There’s a habit you want out of your life. Maybe it’s the phone in bed, the energy drink at four in the afternoon, the snooze button, or one more episode when you said one more episode an hour ago. Whatever it is, you’ve decided it stops, and the first thing you want to know is how long that takes. The number you’ve probably heard is 21 days. It’s a tidy number. It isn’t the answer.
Here is what the research can honestly give you. In the one study that has timed people day by day while they tried to break a habit, the pull of the old habit faded fastest in the first week or two and then leveled off, at a point that was different for everyone: within days for some, closer to two months for a few, and for some people it barely moved at all.
Underneath that number sits the part that matters more for what you do next. An old habit doesn’t get deleted. It gets outgrown and goes quiet, and the setting that built it can wake it up again. That sounds like bad news, and it’s the opposite, because it means you can stop counting days and start changing the settings. Here is where the 21 days came from, what was actually measured, what makes your own version faster or slower, and what a slip means.
Where the 21 days comes from
A plastic surgeon. In 1960 Maxwell Maltz published a self-help book called Psycho-Cybernetics, and in the preface he asked readers to practice its exercises for at least three weeks before judging them. His reason was what he had noticed in his patients: “It usually requires a minimum of about 21 days to effect any perceptible change in a mental image.” After surgery, he wrote, the average patient took about 21 days to get used to a new face. After an amputation, the feeling that the limb was still there lasted about 21 days. People needed about three weeks in a new house before it felt like home.
Read it again and notice what isn’t in it: a habit. He was describing how long it takes to get used to a change, in people adjusting to a change that had already happened, and he said “minimum”. Sixty years of retelling turned the minimum into the whole answer and a new face into any habit you like. When researchers finally timed habits in everyday life, a flat 21 days is not what they found, for building one or for breaking one.
So if you’re past 21 days and still fighting the thing, nothing has gone wrong with you. The deadline was never real.
How long it takes, as far as anyone has measured it
Most pages answer this question with studies of people building a new habit. Those numbers are real and we’ll get to them, but they answer a different question. The first study to follow people day by day while they tried to break a habit was published in 2025, by researchers in Switzerland, and it’s worth knowing what they did, because the number that gets quoted from it, “1 to 65 days”, means less than it sounds.
194 adults each picked one habit they wanted less of: sitting, unhealthy snacking, alcohol, or smoking. For a week they wrote down the situations that set it off. Then each person chose one situation to work on (coffee in the morning, the sofa after dinner), decided whether they would do something else in that situation, stop themselves, or avoid the situation altogether, and wrote it down as a plan. Most chose to replace the habit with something. From then on, for 84 days, they answered a short questionnaire every evening about how automatic the habit felt in that situation. Most of them were women, most were well educated, and all of them wanted this badly enough to fill in a form every night for three months.
The typical pattern was a drop that was fastest in the first days and then flattened out. Where the drop was clean enough to time, it took anywhere from one day to 65 to level off, and the typical person was there in under two weeks.
Three things the pages quoting that range leave out.
Leveled off is not gone. The authors say so themselves. Habit strength settled at a lower level, not at zero, and that lower level was different for everyone. For some it was still a fair pull.
The clean curve was about one person in five. For the rest, some habits barely moved, a few got stronger over the three months, and some dropped and then climbed back, which from the inside looks a lot like a relapse.
It measured how automatic the habit felt, not how often it happened. Rating how automatic it feels is how researchers usually measure a habit in daily life, and it is a fair measure with known limits, but it means “one day” at the fast end is a result the authors treat with suspicion. They say it doesn’t fit how habits work.
One more finding that matters for you: which habit it was made no difference to the shape of the fade. Sitting habits started out strongest and alcohol habits weakest, and after that the slope was personal. About three quarters of the variation in the data was variation between people, not between habits.
So, a number: for many people the pull is noticeably weaker inside a few weeks. A date when the habit is gone: there isn’t one, because of what an old habit is. And one line on the alcohol and smoking groups, because they were in this study: if the habit you want to break is alcohol, nicotine, other drugs or gambling, an article is the wrong tool for it. Talk to your doctor, who will know what help exists where you live.
Why an old habit never fully goes
Anyone who has moved house knows the first part of this. For a while your hand goes to where the light switch used to be. If you play games, you know the second part too: you remap a key, and in a calm moment your finger finds the new one, but in the middle of a fight it hits the old one. Nobody taught your hand either of those. Doing the same thing in the same place, over and over, taught it. That is all a habit is: a link between a situation and an action, built by repetition, which fires before you’ve decided anything.
What happens to that link when you stop? Research on learning, in animals and in people, has an answer that has held up for decades: it doesn’t get erased. New learning gets laid on top of the old, and from then on the situation decides which of the two comes up. The new learning, “I don’t do that here”, is also more tied to the place it was learned than the old habit was. In plain terms, the “I don’t” you’ve built at your own desk may not travel to a hotel room, your parents’ house at Christmas, or a Friday with the old group. The researchers who review everyday habits say the same: old traces aren’t necessarily replaced, and the old routine or the old setting can wake them.
A study at a cinema shows how much the setting does. People with a strong habit of eating popcorn at the movies were handed a box that was either fresh or a week old. In a cinema, with trailers playing, they ate about as much of the stale popcorn as the fresh, roughly two thirds of the box, while rating it as nasty. People with the same strong habit, put in a meeting room watching music videos, ate less of the stale. And back in the cinema, when they were asked to eat with their other hand, the strong-habit eaters started eating less of the stale popcorn too, a difference the researchers call marginal. Small studies, students, one snack. But the shape is clear: the setting runs the habit, and anything that makes the action clumsy hands the decision back to you.
The same thing turns up in real life. When students transferred to a new university, their exercise, TV and newspaper habits survived only when where they did it and who was around stayed similar; where that changed, their behavior tended to follow what they currently intended, like anyone without the habit. When a charity moved its office to another town, its workers’ old commuting habit didn’t vanish on moving day: it faded over the following weeks while the new one grew. And in a trial with 800 households, a program to change everyday routines worked better for the people who had moved house within the last three months. A change of place does a lot of the work for you.
Building a habit and breaking one run on different clocks
Now the numbers you’ve probably seen. A study in London followed 96 people who each chose a new daily habit, a glass of water after breakfast, a run before dinner, that kind of thing, and rated every day how automatic it felt. For the people whose curve could be timed, it took from 18 days to 254 to level off, and the middle of that group was around 66 days. A 2024 review pooled 20 studies with about 2,600 adults; the four that timed it found the middle person leveling off around two months, the averages higher, and individuals anywhere from four days to nearly a year, and in one of them only about a quarter of the participants got all the way to automatic inside the study. Those are the numbers for building a habit: months rather than weeks, and very personal.
They get quoted as the answer to breaking one because nobody had timed breaking. But they still matter to you, for a reason the cinema study hints at. If you’re replacing the habit with something, which is what most people in the Swiss study chose to do, then two clocks are running at once. The old pull fading, which for many people is a matter of weeks. And the new action becoming automatic, which is a matter of months. The commuters in the office move were on both clocks at the same time: the old habit faded while the new one grew. The stretch in between is where it feels like you’re doing everything by hand. You are. That is what this stage is, not a sign it isn’t working.
How fast the new side gets automatic has a lot to do with how often it gets repeated. In two very large sets of records, hospital staff built a handwashing habit in weeks and gym members built a gym habit in months, and the likely reason is how often each gets repeated: hands get washed many times a shift, the gym happens a few times a week. Worth knowing before you choose what to replace the habit with.
What makes it take longer, or shorter
Four things, as far as the research can show, and one thing that doesn’t seem to matter.
How strong it is now. In diary studies of people trying to stop everyday things (snacking, sleeping in, nail biting), strong habits were plainly harder to control than weak ones, and in a field test, written if-then plans reduced smoking for people with a weak or moderate smoking habit and did nothing for people with a strong one. The thing you’ve done every evening in the same chair for six years is a stronger link than the thing you picked up last winter, and it will take more than a plan on paper.
Whether you still meet the cue. This is the biggest lever. If the situation that fires the habit is gone, the behavior can fall away within weeks, the way the students’ TV habits did; the link waits, but it waits quietly. If you’re still in the same chair at the same hour with the same phone, the link gets a vote every single day.
Whether something new happens at the cue. A situation with nothing attached keeps calling up the old action. A situation with a new action attached starts the other clock. The habit researchers who have looked at what people actually do suggest that stopping yourself works best as a short-term tool while a new response takes hold, not as a permanent way of life.
How tired, stressed or rushed you are. Not a factor in the timeline exactly, but the reason day 40 can feel like day one. Habits are the brain’s default when attention is short, so under stress or on no sleep the automatic answer wins more often. That isn’t your character showing through, it’s the default.
What doesn’t seem to matter: which habit it is. In the Swiss study the four habits faded in the same shape. The differences were in where people started and in the people themselves, not in whether it was a snack or a cigarette.
What to do in the meantime
Everything above points the same way, and it happens to be how we coach habits: don’t fight the link, change the settings. Our rule for a system that works is that it needs very little willpower, and the way to get there is high resistance on the stuff you shouldn’t do and low resistance on the stuff you should do.
In practice that’s unglamorous. The phone charges in the kitchen, so the phone in bed needs a trip down the hall. The energy drink isn’t in the fridge, so the four o’clock one needs a walk to the store (and if you’ve been wondering why it stopped working anyway, that is its own article). The alarm is across the room, so the snooze button needs you on your feet. None of these makes the habit impossible. They turn a reflex into a decision, and a decision is something you can make.
Then decide what the new action is, in advance, and write it as a plan that starts with the situation: when I sit down after dinner, I make tea before anything else. Plans in that form have been tested hundreds of times and they help, more when the person actually wants the change and rehearses the plan, and less against a strong old habit on their own, which is why the settings come first. (The same plans do a lot of the work in how to stop procrastinating.)
For the first weeks, watch for it on purpose. In those diary studies, the strategy that worked best on strong habits wasn’t distraction or leaving the room; it was plain vigilance: thinking “don’t do it” and watching for slips. An experiment afterwards showed why. Monitoring didn’t weaken the habit at all; it strengthened the brake. That is exactly what you need early on and exactly what you can’t keep up forever, which is why it goes together with the new action and the changed settings, not instead of them.
And stop counting days. The line we give clients about weight loss is to detach yourself from the illusion that you need to reach your goal in 7, 21 or 30 days, and it holds here word for word. If you want something to count, count the settings you’ve changed. If you’re waiting to feel motivated first, don’t; the people who look like they have iron self-control tend to lean on habits rather than willpower, which is what you’re building. If you’d rather have a coach build those settings with you, that is what we do at Gains For Gamers.
What a slip means
A question our clients ask a lot, in one form or another: do I need to restart after a bad week? No. Nothing in the research works like a counter that goes back to zero.
In the London study of people building a habit, missing a single day made no real difference to how the habit formed. In the Swiss study of people breaking one, the strength of the habit wobbled from day to day on its way down; a bad evening was part of the data, not the end of it. And the learning research says the quiet part out loud: your good weeks didn’t erase the old link, and one bad night doesn’t erase what you’ve built either. Both are still there, and the setting decides which one runs. What a slip actually tells you is which setting still has the link live. That’s useful. That’s the one to change next.
This is why we coach to 80 percent and not 100. A month at 80 percent beats a perfect month followed by two at 20, and a plan that only works on perfect weeks was built for someone else’s life. Treat the slip the way a scientist treats a result that didn’t go as planned: as information. You either win, or you learn. The only way to lose is to stop altogether.
And if the bad week was a holiday, a move, a hotel, somewhere the old link had never been challenged, you now know why. Back home, in the settings you changed, the quiet version of the habit is still the one you built.
A habit isn’t broken on a date. It’s outgrown in a place. Change the place, and the clock runs in your favor.
Questions we get asked
Is it harder to break a habit than to build one?
Nobody has timed both in the same people, so there’s no clean comparison. What the researchers who have studied both say is that breaking an existing habit takes different and more effortful strategies than starting a new one, because the old link stays in place and fires on its own until something else sits in its spot. In timings, the pull of an old habit typically faded within weeks in the one study that measured it, and a new habit typically took months to become automatic in the studies that timed that. The fade is quicker, the replacement is slower, and the old link outlasts both.
Why does the habit come back when I’m stressed or tired?
Because a habit is the brain’s default when attention is short. When you’re rested and calm, the newer, deliberate response gets a say; under stress or on no sleep, the automatic one wins more often. It’s not a verdict on your willpower, and it’s predictable enough to plan for: the settings you changed are what protect you on those days, which is why we tell clients to have a plan B and a plan C for low-energy days rather than one plan that needs a good day to work.
How do I work out what triggers it?
Before you change anything, spend a few days writing down when it happens: where you were, what time, who was there, what you had just done, how you felt. The Swiss study had people do exactly that for a week before choosing what to work on, and it’s a sensible way to find out what you’re dealing with, because triggers are often not obvious from the inside. In the diary studies, people who tried to control a strong habit by leaving the situation did worse than people who simply watched for it, and the researchers’ likely explanation is that they hadn’t pinned the trigger down. Triggers in these studies were places, times, people and feelings, and once you can name yours you can change it.
Is it better to replace a habit than to just stop it?
Most people in the Swiss study chose to replace theirs, and the study didn’t compare whether replacing beat stopping, so there’s no number here. The reasoning from the learning research is on the side of replacing: a situation with nothing new attached keeps calling up the old action, while a situation with a new action attached starts building the new link. Our own preference is the same, with one addition: make the replacement the lowest-effort option in the room, because a replacement that needs willpower every evening is the old fight with a new name.
Thanks for reading. I’m Daniel, and I wrote this. I’m a certified personal trainer with more than 10,000 hours on the gym floor with clients, and as founder and head coach of Gains For Gamers, I get to combine the best of my two favorite worlds: gaming and fitness.
Sources
Maltz M. Psycho-Cybernetics. Prentice-Hall, 1960. The preface, “Reserve judgment for 21 days”, read in the 2023 edition, which reprints the original preface.
Edgren R, Baretta D, Inauen J. The temporal trajectories of habit decay in daily life: an intensive longitudinal study on four health-risk behaviors. Applied Psychology: Health and Well-Being, 2025;17(1):e12612. The Swiss study: 194 adults, 91 days. DOI.
Bouton ME. Context, ambiguity, and unlearning: sources of relapse after behavioral extinction. Biological Psychiatry, 2002;52(10):976-986. Old learning is not erased by new learning, and the context decides which is retrieved. DOI.
Wood W, Rünger D. Psychology of habit. Annual Review of Psychology, 2016;67:289-314. The review of everyday habits: old traces not necessarily replaced, habits as the default under stress, life transitions as a window. DOI.
Gardner B, Richards R, Lally P, Rebar A, Thwaite T, Beeken RJ. Breaking habits or breaking habitual behaviours? Old habits as a neglected factor in weight loss maintenance. Appetite, 2021;162:105183. DOI.
Neal DT, Wood W, Wu M, Kurlander D. The pull of the past: when do habits persist despite conflict with motives? Personality and Social Psychology Bulletin, 2011;37(11):1428-1437. The cinema popcorn studies. DOI.
Wood W, Tam L, Witt MG. Changing circumstances, disrupting habits. Journal of Personality and Social Psychology, 2005;88(6):918-933. 115 students transferring university. DOI.
Walker I, Thomas GO, Verplanken B. Old habits die hard: travel habit formation and decay during an office relocation. Environment and Behavior, 2015;47(10):1089-1106. DOI.
Verplanken B, Roy D. Empowering interventions to promote sustainable lifestyles: testing the habit discontinuity hypothesis in a field experiment. Journal of Environmental Psychology, 2016;45:127-134. 800 households; the three-month window. DOI.
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;40(6):998-1009. The London study: 96 volunteers, 12 weeks. DOI.
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;12(23):2488. DOI.
Buyalskaya A, Ho H, Milkman KL, Li X, Duckworth AL, Camerer C. What can machine learning teach us about habit formation? Evidence from exercise and hygiene. Proceedings of the National Academy of Sciences, 2023;120(17):e2216115120. DOI.
Quinn JM, Pascoe A, Wood W, Neal DT. Can’t control yourself? Monitor those bad habits. Personality and Social Psychology Bulletin, 2010;36(4):499-511. The diary studies and the monitoring experiment. DOI.
Webb TL, Sheeran P, Luszczynska A. Planning to break unwanted habits: habit strength moderates implementation intention effects on behaviour change. British Journal of Social Psychology, 2009;48(3):507-523. DOI.
Sheeran P, Listrom O, Gollwitzer PM. The when and how of planning: meta-analysis of the scope and components of implementation intentions in 642 tests. European Review of Social Psychology, 2025;36(1):162-194. DOI.
Galla BM, Duckworth AL. More than resisting temptation: beneficial habits mediate the relationship between self-control and positive life outcomes. Journal of Personality and Social Psychology, 2015;109(3):508-525. DOI.
Gardner B, Lally P, Wardle J. Making health habitual: the psychology of ‘habit-formation’ and general practice. British Journal of General Practice, 2012;62(605):664-666. Breaking existing habits needs different, more effortful strategies than making new ones. DOI.
This page is general information from coaches, not medical advice. If you have a medical condition or an injury, or take medication, talk to your doctor before you change your training or diet.