How long does it take to form a habit: 21 or 66 days?

October 13, 2026 · 6 min read

"Twenty-one days and the habit is yours." Coaches, bloggers and apps repeat this number constantly. The problem is that it has no scientific basis. If you have ever honestly kept going for three weeks and then watched everything fall apart, it was not you. It was the wrong expectation.

Let's look at where 21 days came from, what the 66-day study actually showed, and how to use that knowledge so a new action really becomes automatic.

Where the 21-day myth came from

In 1960 the plastic surgeon Maxwell Maltz published Psycho-Cybernetics. Maltz noticed that his patients usually needed at least three weeks to get used to their new faces after surgery.

That was a doctor's observation, not an experiment. And it was about the minimum time it takes to adjust to a change, not how long it takes to build new behaviour. Retold dozens of times, "at least 21 days" became "exactly 21 days". The number was convenient: short, round, and it promised fast results. That is why it stuck.

What the research actually showed

In 2009 psychologist Phillippa Lally and her colleagues at University College London ran one of the first studies of how habits form in real life rather than in a lab. The results were published in the European Journal of Social Psychology.

How it worked:

96 volunteers each chose one new behaviour related to eating, drinking or exercise, such as "eat a piece of fruit with lunch" or "run for 15 minutes after dinner".
The behaviour was tied to a specific situation: a time, a place or a preceding action.
For 12 weeks participants recorded every day whether they had done it and how automatic it felt — effortless, "on its own".

The main findings:

On average, automaticity reached a plateau after about 66 days.
The range was huge: from 18 to 254 days. Simple behaviours (a glass of water at breakfast) became automatic faster; harder ones (50 sit-ups before morning coffee) took longer.
Missing a single day did not derail the process. People who occasionally missed a day kept moving towards automaticity almost as well as those who never missed.

A habit isn't "holding out for N days". It's the point where the situation, not willpower, starts the action.

Why 66 is not the new magic number

It's important not to fall into the same trap twice. 66 days is the median from one particular study, not a promise. For some people a month will be enough; others will need half a year.

But the number does have practical value. It sets more honest expectations:

Your expectations become realistic. If you know it's too early to expect ease on day 21, you won't conclude that "something is wrong with me".
The horizon is long enough. 66 days is more than two months. In that time a goal genuinely moves rather than just being "tried out".
There's a natural checkpoint. After 66 days you can look honestly at what has become automatic, what hasn't, and what to keep.

What speeds up habit formation

Research on habits and intentions points to a few rules that genuinely work.

1. Tie it to a situation

A habit is a link: "situation → action". So the action needs a time and a place, or should hook onto something you already do. Not "I'll read more", but "at 9:30 pm, when I get into bed, I read 10 pages". More on this in the article on if–then plans.

2. Keep it small

The simpler the action, the faster it becomes automatic. Start with a version you couldn't fail to do even on a bad day, and increase it later.

3. Repeat it in the same conditions

Automaticity grows from repetition in a similar context. If you run in the morning one day, the evening the next and at lunch the day after, the habit has nothing to latch onto.

4. Stay calm about missed days

Since one missed day doesn't break the process, the key is not to turn it into "I've blown it anyway". The danger isn't the missed day — it's the second and third in a row. More in the article Missed a day? What to do.

How Ritm is built around this

The Ritm program is built on these findings:

66 days split into four phases: "Foundation" (days 1–14), "Rehearsal" (15–28), "Leverage" (29–42) and "Scale" (43–66).
One action a day — with a time, a place and a criterion for "what will be done when I finish".
A plan for obstacles: what I'll do if something gets in the way.
A missed day resets nothing. The counter counts closed days, not calendar days, and the program waits for you on the same day.

Frequently asked questions

So is 21 days a complete lie?

It's a distortion rather than a lie. 21 days was an observation about the minimum time it takes to adjust to change, and it turned into "the time it takes to form a habit". For simple actions three weeks may be enough, but don't count on it.

Why does a habit form in 18 days for one person and 254 for another?

It depends on how hard the action is, how stable the situation you do it in is, and how pleasant the action is in itself. A glass of water in the morning is easier than a daily workout.

What if the action still takes effort on day 66?

That's normal for harder habits. Look at what gets in the way: an unstable time, too much volume, no plan for obstacles. Make the action smaller and keep going — the number in the study was a median, not a deadline.

Can I build several habits at once?

You can, but each new habit takes attention. One goal and one key action a day is more reliable; add the rest once the first has become automatic.

Sources

Lally P., van Jaarsveld C. H. M., Potts H. W. W., Wardle J. How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 2010, 40(6), 998–1009.
Maltz M. Psycho-Cybernetics. Prentice-Hall, 1960.
Wood W., Neal D. T. A new look at habits and the habit–goal interface. Psychological Review, 2007, 114(4), 843–863.