How to make a habit stick: what habit studies show

To make a habit stick, pick one small action, tie it to a moment that already happens every day, and repeat it on that cue until it starts without a decision. Write the plan as “after X, I will Y,” a format that had a medium-to-large effect on reaching goals across 94 tests. Expect it to take about two months, often longer, and don’t restart the count after a missed day, because in the best-known habit study one miss did not materially slow the process.

That is the short answer. The longer one is more useful, because most habit advice skips the parts of the studies that explain why your last attempt faded around week three.

Why do habits fail even when you really want them?

Wanting something and doing it daily rely on different machinery. A habit, in the research sense, is an action that a familiar situation triggers before you’ve thought about it. The kettle clicks off and your hand reaches for the mug. Wanting comes into it at the start, and then the situation takes over.

Wendy Wood and her colleagues showed this in two diary studies published in 2002. People reported what they were doing and thinking every hour. During behavior they did almost daily in the same setting, their minds were usually somewhere else, and they felt less stressed than during unfamiliar actions. Habits free up attention. In a 2016 review, Wood and Dennis Rünger describe habits as the brain’s efficient default, the thing you do when you aren’t steering.

This is why motivation alone tends to fail. In a 2002 trial by Sarah Milne, Sheina Orbell and Paschal Sheeran, 248 students were split into three groups. A control group simply reported their exercise. A second group got a persuasive message about the risks of inactivity and the benefits of exercise. It raised their intentions to exercise but didn’t significantly change how much they exercised. A third group got the same message plus one extra step. They wrote down when and where they would exercise in the coming week. The authors describe the effect of that combination on actual exercise as “dramatic.” Motivation changed what people meant to do. The plan changed what they did.

So the job is to stop relying on how much you want it, and build a cue that does the remembering for you.

How long does it take for a habit to stick?

Longer than 21 days, and less predictably than 66.

The 66-day figure comes from a 2010 study by Phillippa Lally and colleagues at University College London. Ninety-six volunteers chose one daily eating, drinking or exercise behavior, like a glass of water with lunch, and rated every day how automatic it felt, for 12 weeks. Automaticity rose quickly at first and then flattened out, like a learning curve. The time to reach near-peak automaticity ranged from 18 to 254 days. Of the 82 people who gave enough data, 39 had curves that fit the model well, and the often-quoted average of about 66 days comes from that group. So the famous number describes fewer than half the volunteers, and the spread around it is huge.

Later work tells the same story with more people:

• A 2021 randomized trial by Jan Keller and colleagues followed 192 adults building a nutrition habit for 84 days. Among those who formed a habit, the median time to peak automaticity was 59 days. A 2024 review notes that only 23% of that trial’s participants reached the threshold for a formed habit at all.
• A 2023 PNAS study by Anastasia Buyalskaya, Katy Milkman, Angela Duckworth and colleagues used machine learning on more than 12 million gym check-ins and 40 million hospital handwashing events. Gym habits typically took months to form. Handwashing at work took weeks.
• Navin Kaushal and Ryan Rhodes followed 111 new gym members for 12 weeks and found that about four workouts a week for six weeks was the minimum to establish an exercise habit.

The 2024 systematic review by Ben Singh and colleagues pulled 20 studies and 2,601 participants together. Across the four studies that measured timing, median times ran from 59 to 66 days and individual times from 4 to 335 days. The authors conclude that health habits usually take two to five months, and they call the 21-day idea refuted. Their own caveat matters too. Most included studies carried a high risk of bias, so treat those numbers as a range, not a schedule.

Two things follow. First, a hard week three is normal, because you are still on the steep part of the curve. Second, don’t measure progress by how many days in a row you’ve done it. Measure it by whether the action takes less pushing than it did a month ago.

And if you miss a day, keep going. In Lally’s study, missing a single opportunity did not materially affect habit formation. Consistency over the whole stretch mattered more. People who performed the behavior more consistently had cleaner habit curves.

What is an implementation intention, and does it work?

An implementation intention is a plan in the form “When situation X happens, I will do Y.” The psychologist Peter Gollwitzer named it in the 1990s, and it is the best-tested tool in the habit toolkit.

In a 2006 meta-analysis, Gollwitzer and Paschal Sheeran pooled 94 independent tests and found a medium-to-large effect on reaching goals, d = 0.65. That number is a standardized difference between groups, not a percentage. In plain terms, people who wrote if-then plans did noticeably better than people who only set the goal.

The plan works better when it is specific. Katherine Milkman’s team mailed flu-clinic reminders to 3,272 employees at a large company in a 2011 PNAS field study. Vaccination in the control group was 33.1%. Asking people to write down the date they would go made no reliable difference. Asking them to write down the date and the time raised vaccination by 4.2 percentage points. One extra blank on a form moved behavior, and only the more precise version did.

Be honest about what this evidence covers. Most implementation intention studies measure whether people did the thing over days or weeks. That is the first half of forming a habit. Whether it becomes automatic depends on what happens next, which is the same cue, over and over.

A good if-then plan has three parts:

1. A cue you can see or feel, such as “after I pour my first coffee,” rather than “in the morning.”
2. One action, small enough that you could do it on a bad day.
3. A place, if the action needs one: “at the kitchen counter.”

Should you tie a habit to a routine or to a time of day?

You’ll read everywhere that attaching a new habit to an existing routine works best. BJ Fogg’s book Tiny Habits (2019) built a whole method around this: an “anchor” moment you already do, a tiny new behavior right after it, and a small celebration. Many writers call the same move habit stacking.

The evidence is more mixed than the advice. In Keller’s 2021 trial, people who linked their nutrition habit to a daily routine and people who linked it to a clock time formed habits at similar rates. What predicted automaticity in both groups was how often people carried out the plan when the cue came.

Where the routine is anchored does seem to matter, though. In an exploratory study of 50 people by Gaby Judah, Benjamin Gardner and Robert Aunger, those told to floss after brushing formed stronger flossing habits than those told to floss before, and they still flossed more at eight months. One reading is that brushing is a clean ending, and an ending is a better trigger than a beginning. Time of day may matter as well. In a 2017 trial by Marion Fournier and colleagues, a stretching habit done on waking was estimated to become automatic in about 106 days, against about 154 days for the same stretch done at bedtime. That was 48 students, and the curves were extrapolated past the 90-day study, so hold it loosely.

Here is how I’d read it. The type of cue matters less than whether it shows up every day in the same form. A routine anchor has one practical advantage. It travels with you. “After I put my keys in the bowl” still works when you come home at 9 p.m. instead of 6.

Why phone reminders can stop a habit from forming

A reminder looks like help, and for remembering, it is. For automaticity, it can get in the way.

Katarzyna Stawarz, Anna Cox and Ann Blandford tested this in a four-week study presented at CHI 2015. Reminders helped people repeat the behavior but hindered habit development. Event-based cues, where people tied the action to something already happening in their day, increased automaticity. The same team reviewed 115 habit apps and found that most were built around tracking and alerts, with little support for event-based cues.

The likely reason is simple. If the buzz in your pocket is what starts the action, the buzz becomes the cue. Take the phone away and the habit has nothing to hold onto.

If the buzz in your pocket is what starts the action, the buzz becomes the cue.

Reminders still have a place early on, while you are learning to notice the anchor. Use them as training wheels. Set the reminder for just after the anchor moment, and turn it off once you catch yourself doing the action before it fires.

How small should a new habit be?

Smaller than feels worthwhile.

Benjamin Gardner, Phillippa Lally and Jane Wardle, writing guidance for family doctors in 2012, advise starting with small, manageable changes, because early failure discourages people and early success builds confidence for the next change. In Kaushal and Rhodes’s gym study, low complexity of the exercise routine predicted habit formation. And Singh’s 2024 review noticed that simple, repetitive behaviors with clear cues, like flossing, tended to show larger effects than complex ones like eating well, though the authors say this could reflect better interventions rather than easier habits.

Small also protects you on bad days. A rule like “two minutes of stretching after I brush my teeth” survives a migraine, a late shift and a hard phone call. “A 30-minute workout every morning” survives about one of those. You can always do more once the action has started. The habit you are building is the starting.

Does it help if you enjoy the habit?

Yes, and probably more than most advice admits.

Kaitlin Woolley and Ayelet Fishbach studied New Year’s resolutions, workouts and study sessions in 2017. People said they did these things for long-term rewards like health or grades. But what predicted whether they kept going was immediate enjoyment. Delayed rewards predicted setting the goal. Immediate ones predicted sticking with it. Kaushal and Rhodes found the same pattern for exercise, where enjoyment of the workouts predicted growth in habit strength. In the flossing study, more positive attitudes toward flossing predicted stronger habits regardless of how often people flossed.

Rewards bolted on from outside work less well. In Stawarz’s study, positive reinforcement had little effect on automaticity. So if you can choose between a version of the habit you like and a “better” version you dread, choose the one you like. A walk with a podcast you look forward to beats a run you avoid.

Is there a best time to start a new habit?

Possibly, and it’s when your routines are already shaken up.

Bas Verplanken and Deborah Roy tested this “habit discontinuity” idea in a 2016 field experiment with 800 people. A program encouraging sustainable behavior worked better for people who had recently moved house than for people who hadn’t. The window of opportunity seemed to last about three months after the move. The effect was modest, and the behaviors were self-reported, but the idea is sound: old cues are gone, so new ones have less competition.

A move, a new job, a new semester, a breakup or the first weeks with a new baby all clear out old cues. They are hard times to do anything extra. They are also good times to decide where one new small action will live.

A short method you can use tonight

1. Choose one habit. No study I know of settles how many habits you can build at once, but each one needs its own daily repetitions, and repetition is the part that counts. One is easier to protect.
2. Shrink it until it takes under two minutes and you could do it on your worst day.
3. Find an anchor that happens every day, in the same form, and ends cleanly: the kettle clicking off, your keys hitting the bowl, brushing your teeth.
4. Write the plan down as “After I [anchor], I will [action] at [place].”
5. Make the action pleasant if you can. Choose the version you’ll look forward to.
6. If you use a reminder, set it for just after the anchor and drop it once the anchor alone starts to work.
7. When you miss a day, do it on the next anchor. Don’t double up and don’t restart a count.
8. Check in after a month by asking whether it takes less effort than it did, and expect the answer to keep changing for another month or two.

Writing the plan down, and writing a line at night about whether the anchor showed up, also gives you a record you can learn from. If you already keep an end-of-day journal, one sentence about the habit is enough. In Auserene, the evening conversation reads your notes from the day before it asks anything, so a half-line like “skipped the stretch, late meeting” can become part of how you look back on the day. For the wider case for writing things down, see what the research says about journaling.

What should you do when a habit isn’t sticking?

Read your misses like data. They usually fall into one of four patterns.

The cue is vague. “In the evening” is a time, not a moment. Swap it for something you can point to.

The action is too big. If you skip it on tired days, it needs to get smaller. You can always do more once you’ve started.

The context keeps changing. Lally’s study linked more consistent performance to cleaner habit curves. If your mornings vary wildly, anchor to something that travels with you, like your first coffee or arriving home.

You don’t like it. Consider a different version of the same goal. A habit you dread will keep needing willpower long after the two-month mark.

Some habits are tied up with how you feel: sleeping earlier, drinking less, getting out of bed when your mood is low. If a habit like that keeps slipping, the problem may not be the plan. It can be worth talking to a doctor or a therapist, and if you are struggling right now, these crisis resources list people you can reach today. Writing down what was going on when a habit slipped, and putting a name to the feeling, can also show you patterns that a checkbox never will. In Auserene, that’s what the notes you write during the day are for.

The habit numbers above, with sources, are also collected in our statistics roundup.

What actually makes a habit stick

The studies agree on less than the advice suggests, and the overlap is small enough to hold in one hand. Repeat a small action on the same cue. Keep going after a miss. Pick a version you enjoy. Give it months.

The rest is texture. Routine or clock, morning or night, tiny or small, these choices change the odds a little. The thing that changes them a lot is the boring part, doing it again tomorrow after the kettle clicks, until one day you notice you already did.