The Power of Habit

Duhigg, then a New York Times reporter, spent years on the science and stories of habit formation: how a Procter & Gamble researcher saved Febreze, how Tony Dungy built a Super Bowl team, how Alcoa's CEO transformed safety culture. The mechanism is the habit loop: cue → routine → reward. You can't kill an old habit, but you can keep the cue and reward and substitute a new routine. Keystone habits trigger cascades of other change.

Eugene Pauly was a 71-year-old retired aerospace engineer in San Diego when a viral encephalitis attack in 1993 destroyed parts of his medial temporal lobe. He could no longer recognize his wife Beverly's photograph or draw a map to his own house. But every afternoon at 5pm, he walked from his living room to the kitchen, opened the cabinet, and made himself dinner. His habits had survived the brain damage that erased his memories.

Duhigg recounts how Alcoholics Anonymous's twelve steps actually work. Researchers initially couldn't explain AA's effectiveness — the program has no controlled-trial design, no licensed therapists. But the cue (stress, social anxiety, end of work day) and reward (relief, social belonging) stayed the same in members' lives — only the routine changed. Drinking gave way to calling a sponsor, sharing in a meeting, repeating slogans. Same loop, new middle.

Procter & Gamble launched Febreze in 1996 as an odor remover. Early results were catastrophic — sales were near zero. Drake Stimson, a P&G researcher, interviewed customers and met a Phoenix housekeeper who described the spritz at the end of cleaning as 'like a little mini-celebration.' The product wasn't actually about removing odors. It was about marking the completion of cleaning. P&G rebuilt the campaign around that moment. Sales doubled within two months and hit $230 million the first year.

Paul O'Neill took over Alcoa in October 1987 and stunned Wall Street analysts at his first speech: he announced his sole priority was worker safety. Analysts called him a moron and recommended their clients sell. But fixing safety required fixing communication, which fixed quality, which fixed efficiency. By the time O'Neill retired in 1999, Alcoa's market value had quintupled, and the company's worker injury rate had dropped to 1/20th the U.S. average.

In 1996, Florida State undergraduates were brought into a lab where freshly baked chocolate-chip cookies cooled on a tray. Half were told to eat the cookies; half were told to resist and eat radishes instead. Both groups were then given an unsolvable geometry puzzle. The cookie-resisters gave up on the puzzle in 8 minutes; the radish-skippers persisted for 19. Roy Baumeister's experiment seemed to prove willpower was a finite muscle that depleted with use.

Duhigg profiles a man named John, an alcoholic who'd relapsed after multiple solo attempts. He maintained six years sober only after he started showing up nightly to a small church basement AA group in his neighborhood. The community didn't grant him willpower — it carried him during the days his own willpower failed. The same pattern shows up across rehabilitation: the support network is the variable that predicts long-term success.

On November 18, 1987, a small fire in a wooden escalator at London's King's Cross Underground station grew into a flashover that killed 31 people. The official Fennell Report found that no single person had failed — the institutional habit of departmental territorialism had. The ticket inspector who saw the fire wasn't allowed to call the safety inspector who handled escalators. The escalator inspector wasn't allowed to override the station manager. Each waited for someone else to act. By then it was too late.