Other companies analyze photos that consumers post online, cataloging if they are obese or skinny, short or tall, hairy or bald, and what kinds of products they might want to buy as a result. (Target, in a statement, declined to indicate what demographic companies it does business with and what kinds of information it studies.)

“It used to be that companies only knew what their customers wanted them to know,” said Tom Davenport, one of the leading researchers on how businesses use data and analytics. “That world is far behind us. You’d be shocked how much information is out there-and every company buys it, because it’s the only way to survive.”

If you use your Target credit card to purchase a box of Popsicles once a week, usually around 6:30 p.m. on a weekday, and megasized trash bags every July and October, Target’s statisticians and computer programs will determine that you have kids at home, tend to stop for groceries on your way back from work, and have a lawn that needs mowing in the summer and trees that drop leaves in the fall. It will look at your other shopping patterns and notice that you sometimes buy cereal, but never purchase milk-which means that you must be buying it somewhere else. So Target will mail you coupons for 2 percent milk, as well as for chocolate sprinkles, school supplies, lawn furniture, rakes, and-since it’s likely you’ll want to relax after a long day at work-beer. The company will guess what you habitually buy, and then try to convince you to get it at Target. The firm has the capacity to personalize the ads and coupons it sends to every customer, even though you’ll probably never realize you’ve received a different flyer in the mail than your neighbors.

“With the Guest ID, we have your name, address, and tender, we know you’ve got a Target Visa, a debit card, and we can tie that to your store purchases,” Pole told an audience of retail statisticians at a conference in 2010. The company can link about half of all in-store sales to a specific person, almost all online sales, and about a quarter of online browsing.

At that conference, Pole flashed a slide showing a sample of the data Target collects, a diagram that caused someone in the audience to whistle in wonder when it appeared on the screen: [198]

The problem with all this data, however, is that it’s meaningless without statisticians to make sense of it. To a layperson, two shoppers who both buy orange juice look the same. It requires a special kind of mathematician to figure out that one of them is a thirty-four-year-old woman purchasing juice for her kids (and thus might appreciate a coupon for a Thomas the Tank Engine DVD) and the other is a twenty-eight-year-old bachelor who drinks juice after going for a run (and thus might respond to discounts on sneakers). Pole and the fifty other members of Target’s Guest Data and Analytical Services department were the ones who found the habits hidden in the facts.

“We call it the ‘guest portrait,’ ” Pole told me. “The more I know about someone, the better I can guess their buying patterns. I’m not going to guess everything about you every time, but I’ll be right more often than I’m wrong.”

By the time Pole joined Target in 2002, the analytics department had already built computer programs to identify households containing children and, come each November, send their parents catalogs of bicycles and scooters that would look perfect under the Christmas tree, as well as coupons for school supplies in September and advertisements for pool toys in June. The computers looked for shoppers buying bikinis in April, and sent them coupons for sunscreen in July and weight-loss books in December. If it wanted, Target could send each customer a coupon book filled with discounts for products they were fairly certain the shoppers were going to buy, because they had already purchased those exact items before.

Target isn’t alone in its desire to predict consumers’ habits. Almost every major retailer, including Amazon.com, Best Buy, Kroger supermarkets, 1-800-Flowers, Olive Garden, Anheuser-Busch, the U.S. Postal Service, Fidelity Investments, Hewlett-Packard, Bank of America, Capital One, and hundreds of others, have “predictive analytics” departments devoted to figuring out consumers’ preferences. “But Target has always been one of the smartest at this,” said Eric Siegel, who runs a conference called Predictive Analytics World. “The data doesn’t mean anything on its own. Target’s good at figuring out the really clever questions.”

It doesn’t take a genius to know that someone buying cereal probably also needs milk. But there were other, much harder-and more profitable-questions to be answered.

Which is why, a few weeks after Pole was hired, his colleagues asked if it was possible to determine who was pregnant, even if that woman didn’t want anyone to know.

In 1984, a visiting professor at UCLA named Alan Andreasen published a paper that set out to answer a basic question: Why do some people suddenly change their shopping routines?

Andreasen’s team had spent the previous year conducting telephone surveys with consumers around Los Angeles, interrogating them about their recent shopping trips. Whenever someone answered the phone, the scientists would barrage them with questions about which brands of toothpaste and soap they had purchased and if their preferences had shifted. All told, they interviewed almost three hundred people. Like other researchers, they found that most people bought the same brands of cereal and deodorant week after week. Habits reigned supreme.

Except when they didn’t.

For instance, 10.5 percent of the people Andreasen surveyed had switched toothpaste brands in the previous six months. More than 15 percent had started buying a new kind of laundry detergent.

Andreasen wanted to know why these people had deviated from their usual patterns. What he discovered has become a pillar of modern marketing theory: People’s buying habits are more likely to change when they go through a major life event. When someone gets married, for example, they’re more likely to start buying a new type of coffee. When they move into a new house, they’re more apt to purchase a different kind of cereal. When they get divorced, there’s a higher chance they’ll start buying different brands of beer. [199] Consumers going through major life events often don’t notice, or care, that their shopping patterns have shifted. However, retailers notice, and they care quite a bit. [200]

“Changing residence, getting married or divorced, losing or changing a job, having someone enter or leave the household,” Andreasen wrote, are life changes that make consumers more “vulnerable to intervention by marketers.”

And what’s the biggest life event for most people? What causes the greatest disruption and “vulnerability to marketing interventions”? Having a baby. There’s almost no greater upheaval for most customers than the arrival of a child. As a result, new parents’ habits are more flexible at that moment than at almost any other period in an adult’s life.

So for companies, pregnant women are gold mines.

New parents buy lots of stuff-diapers and wipes, cribs and Onesies, blankets and bottles-that stores such as Target sell at a significant profit. One survey conducted in 2010 estimated that the average parent spends $6,800 on baby items before a child’s first birthday. [201]

But that’s just the tip of the shopping iceberg. Those initial expenditures are peanuts compared with the profits a store can earn by taking advantage of a new parent’s shifting shopping habits. If exhausted moms and sleep-deprived dads start purchasing baby formula and diapers at Target, they’ll start buying their groceries, cleaning supplies, towels, underwear, and-well, the sky’s the limit-from Target as well. Because it’s easy. To a new parent, easy matters most of all.

“As soon as we get them buying diapers from us, they’re going to start buying everything else, too,” Pole told me. “If you’re rushing through the store, looking for bottles, and you pass orange juice, you’ll grab a carton. Oh, and there’s that new DVD I want. Soon, you’ll be buying cereal and paper towels from us, and keep coming back.”

New parents are so valuable that major retailers will do almost anything to find them, including going inside maternity wards, even if their products have nothing to do with infants. One New York hospital, for instance, provides every new mother with a gift bag containing samples of hair gel, face wash, shaving cream, an energy bar, shampoo, and a soft-cotton T-shirt. Inside are coupons for an online photo service, hand soap, and a local gym. There are also samples of diapers and baby lotions, but they’re lost among the nonbaby supplies. In 580 hospitals across the United States, new mothers get gifts from the Walt Disney Company, which in 2010 started a division specifically aimed at marketing to the parents of infants. Procter & Gamble, Fisher-Price, and other firms have similar giveaway programs. Disney estimates the North American new baby market is worth $36.3 billion a year. [202]

But for companies such as Target, approaching new moms in the maternity ward is, in some senses, too late. By then, they’re already on everyone else’s radar screen. Target didn’t want to compete with Disney and Procter & Gamble; they wanted to beat them. Target’s goal was to start marketing to parents before the baby arrived-which is why Andrew Pole’s colleagues approached him that day to ask about building a pregnancy-prediction algorithm. If they could identify expecting mothers as early as their second trimester, they could capture them before anyone else.

The only problem was that figuring out which customers are pregnant is harder than it seems. Target had a baby shower registry, and that helped identify some pregnant women-and what’s more, all those soon-to-be mothers willingly handed over valuable information, like their due dates, that let the company know when to send them coupons for prenatal vitamins or diapers. But only a fraction of Target’s pregnant customers used the registry.

Then there were other customers who executives suspected were pregnant because they purchased maternity clothing, nursery furniture, and boxes of diapers. Suspecting and knowing, however, are two different things. How do you know whether someone buying diapers is pregnant or buying a gift for a pregnant friend? What’s more, timing matters. A coupon that’s useful a month before the due date might get put in the trash a few weeks after the baby arrives.

Pole started working on the problem by scouring the information in Target’s baby shower registry, which let him observe how the average woman’s shopping habits changed as her due date approached. The registry was like a laboratory where he could test hunches. Each expectant mother handed over her name, her spouse’s name, and her due date. Target’s data warehouse could link that information to the family’s Guest IDs. As a result, whenever one of these women purchased something in a store or online, Pole, using the due date the woman provided, could plot the trimester in which the purchase occurred. Before long, he was picking up patterns.

Expectant mothers, he discovered, shopped in fairly predictable ways. Take, for example, lotions. Lots of people buy lotion, but a Target data analyst noticed that women on the baby registry were buying unusually large quantities of unscented lotion around the beginning of their second trimester. Another analyst noted that sometime in the first twenty weeks, many pregnant women loaded up on vitamins, such as calcium, magnesium, and zinc. Lots of shoppers purchase soap and cotton balls every month, but when someone suddenly starts buying lots of scent-free soap and cotton balls, in addition to hand sanitizers and an astounding number of washcloths, all at once, a few months after buying lotions and magnesium and zinc, it signals they are getting close to their delivery date.

As Pole’s computer program crawled through the data, he was able to identify about twenty-five different products that, when analyzed together, allowed him to, in a sense, peer inside a woman’s womb. Most important, he could guess what trimester she was in-and estimate her due date-so Target could send her coupons when she was on the brink of making new purchases. By the time Pole was done, his program could assign almost any regular shopper a “pregnancy prediction” score.

Jenny Ward, a twenty-three-year-old in Atlanta who bought cocoa butter lotion, a purse large enough to double as a diaper bag, zinc, magnesium, and a bright blue rug? There’s an 87 percent chance that she’s pregnant and that her delivery date is sometime in late August. [203] Liz Alter in Brooklyn, a thirty-five-year-old who purchased five packs of washcloths, a bottle of “sensitive skin” laundry detergent, baggy jeans, vitamins containing DHA, and a slew of moisturizers? She’s got a 96 percent chance of pregnancy, and she’ll probably give birth in early May. Caitlin Pike, a thirty-nine-year-old in San Francisco who purchased a $250 stroller, but nothing else? She’s probably buying for a friend’s baby shower. Besides, her demographic data shows she got divorced two years ago.

Pole applied his program to every shopper in Target’s database. When it was done, he had a list of hundreds of thousands of women who were likely to be pregnant that Target could inundate with advertisements for diapers, lotions, cribs, wipes, and maternity clothing at times when their shopping habits were particularly flexible. If a fraction of those women or their husbands started doing their shopping at Target, it would add millions to the company’s bottom line.

Then, just as this advertising avalanche was about to begin, someone within the marketing department asked a question: How are women going to react when they figure out how much Target knows?

“If we send someone a catalog and say, ‘Congratulations on your first child!’ and they’ve never told us they’re pregnant, that’s going to make some people uncomfortable,” Pole told me. “We are very conservative about compliance with all privacy laws. But even if you’re following the law, you can do things where people get queasy.”

There’s good reason for such worries. About a year after Pole created his pregnancy prediction model, a man walked into a Minnesota Target and demanded to see the manager. He was clutching an advertisement. He was very angry.

“My daughter got this in the mail!” he said. “She’s still in high school, and you’re sending her coupons for baby clothes and cribs? Are you trying to encourage her to get pregnant?”

The manager didn’t have any idea what the man was talking about. He looked at the mailer. Sure enough, it was addressed to the man’s daughter and contained advertisements for maternity clothing, nursery furniture, and pictures of smiling infants gazing into their mothers’ eyes.

The manager apologized profusely, and then called, a few days later, to apologize again.

The father was somewhat abashed.

“I had a talk with my daughter,” he said. “It turns out there’s been some activities in my house I haven’t been completely aware of.” He took a deep breath. “She’s due in August. I owe you an apology.”

Target is not the only firm to have raised concerns among consumers. Other companies have been attacked for using data in far less intrusive ways. In 2011, for instance, a New York resident sued McDonald’s, CBS, Mazda, and Microsoft, alleging those companies’ advertising agency monitored people’s Internet usage to profile their buying habits. [204] There are ongoing class action lawsuits in California against Target, Walmart, Victoria’s Secret, and other retail chains for asking customers to give their zip codes when they use credit cards, and then using that information to ferret out their mailing addresses. [205]

Using data to predict a woman’s pregnancy, Pole and his colleagues knew, was a potential public relations disaster. So how could they get their advertisements into expectant mothers’ hands without making it appear they were spying on them? How do you take advantage of someone’s habits without letting them know you’re studying every detail of their lives?¹

II.

In the summer of 2003, a promotion executive at Arista Records named Steve Bartels began calling up radio DJs to tell them about a new song he was certain they would love. It was called “Hey Ya!” by the hip-hop group OutKast.

“Hey Ya!” was an upbeat fusion of funk, rock, and hip-hop with a dollop of Big Band swing, from one of the most popular bands on earth. It sounded like nothing else on the radio. “It made the hair on my arms stand up the first time I heard it,” Bartels told me. “It sounded like a hit, like the kind of song you’d be hearing at bar mitzvahs and proms for years.” Around the Arista offices, executives sang the chorus-“shake it like a Polaroid picture”-to one another in the hallways. This song, they all agreed, is going to be huge.

That certainty wasn’t based solely on intuition. At the time, the record business was undergoing a transformation similar to the data-driven shifts occurring at Target and elsewhere. Just as retailers were using computer algorithms to forecast shoppers’ habits, music and radio executives were using computer programs to forecast listeners’ habits. A company named Polyphonic HMI-a collection of artificial intelligence experts and statisticians based in Spain-had created a program called Hit Song Science that analyzed the mathematical characteristics of a tune and predicted its popularity. By comparing the tempo, pitch, melody, chord progression, and other factors of a particular song against the thousands of hits stored in Polyphonic HMI’s database, Hit Song Science could deliver a score that forecasted if a tune was likely to succeed. [206]

The program had predicted that Norah Jones’s Come Away with Me, for instance, would be a hit after most of the industry had dismissed the album. (It went on to sell ten million copies and win eight Grammys.) It had predicted that “Why Don’t You and I” by Santana would be popular, despite DJs’ doubts. (It reached number three on the Billboard Top 40 list.)

When executives at radio stations ran “Hey Ya!” through Hit Song Science, it did well. In fact, it did better than well: The score was among the highest anyone had ever seen.

“Hey Ya!,” according to the algorithm, was going to be a monster hit.

On September 4, 2003, in the prominent slot of 7:15 p.m., the Top 40 station WIOQ in Philadelphia started playing “Hey Ya!” on the radio. It aired the song seven more times that week, and a total of thirty-seven times throughout the month. [207]

At the time, a company named Arbitron was testing a new technology that made it possible to figure out how many people were listening to a particular radio station at a given moment, and how many switched channels during a specific song. WIOQ was one of the stations included in the test. The station’s executives were certain “Hey Ya!” would keep listeners glued to their radios.

Then the data came back.

Listeners didn’t just dislike “Hey Ya!” They hated it according to the data. [208] They hated it so much that nearly a third of them changed the station within the first thirty seconds of the song. It wasn’t only at WIOQ, either. Across the nation, at radio stations in Chicago, Los Angeles, Phoenix, and Seattle, whenever “Hey Ya!” came on, huge numbers of listeners would click off.

“I thought it was a great song the first time I heard it,” said John Garabedian, the host of a syndicated Top 40 radio show heard by more than two million people each weekend. “But it didn’t sound like other songs, and so some people went nuts when it came on. One guy told me it was the worst thing he had ever heard.

“People listen to Top 40 because they want to hear their favorite songs or songs that sound just like their favorite songs. When something different comes on, they’re offended. They don’t want anything unfamiliar.”

Arista had spent a lot of money promoting “Hey Ya!” The music and radio industries needed it to be a success. Hit songs are worth a fortune-not only because people buy the song itself, but also because a hit can convince listeners to abandon video games and the Internet for radio. A hit can sell sports cars on television and clothing inside trendy stores. Hit songs are at the root of dozens of spending habits that advertisers, TV stations, bars, dance clubs-even technology firms such as Apple-rely on.

Now, one of the most highly anticipated songs-a tune that the algorithms had predicted would become the song of the year-was flailing. Radio executives were desperate to find something that would make “Hey Ya!” into a hit. [209]

That question-how do you make a song into a hit?-has been puzzling the music industry ever since it began, but it’s only in the past few decades that people have tried to arrive at scientific answers. One of the pioneers was a onetime station manager named Rich Meyer who, in 1985, with his wife, Nancy, started a company called Mediabase in the basement of their Chicago home. They would wake up every morning, pick up a package of tapes of stations that had been recorded the previous day in various cities, and count and analyze every song that had been played. Meyer would then publish a weekly newsletter tracking which tunes were rising or declining in popularity.

In his first few years, the newsletter had only about a hundred subscribers, and Meyer and his wife struggled to keep the company afloat. However, as more and more stations began using Meyer’s insights to increase their audiences-and, in particular, studying the formulas he devised to explain listening trends-his newsletter, the data sold by Mediabase, and then similar services provided by a growing industry of data-focused consultants, overhauled how radio stations were run.

One of the puzzles Meyer most loved was figuring out why, during some songs, listeners never seemed to change the radio dial. Among DJs, these songs are known as “sticky.” Meyer had tracked hundreds of sticky songs over the years, trying to divine the principles that made them popular. His office was filled with charts and graphs plotting the characteristics of various sticky songs. Meyer was always looking for new ways to measure stickiness, and about the time “Hey Ya!” was released, he started experimenting with data from the tests that Arbitron was conducting to see if it provided any fresh insights.

Some of the stickiest songs at the time were sticky for obvious reasons-“Crazy in Love” by Beyoncé and “Señorita” by Justin Timberlake, for instance, had just been released and were already hugely popular, but those were great songs by established stars, so the stickiness made sense. Other songs, though, were sticky for reasons no one could really understand. For instance, when stations played “Breathe” by Blu Cantrell during the summer of 2003, almost no one changed the dial. The song is an eminently forgettable, beat-driven tune that DJs found so bland that most of them only played it reluctantly, they told music publications. But for some reason, whenever it came on the radio, people listened, even if, as pollsters later discovered, those same listeners said they didn’t like the song very much. Or consider “Here Without You” by 3 Doors Down, or almost any song by the group Maroon 5. Those bands are so featureless that critics and listeners created a new music category-“bath rock”-to describe their tepid sounds. Yet whenever they came on the radio, almost no one changed the station.

Then there were songs that listeners said they actively disliked, but were sticky nonetheless. Take Christina Aguilera or Celine Dion. In survey after survey, male listeners said they hated Celine Dion and couldn’t stand her songs. But whenever a Dion tune came on the radio, men stayed tuned in. Within the Los Angeles market, stations that regularly played Dion at the end of each hour-when the number of listeners was measured-could reliably boost their audience by as much as 3 percent, a huge figure in the radio world. Male listeners may have thought they disliked Dion, but when her songs played, they stayed glued. [210]

One night, Meyer sat down and started listening to a bunch of sticky songs in a row, one right after the other, over and over again. As he did, he started to notice a similarity among them. It wasn’t that the songs sounded alike. Some of them were ballads, others were pop tunes. However, they all seemed similar in that each sounded exactly like what Meyer expected to hear from that particular genre. They sounded familiar-like everything else on the radio-but a little more polished, a bit closer to the golden mean of the perfect song.

“Sometimes stations will do research by calling listeners on the phone, and play a snippet of a song, and listeners will say, ‘I’ve heard that a million times. I’m totally tired of it,’ ” Meyer told me. “But when it comes on the radio, your subconscious says, ‘I know this song! I’ve heard it a million times! I can sing along!’ Sticky songs are what you expect to hear on the radio. Your brain secretly wants that song, because it’s so familiar to everything else you’ve already heard and liked. It just sounds right.”

There is evidence that a preference for things that sound “familiar” is a product of our neurology. Scientists have examined people’s brains as they listen to music, and have tracked which neural regions are involved in comprehending aural stimuli. Listening to music activates numerous areas of the brain, including the auditory cortex, the thalamus, and the superior parietal cortex. [211] These same areas are also associated with pattern recognition and helping the brain decide which inputs to pay attention to and which to ignore. The areas that process music, in other words, are designed to seek out patterns and look for familiarity. This makes sense. Music, after all, is complicated. The numerous tones, pitches, overlapping melodies, and competing sounds inside almost any song-or anyone speaking on a busy street, for that matter-are so overwhelming that, without our brain’s ability to focus on some sounds and ignore others, everything would seem like a cacophony of noise. [212]

Our brains crave familiarity in music because familiarity is how we manage to hear without becoming distracted by all the sound. Just as the scientists at MIT discovered that behavioral habits prevent us from becoming overwhelmed by the endless decisions we would otherwise have to make each day, listening habits exist because, without them, it would be impossible to determine if we should concentrate on our child’s voice, the coach’s whistle, or the noise from a busy street during a Saturday soccer game. Listening habits allow us to unconsciously separate important noises from those that can be ignored.

That’s why songs that sound “familiar”-even if you’ve never heard them before-are sticky. Our brains are designed to prefer auditory patterns that seem similar to what we’ve already heard. When Celine Dion releases a new song-and it sounds like every other song she’s sung, as well as most of the other songs on the radio-our brains unconsciously crave its recognizability and the song becomes sticky. You might never attend a Celine Dion concert, but you’ll listen to her songs on the radio, because that’s what you expect to hear as you drive to work. Those songs correspond perfectly to your habits.

This insight helped explain why “Hey Ya!” was failing on the radio, despite the fact that Hit Song Science and music executives were sure it would be a hit. The problem wasn’t that “Hey Ya!” was bad. The problem was that “Hey Ya!” wasn’t familiar. Radio listeners didn’t want to make a conscious decision each time they were presented with a new song. Instead, their brains wanted to follow a habit. Much of the time, we don’t actually choose if we like or dislike a song. It would take too much mental effort. Instead, we react to the cues (“This sounds like all the other songs I’ve ever liked”) and rewards (“It’s fun to hum along!”) and without thinking, we either start singing, or reach over and change the station.

THE FAMILIARITY LOOP

In a sense, Arista and radio DJs faced a variation of the problem Andrew Pole was confronting at Target. Listeners are happy to sit through a song they might say they dislike, as long as it seems like something they’ve heard before. Pregnant women are happy to use coupons they receive in the mail, unless those coupons make it obvious that Target is spying into their wombs, which is unfamiliar and kind of creepy. Getting a coupon that makes it clear Target knows you’re pregnant is at odds from what a customer expects. It’s like telling a forty-two-year-old investment banker that he sang along to Celine Dion. It just feels wrong.

So how do DJs convince listeners to stick with songs such as “Hey Ya!” long enough for them to become familiar? How does Target convince pregnant women to use diaper coupons without creeping them out?

By dressing something new in old clothes, and making the unfamiliar seem familiar.

III.

In the early 1940s, the U.S. government began shipping much of the nation’s domestic meat supply to Europe and the Pacific theater to support troops fighting in World War II. Back home, the availability of steaks and pork chops began to dwindle. By the time the United States entered the war in late 1941, New York restaurants were using horse meat for hamburgers and a black market for poultry had emerged. [213] Federal officials became worried that a lengthy war effort would leave the nation starved of protein. This “problem will loom larger and larger in the United States as the war goes on,” former president Herbert Hoover wrote to Americans in a government pamphlet in 1943. “Our farms are short of labor to care for livestock; and on top of it all we must furnish supplies to the British and Russians. Meats and fats are just as much munitions in this war as are tanks and aeroplanes.”

Concerned, the Department of Defense approached dozens of the nation’s leading sociologists, psychologists, and anthropologists-including Margaret Mead and Kurt Lewin, who would go on to become celebrity academics-and gave them an assignment: Figure out how to convince Americans to eat organ meats. Get housewives to serve their husbands and children the protein-rich livers, hearts, kidneys, brains, stomachs, and intestines that were left behind after the rib eyes and roast beef went overseas.

At the time, organ meat wasn’t popular in America. A middle-class woman in 1940 would sooner starve than despoil her table with tongue or tripe. So when the scientists recruited into the Committee on Food Habits met for the first time in 1941, they set themselves a goal of systematically identifying the cultural barriers that discouraged Americans from eating organ meat. In all, more than two hundred studies were eventually published, and at their core, they all contained a similar finding: To change people’s diets, the exotic must be made familiar. And to do that, you must camouflage it in everyday garb. [214]

To convince Americans to eat livers and kidneys, housewives had to know how to make the foods look, taste, and smell as similar as possible to what their families expected to see on the dinner table, the scientists concluded. For instance, when the Subsistence Division of the Quartermaster Corps-the people in charge of feeding soldiers-started serving fresh cabbage to troops in 1943, it was rejected. So mess halls chopped and boiled the cabbage until it looked like every other vegetable on a soldier’s tray-and the troops ate it without complaint. “Soldiers were more likely to eat food, whether familiar or unfamiliar, when it was prepared similar to their prior experiences and served in a familiar fashion,” a present-day researcher evaluating those studies wrote. [215]

The secret to changing the American diet, the Committee on Food Habits concluded, was familiarity. Soon, housewives were receiving mailers from the government telling them “every husband will cheer for steak and kidney pie.” [216] Butchers started handing out recipes that explained how to slip liver into meatloaf.

A few years after World War II ended, the Committee on Food Habits was dissolved. By then, however, organ meats had been fully integrated into the American diet. One study indicated that offal consumption rose by 33 percent during the war. By 1955, it was up 50 percent. [217] Kidney had become a staple at dinner. Liver was for special occasions. America’s dining patterns had shifted to such a degree that organ meats had become emblems of comfort.

Since then, the U.S. government has launched dozens of other efforts to improve our diets. For example, there was the “Five a Day” campaign, intended to encourage people to eat five fruits or vegetables, the USDA’s food pyramid, and a push for low-fat cheeses and milks. None of them adhered to the committee’s findings. None tried to camouflage their recommendations in existing habits, and as a result, all of the campaigns failed. To date, the only government program ever to cause a lasting change in the American diet was the organ meat push of the 1940s.

However, radio stations and massive companies-including Target-are a bit savvier.

To make “Hey Ya!” a hit, DJs soon realized, they needed to make the song feel familiar. And to do that, something special was required.

The problem was that computer programs such as Hit Song Science were pretty good at predicting people’s habits. But sometimes, those algorithms found habits that hadn’t actually emerged yet, and when companies market to habits we haven’t adopted or, even worse, are unwilling to admit to ourselves-like our secret affection for sappy ballads-firms risk going out of business. If a grocery store boasts “We have a huge selection of sugary cereals and ice cream!” shoppers stay away. If a butcher says “Here’s a piece of intestine for your dinner table,” a 1940s housewife serves tuna casserole instead. When a radio station boasts “Celine Dion every half hour!” no one tunes in. So instead, supermarket owners tout their apples and tomatoes (while making sure you pass the M &M’s and Häagen-Dazs on the way to the register), butchers in the 1940s call liver “the new steak,” and DJs quietly slip in the theme song from Titanic.

“Hey Ya!” needed to become part of an established listening habit to become a hit. And to become part of a habit, it had to be slightly camouflaged at first, the same way housewives camouflaged kidney by slipping it into meatloaf. So at WIOQ in Philadelphia-as well as at other stations around the nation-DJs started making sure that whenever “Hey Ya!” was played, it was sandwiched between songs that were already popular. “It’s textbook playlist theory now,” said Tom Webster, a radio consultant. “Play a new song between two consensus popular hits.”

DJs, however, didn’t air “Hey Ya!” alongside just any kind of hit. They sandwiched it between the types of songs that Rich Meyer had discovered were uniquely sticky, from artists like Blu Cantrell, 3 Doors Down, Maroon 5, and Christina Aguilera. (Some stations, in fact, were so eager they used the same song twice.)

Consider, for instance, the WIOQ playlist for September 19, 2003:

11:43 “Here Without You” by 3 Doors Down

11:54 “Breathe” by Blu Cantrell

11:58 “Hey Ya!” by OutKast

12:01 “Breathe” by Blu Cantrell

Or the playlist for October 16:

9:41 “Harder to Breathe” by Maroon 5

9:45 “Hey Ya!” by OutKast

9:49 “Can’t Hold Us Down” by Christina Aguilera

10:00 “Frontin’ ” by Pharrell

November 12:

9:58 “Here Without You” by 3 Doors Down

10:01 “Hey Ya!” by OutKast

10:05 “Like I Love You” by Justin Timberlake

10:09 “Baby Boy” by Beyoncé

“Managing a playlist is all about risk mitigation,” said Webster. “Stations have to take risks on new songs, otherwise people stop listening. But what listeners really want are songs they already like. So you have to make new songs seem familiar as fast as possible.”

When WIOQ first started playing “Hey Ya!” in early September-before the sandwiching started-26.6 percent of listeners changed the station whenever it came on. By October, after playing it alongside sticky hits, that “tune-out factor” dropped to 13.7 percent. By December, it was 5.7 percent. Other major radio stations around the nation used the same sandwiching technique, and the tune-out rate followed the same pattern.

And as listeners heard “Hey Ya!” again and again, it became familiar. Once the song had become popular, WIOQ was playing “Hey Ya!” as many as fifteen times a day. People’s listening habits had shifted to expect-crave, even-“Hey Ya!” A “Hey Ya!” habit emerged. The song went on to win a Grammy, sell more than 5.5 million albums, and earn radio stations millions of dollars. “This album cemented OutKast in the pantheon of superstars,” Bartels, the promotion executive, told me. “This is what introduced them to audiences outside of hip-hop. It’s so fulfilling now when a new artist plays me their single and says, This is going to be the next ‘Hey Ya!’”

After Andrew Pole built his pregnancy-prediction machine, after he identified hundreds of thousands of female shoppers who were probably pregnant, after someone pointed out that some-in fact, most-of those women might be a little upset if they received an advertisement making it obvious Target knew their reproductive status, everyone decided to take a step back and consider their options.

The marketing department thought it might be wise to conduct a few small experiments before rolling out a national campaign. They had the ability to send specially designed mailers to small groups of customers, so they randomly chose women from Pole’s pregnancy list and started testing combinations of advertisements to see how shoppers reacted.

“We have the capacity to send every customer an ad booklet, specifically designed for them, that says, ‘Here’s everything you bought last week, and a coupon for it,’ ” one Target executive with firsthand knowledge of Pole’s pregnancy predictor told me. “We do that for grocery products all the time.

“With the pregnancy products, though, we learned that some women react badly. Then we started mixing in all these ads for things we knew pregnant women would never buy, so the baby ads looked random. We’d put an ad for a lawnmower next to diapers. We’d put a coupon for wineglasses next to infant clothes. That way, it looked like all the products were chosen by chance.

“And we found out that as long as a pregnant woman thinks she hasn’t been spied on, she’ll use the coupons. She just assumes that everyone else on her block got the same mailer for diapers and cribs. As long as we don’t spook her, it works.”

The answer to Target and Pole’s question-how do you advertise to a pregnant woman without revealing that you know she’s pregnant?-was essentially the same one that DJs used to hook listeners on “Hey Ya!” Target started sandwiching the diaper coupons between nonpregnancy products that made the advertisements seem anonymous, familiar, comfortable. They camouflaged what they knew.

Soon, Target’s “Mom and Baby” sales exploded. The company doesn’t break out sales figures for specific divisions, but between 2002-when Pole was hired-and 2009, Target’s revenues grew from $44 billion to $65 billion. In 2005, the company’s president, Gregg Steinhafel, boasted to a room full of investors about the company’s “heightened focus on items and categories that appeal to specific guest segments such as mom and baby.

“As our database tools grow increasingly sophisticated, Target Mail has come into its own as a useful tool for promoting value and convenience to specific guest segments such as new moms or teens,” he said. “For example, Target Baby is able to track life stages from prenatal care to car seats and strollers. In 2004, the Target Baby Direct Mail Program drove sizable increases in trips and sales.” [218]

Whether selling a new song, a new food, or a new crib, the lesson is the same: If you dress a new something in old habits, it’s easier for the public to accept it.

IV.

The usefulness of this lesson isn’t limited to large corporations, government agencies, or radio companies hoping to manipulate our tastes. These same insights can be used to change how we live.

In 2000, for instance, two statisticians were hired by the YMCA-one of the nation’s largest nonprofit organizations-to use the powers of data-driven fortune-telling to make the world a healthier place. The YMCA has more than 2,600 branches in the United States, most of them gyms and community centers. About a decade ago, the organization’s leaders began worrying about how to stay competitive. They asked a social scientist and a mathematician-Bill Lazarus and Dean Abbott-for help.

The two men gathered data from more than 150,000 YMCA member satisfaction surveys that had been collected over the years and started looking for patterns. At that point, the accepted wisdom among YMCA executives was that people wanted fancy exercise equipment and sparkling, modern facilities. The YMCA had spent millions of dollars building weight rooms and yoga studios. When the surveys were analyzed, however, it turned out that while a facility’s attractiveness and the availability of workout machines might have caused people to join in the first place, what got them to stay was something else.

Retention, the data said, was driven by emotional factors, such as whether employees knew members’ names or said hello when they walked in. People, it turns out, often go to the gym looking for a human connection, not a treadmill. If a member made a friend at the YMCA, they were much more likely to show up for workout sessions. In other words, people who join the YMCA have certain social habits. If the YMCA satisfied them, members were happy. So if the YMCA wanted to encourage people to exercise, it needed to take advantage of patterns that already existed, and teach employees to remember visitors’ names. It’s a variation of the lesson learned by Target and radio DJs: to sell a new habit-in this case exercise-wrap it in something that people already know and like, such as the instinct to go places where it’s easy to make friends.

“We’re cracking the code on how to keep people at the gym,” Lazarus told me. “People want to visit places that satisfy their social needs. Getting people to exercise in groups makes it more likely they’ll stick with a workout. You can change the health of the nation this way.”

Someday soon, say predictive analytics experts, it will be possible for companies to know our tastes and predict our habits better than we know ourselves. However, knowing that someone might prefer a certain brand of peanut butter isn’t enough to get them to act on that preference. To market a new habit-be it groceries or aerobics-you must understand how to make the novel seem familiar.

The last time I spoke to Andrew Pole, I mentioned that my wife was seven months pregnant with our second child. Pole himself has children, and so we talked a bit about kids. My wife and I shop at Target on occasion, I said, and about a year earlier we had given the company our address, so we could start getting coupons in the mail. Recently, as my wife’s pregnancy had progressed, I’d been noticing a subtle upswing in the number of advertisements for diapers, lotions, and baby clothes arriving at our house.

I was planning on using some of those coupons that very weekend, I told him. I was also thinking of buying a crib, and some drapes for the nursery, and maybe some Bob the Builder toys for my toddler. It was really helpful that Target was sending me exactly the right coupons for what I needed to buy.

“Just wait till the baby comes,” Pole said. “We’ll be sending you coupons for things you want before you even know you want them.”

¹The reporting in this chapter is based on interviews with more than a dozen current and former Target employees, many of them conducted on a not-for-attribution basis because sources feared dismissal from the company or other retribution. Target was provided with an opportunity to review and respond to the reporting in this chapter, and was asked to make executives involved in the Guest Analytics department available for on-the-record interviews. The company declined to do so and declined to respond to fact-checking questions except in two emails. The first said: “At Target, our mission is to make Target the preferred shopping destination for our guests by delivering outstanding value, continuous innovation and an exceptional guest experience by consistently fulfilling our ‘Expect More. Pay Less.’ brand promise. Because we are so intently focused on this mission, we have made considerable investments in understanding our guests’ preferences. To assist in this effort, we’ve developed a number of research tools that allow us to gain insights into trends and preferences within different demographic segments of our guest population. We use data derived from these tools to inform our store layouts, product selection, promotions and coupons. This analysis allows Target to provide the most relevant shopping experience to our guests. For example, during an in-store transaction, our research tool can predict relevant offers for an individual guest based on their purchases, which can be delivered along with their receipt. Further, opt-in programs such as our baby registry help Target understand how guests’ needs evolve over time, enabling us to provide new mothers with money-saving coupons. We believe these efforts directly benefit our guests by providing more of what they need and want at Target-and have benefited Target by building stronger guest loyalty, driving greater shopping frequency and delivering increased sales and profitability.” A second email read: “Almost all of your statements contain inaccurate information and publishing them would be misleading to the public. We do not intend to address each statement point by point. Target takes its legal obligations seriously and is in compliance with all applicable federal and state laws, including those related to protected health information.”