A habit is a shortcut your setting triggers, and this is how one forms

Key takeaways

• In hour-by-hour diaries, about 43% of what people did happened almost every day, usually in the same place, which shows how much of daily life runs on habit¹.
• The “21 days” rule traces to a 1960 self-help book by a plastic surgeon describing how patients got used to a new face, not how habits form³.
• In a well-known study, new daily health habits took a median of 66 days to level off, with a range from 18 days to a projected 254²⁴.
• Missing one day barely dented progress in that study, but in records from about 30,000 gym members, longer gaps between visits made a habit less likely²⁶⁷.
• In an English field experiment, a green-living program changed people’s routines only in roughly the first three months after a move¹⁰.

A review by psychologists Wendy Wood and Dennis Rünger describes a telling 2011 experiment. Researchers gave people stale popcorn in a local movie theater¹.

People with a strong habit of eating popcorn at the movies ate more of it than people without that habit, even though they didn’t like it¹.

Then the researchers ran the test in a campus meeting room, where people watched music videos instead. There, people with strong movie-popcorn habits followed their taste buds and ate little of the stale batch¹.

The popcorn hadn’t changed. The setting had.

That study captures decades of habit research. A habit isn’t a preference or a promise. It’s an action your surroundings switch on, often before you’ve decided anything.

If you’re building a routine, such as a morning walk or herbal tea at bedtime, that matters more than any motivation trick. Below, we cover what research shows about how habits form and how long it takes.

A habit is a shortcut that your surroundings switch on

Psychologists define a habit as a link in memory between a setting and an action. You build the link by doing the same thing in the same situation again and again. Once it’s strong, the setting triggers the action with little thought¹.

Habits have two telltale features, according to Wood and Rünger’s 2016 review in the Annual Review of Psychology. Recurring cues set them off, and they don’t bend much when your goals change in the short term¹. The popcorn eaters’ goal was to enjoy a snack. Their habit didn’t care.

A cue can be a place, a time of day, a person, a mood or the action that comes just before. In studies where people logged what they were doing once an hour, about 43% of their actions happened almost daily and usually in the same place¹.

In a 1998 analysis of 64 studies, people’s intentions predicted things they did now and then better than things they did all the time¹. And on days when people’s plans to exercise were weaker than usual, those with exercise habits fell back on the habit¹.

So the habit system cuts both ways. A bad habit can beat your best intentions. A good habit can carry you on days your motivation is low.

Your brain hands repeated actions to a different circuit

Deep in the brain sit structures called the basal ganglia, which work in loops with the brain’s wrinkled outer layer¹.

Both loops pass through a structure called the striatum. One supports goal-directed choices, where you weigh what you want and how to get it. The other, which runs through a different part of the striatum and connects with the brain’s movement and touch areas, supports habits¹.

As a habit grows stronger, brain activity shifts from the goal loop toward the habit loop¹. The brain chemical dopamine acts like a teacher along the way. When a result turns out better than expected, the dopamine signal helps stamp in the link between cue and action¹.

Russell Poldrack, a neurobiologist then at the University of Texas at Austin, told NIH News in Health in 2012 that good and bad habits run on the same brain machinery⁹. He explained that dopamine strengthens pleasure-linked habits and fuels the craving to repeat them, which the NIH writers said makes those habits harder to break⁹.

Much of this map comes from animals. In rats, damaging the habit part of the striatum kept lever-pressing habits from forming. Switching it off after a habit had formed made the rats care about the reward again¹.

People have been harder to study. In 2018, a team of researchers reported five separate attempts to create rigid habits in healthy adults through extra lab training, including two repeats of the original human study. All five failed⁸. The authors concluded that habits are hard to produce in a lab in healthy people⁸.

That doesn’t mean human habits are imaginary. The brain story is clearest in animals, and real-world records of people, like those below, fill in the rest.

The 21-day rule came from a surgeon’s observations

The popular 21-day figure traces to Maxwell Maltz, a plastic surgeon who wrote the 1960 self-help book Psycho-Cybernetics³.

Maltz’s preface leaned on what he had seen around him. Surgery patients, he wrote, seemed to need roughly three weeks to accept a changed face. He gave similar timelines for the “phantom limb” sensations of amputees and for a new house to start feeling like home³. He then generalized: in his view, changing a mental image takes at least about three weeks³.

Those were anecdotes, not an experiment³. Benjamin Gardner, a habit researcher then at University College London, offered one theory in 2012 for how the number spread. Readers may have confused habituation, the process of getting used to something, with forming a habit³.

Colin Camerer of the California Institute of Technology, an author of the gym study described below, put it bluntly in 2023: the 21-day estimate wasn’t based on science⁶.

Real habits take weeks to months, and progress slows as it builds

The best-known test came from Phillippa Lally and colleagues at University College London, published in 2010²³. They recruited 96 people, and 82 finished the study².

Volunteers each picked one new healthy action and paired it with something that already happened daily. Choices ranged from a piece of fruit at lunch to a 15-minute run before dinner or 50 sit-ups once the morning coffee was done². Every day for 84 days, they went online to say whether they’d done it and to rate how automatic it felt².

Progress traced a curve: big jumps in automaticity early on, then smaller and smaller gains until scores flattened².

The median was 66 days, meaning half the group got close to their plateau sooner and half took longer⁴. Press summaries often call it an average, but a 2024 review of the study reports it as a median²⁴.

The range ran from 18 days to a projected 254 days. That top figure was an estimate, since some people’s scores were still rising when the 84 days ended².

By day 84, roughly half the volunteers still fell short of the score the researchers used to call something a habit². Exercise habits needed about one and a half times as long to flatten out as new eating or drinking habits².

A 2024 review of 20 studies found only four that measured time to habit. Typical times ran about two months, and the authors advised expecting at least two to five months⁴.

The biggest real-world test used no surveys at all. Researchers at Caltech and other universities used machine learning, a computer method that finds patterns in huge data sets, to study records from about 30,000 members of the 24 Hour Fitness chain over four years⁵⁶. They also studied badge data from more than 3,000 hospital workers who were supposed to wash their hands⁶⁷.

Gym habits took about six months on average to form, the Caltech team reported⁶. Handwashing habits formed within weeks, and most hospital staff got there after nine or 10 shifts⁷. Lead author Anastasia Buyalskaya, now at HEC Paris, noted that gym trips happen less often and involve more steps⁷. Her summary: there’s no magic number of days⁷.

One missed day won’t undo you, but long gaps can

If you miss a day, you haven’t started over. In Lally’s study, a single missed day had little effect on how automatic the action became².

Frequent misses were another story. Skips added up even when they were scattered, and people who missed often topped out at a lower level of automaticity².

The gym records point the same way. For about three-quarters of members, the time since their last visit helped predict whether they’d go again, and longer gaps made a habit less likely⁶⁷.

The day of the week mattered, too. About 69% of members were more likely to show up on the same days each week, with Monday and Tuesday the most popular. Time of day didn’t predict their habits⁶.

In hospitals, the strongest predictor of washing up was whether a worker had done it on the previous shift⁷.

Gym members with the most predictable routines responded less to a program designed to boost their visits⁵⁷. Once a habit forms, outside rewards seem to matter less. That’s good news for a habit you want to keep, and a warning about one you want to break.

A big life change opens a short window

Psychologist Bas Verplanken of the University of Bath calls this idea the habit discontinuity hypothesis. When a life change knocks out your old cues, you’re more open to new information, so it’s a good time to start something new¹⁰¹¹.

His team tested it with 800 adults in Peterborough, England¹⁰. Half had moved within the past six months, and half were matched people who hadn’t moved¹⁰. Some areas got a program with a personal interview, a bag of eco-friendly goods, a local directory and newsletters¹⁰.

The program aimed at everyday routines such as shorter showers, more recycling, washing clothes at cooler temperatures and fewer short car trips¹¹. It worked best for people who had moved most recently. The window lasted about three months after a move¹⁰.

The effects were small, and people reported their own behavior, which can be biased¹⁰. An earlier Bath study found something similar for commuting: university employees who cared about the environment drove to work less if they had recently moved¹².

There’s a catch. Wood and Rünger note that a move doesn’t necessarily erase old habit memories. When the old cues come back, so can the old habit¹. Our article on changing long-held habits covers how to handle that.

So if you’ve just moved or changed jobs, start the routine you want now, before new defaults settle in.

Build a new habit the way the research describes

Print this blueprint or copy it into a notebook, and fill in each blank.

1. Pick one action you chose and like. Self-chosen and enjoyable habits grew stronger in the 2024 review⁴. Write it in a few words: “a 10-minute walk,” not “get fit.”
2. Tie it to a daily cue in the same place. Lally’s volunteers linked each action to something that already happened every day². My cue: __. My place: ____.
3. Keep the days steady. Same-day patterns predicted gym habits better than the time on the clock⁶. My days: ______.
4. Count repetitions, not perfect weeks. One miss barely mattered in Lally’s study². Make a simple mark each time you do it.
5. Don’t let a gap grow. Long gaps predicted lost habits in the gym data⁶. A popular rule of thumb, “never miss twice,” fits this finding, though nobody has tested it directly.
6. Use a fresh start if you have one. If you’ve moved or started a new job in the past three months, begin now¹⁰.
7. Check automaticity once a month. Ask yourself these four questions, written for this article, and count your yes answers.
   – Do I sometimes start before I’ve decided to?
   – Would skipping it feel odd?
   – Do I do it without a reminder?
   – Can I do it while thinking about something else?

Three or four yeses suggest the habit is taking hold. Zero or one at two months is common, so keep going rather than quitting.

Stretch	What usually happens	What to do
Weeks 1 to 3	The fastest gains, but you still need reminding2	Protect the cue and keep the action small
Weeks 4 to 10	Gains slow and progress can feel flat2	Keep repeating; the median plateau came near day 664
Months 3 to 6	Complex habits like gym trips may still be building6	Judge by your monthly check, not by the calendar
After a slip	One missed day barely matters2	Restart at the very next cue

If a habit you want to change involves alcohol, drugs, tobacco or eating that feels out of control, it may be more than a habit. Talk with a clinician. Our article on changing long-held habits lists free help lines.

What the evidence shows

Claim	What studies found	Evidence level
Habits are cued by context and built by repetition	Decades of studies, hour-by-hour diaries and objective gym and hospital records15	Strong evidence
A habit forms in 21 days	Traces to anecdotes in a 1960 book; tracked studies find longer, widely varying times34	Evidence says no
About two months is typical	Medians near two months in a few small studies, with a huge spread and many at high risk of bias24	Promising but early
Complex habits take longer	Exercise took 1.5 times longer in one study; months versus weeks in big data267	Promising but early
One missed day doesn’t erase progress	One self-report study; long gaps hurt in gym records26	Promising but early
Life changes make new habits easier	Small effects in field studies; a window of about three months1012	Promising but early
The brain shifts control to habit circuits	Clear in animals; hard to reproduce in human lab tests18	Promising but early in people

Your body is built to learn routines

Natural health starts from respect for the body’s ability to adapt. Habit is part of that design. Your brain hands repeated actions to a system that runs them with little thought¹.

Seen that way, building a habit is less about forcing yourself and more about teaching your body a rhythm. Many lasting traditions work like this: tea at the same hour, a walk after supper, a quiet moment before a meal. Each one ties a small act to a cue that shows up every day.

The research adds patience to that wisdom. Progress comes fast at first and then slows², like a seedling that shoots up and then spends weeks building roots. Flat weeks aren’t failure.

It also adds a caution. The same machinery that builds a morning walk builds the evening scroll⁹. Choosing your cues on purpose is one of the most self-reliant things you can do for your health, and it costs nothing.

Questions readers ask

The bottom line

A habit is an action your surroundings switch on after many repetitions. It forms slowly, with fast early gains and a long, flat stretch, and the time it takes varies from a few weeks to most of a year. The 21-day rule came from a surgeon’s anecdotes, not research.

To build one, choose a small action you like, tie it to a daily cue in a steady place, count repetitions instead of chasing perfect weeks and don’t let gaps grow. If life has just shaken up your routine, that’s your opening.

Related reading

• Old habits run on cues rather than willpower, and you can rewire them
• Behavior science tested its own tricks at scale, and rewards and repetition held up
• Habit stacking is barely tested, though the idea underneath it holds up better
• A healthy identity supports good habits, while all-or-nothing food rules work against them
• Feeling better after trying something isn’t a test, but you can run one

Sources

1. Wood W, Rünger D. “Psychology of Habit.” Annual Review of Psychology, 2016. dornsife.usc.edu
2. Lally P, van Jaarsveld CHM, Potts HWW, Wardle J. “How are habits formed: Modelling habit formation in the real world.” European Journal of Social Psychology, 2010, as summarized in Jarrett C. “How to form a habit.” BPS Research Digest, British Psychological Society, 2010. bps.org.uk
3. Gardner B, Meisel S. “Busting the 21 days habit formation myth.” Health Behaviour Research Centre blog, University College London, 2012. blogs.ucl.ac.uk
4. Singh B, Murphy A, Maher C, Smith AE. “Time to Form a Habit: A Systematic Review and Meta-Analysis of Health Behaviour Habit Formation and Its Determinants.” Healthcare, 2024. pubmed.ncbi.nlm.nih.gov
5. Buyalskaya A, Ho H, Milkman KL, Li X, Duckworth AL, Camerer CF. “What can machine learning teach us about habit formation? Evidence from exercise and hygiene.” Proceedings of the National Academy of Sciences, 2023. authors.library.caltech.edu
6. California Institute of Technology. “No magic number for time it takes to form habits.” News release via ScienceDaily, April 17, 2023. sciencedaily.com
7. Buyalskaya A. “Changing Your Habits Depends on Context, Not Just Motivation.” HEC Paris Knowledge, January 25, 2024. hec.edu
8. de Wit S, Kindt M, Knot SL, et al. “Shifting the balance between goals and habits: Five failures in experimental habit induction.” Journal of Experimental Psychology: General, 2018. dare.uva.nl
9. National Institutes of Health. “Breaking Bad Habits: Why It’s So Hard to Change.” NIH News in Health, January 2012. newsinhealth.nih.gov
10. Verplanken B, Roy D. “Empowering interventions to promote sustainable lifestyles: Testing the habit discontinuity hypothesis in a field experiment.” Journal of Environmental Psychology, 2016. pure.ulster.ac.uk
11. University of Bath. “Capitalise on life transitions to instil change.” University of Bath announcement, January 27, 2016. bath.ac.uk
12. Verplanken B, Walker I, Davis A, Jurasek M. “Context change and travel mode choice: Combining the habit discontinuity and self-activation hypotheses.” Journal of Environmental Psychology, 2008. researchportal.bath.ac.uk