There is no reliable number for how long it takes to break a habit. The often quoted 66 days comes from a study of 96 people forming new ones, where times ranged from 18 days to a projected 254 and half had not got there by the end. The 21 day figure has no research behind it.
Two figures dominate any discussion of how long it takes to break a habit. One is made up. The other is real and far less solid than the people quoting it suggest.
The 21 day figure is not research
It traces to a plastic surgeon's observations in the 1960s about how long patients took to adjust to their new appearance, which then became a self help staple. It was never a study of habit formation and there is no reason to apply it to your phone.
If someone tells you 21 days with confidence, they have not checked. It is one of the most widely repeated claims in self help and one of the easiest to trace back to nothing.
The 66 day figure is real, and comes with heavy caveats
This one comes from Lally, van Jaarsveld, Potts and Wardle, published in the European Journal of Social Psychology in 2010. Ninety six participants chose a new daily behaviour tied to a consistent cue and tracked it for 84 days.1
Sixty six days is the headline. Now the parts that get removed.
- The range was enormous. From 18 days to a projected 254.1 A figure with that spread is not a timeline you can plan around.
- The 254 was never observed. It is a model extrapolation past the end of the study.2 So is the plateau the whole analysis rests on.
- About half had not got there by day 84.2 The study ended before most participants finished.
- The 66 days comes from a subset. Of 62 individuals whose data was fitted, 39 showed a good fit, and the figure is drawn from those.2
- It was about forming habits, not breaking them. Participants added a behaviour. Stopping one is a different problem and has not been measured the same way.
None of this makes the study bad. It was a careful piece of work with modest claims. The problem is what happened to it afterwards, where a median from a fitted subset became a universal law.
The one genuinely useful finding
Missing a single day did not appear to reduce habit formation.1
That is worth more than the 66 days, because it contradicts the thing most people believe. The streak logic that governs habit apps, where breaking the chain undoes your progress, is not what the data showed. One missed day was not a reset.
Which means the most common reason people abandon an attempt, having missed a day and concluded they have ruined it, is based on a belief the research does not support.
The study also found more complex behaviours took longer.1 Stopping a reflex you perform two hundred times a day is complex.
Why break a habit is the wrong frame for phones
Breaking a habit implies an end state where you no longer have it. For phone use, the evidence does not really support that picture.
A 2025 randomised controlled trial in BMC Medicine had 58 students hold under two hours a day for three weeks, with measured improvements in stress, wellbeing, depressive symptoms and sleep. Screen time then climbed back quickly, and by the six week follow up it was approaching baseline.3
Three weeks of successful change produced no lasting change in behaviour once the constraint was removed. Not because those students were weak, but because the phone is still there, still designed the same way, and the environment did not change when their motivation did.
So the practical reframing: you are probably not going to break this habit. You are going to put something in place that makes it not happen, and keep that thing in place.
That sounds like a lesser achievement and it is the one that works.
It also explains an experience almost everyone has had and misreads. You manage a good month, feel genuinely different, and assume the thing is beaten. Then a bad week arrives, a holiday, an illness, a deadline, and within days you are back where you started and concluding you never really changed. Nothing about that sequence means the month was wasted or that you are incapable. It means the improvement was held in place by effort, and effort is the one resource that a bad week takes away first.
What to do with all this
- Stop counting days. There is no finish line, and waiting for one sets you up to quit when it does not arrive.
- Do not reset on a missed day. The research specifically found one lapse did not undo progress.
- Change the environment rather than the intention. Environmental changes keep working when motivation stops.
- Judge it at three months, not three weeks. Three weeks tells you almost nothing, as the trial above demonstrates.
- Prefer something small you will keep. One blocked hour a night, permanently, beats a heroic month.
Where SETL fits
SETL is built on the assumption that you will not break the habit. It is an iPhone app blocker that takes your chosen apps away at your bedtime automatically, indefinitely, with no streak to maintain and nothing to reset. If you miss a night because of a planned exception, SETL Breaks ends by itself. See also how to stop phone addiction and digital minimalism.
This article is general information, not medical advice. Speak to your doctor about ongoing sleep, attention or mental health concerns.
Sources
- British Psychological Society Research Digest on Lally, van Jaarsveld, Potts and Wardle, How are habits formed, European Journal of Social Psychology, 2010
- The Behavioral Scientist, How Long Does It Take to Form a Habit? What Lally et al. (2010) Actually Found
- Pieh C et al., Smartphone screen time reduction improves mental health, BMC Medicine, 2025