Estimate how many days it may take to break or replace a habit, based on UCL research (median ~66 days). Free, no signup, with a week-by-week plan.
Notice what reliably triggers the habit — a time, place, feeling, or preceding action. Just observing (without forcing change yet) makes the loop visible.
When the cue fires, run your replacement response instead. Expect to slip — the goal is reps, not a perfect record. Each interruption weakens the old pathway.
Make the old habit harder (remove the trigger, add a step) and the new one easier. Track a daily check-in so a visible streak starts pulling you forward.
Around the ~66-day median the new response starts to feel default rather than effortful. Lapses become rarer and easier to recover from.
Protect the streak through travel, stress, and disrupted routines — those are the classic relapse windows. Maintenance, not willpower, is what makes it stick.
The headline you've probably heard — "it takes 21 days to break a habit" — isn't supported by research. The best-known study on how long behaviour change actually takes is Lally et al. (2010), run at University College London. Following 96 people forming a new daily behaviour, the researchers found it took a median of about 66 days for the action to feel automatic — with individuals ranging anywhere from 18 to 254 days. That enormous spread is exactly why this tool gives you a band rather than a single false-precision number.
This calculator anchors near that 66-day median and then shifts the estimate based on the four factors that most reliably move it: how long you've lived with the habit (a deeper groove takes longer to re-route), how often you repeat it (more reps = a stronger loop), how automatic it already feels, and whether you've lined up a replacement behaviour. The output is an honest planning range, not a promise — your own number could land anywhere inside the research's wide window.
Habits run on a loop: a cue triggers a routine that delivers a reward. The cue doesn't vanish just because you've decided to quit — so if you don't give it somewhere to go, the old routine keeps winning. Planning a substitute response to the same cue (stand up and stretch when the 3 p.m. slump hits, instead of reaching for a snack) is what James Clear calls inverting the laws of habit-building. It's one of the most consistent accelerators in the literature, which is why this tool trims the estimate when you have one.
Knowing your timeline is the easy part — showing up every day is the hard part. The week-by-week milestones above are most useful when each one becomes a tiny, trackable check-in. You can log this as a daily habit in HabitBox — it's free, needs no account, keeps your data on-device, runs on iOS and Android, and logs a check-in in one tap so a visible streak does the nudging for you. Pair it with the 2-Minute Rule Timer to make your replacement habit small enough that skipping it feels harder than doing it.
Note: this is a general estimate for everyday habits in healthy adults. Compulsive behaviours and substance dependence are different — if a habit is harming your health or you feel unable to stop, please talk to a qualified professional.
There's no fixed number. The most-cited study — Lally et al. (2010) at University College London — found new behaviours became automatic over a median of about 66 days, but individuals ranged from 18 to 254 days. This calculator maps your situation onto a realistic band within that research, but treat the result as a planning estimate, not a guarantee.
The '21 days' figure is a myth. It traces back to a 1960s observation by plastic surgeon Maxwell Maltz about patients adjusting to changes, not rigorous behaviour-change research. The peer-reviewed Lally study found a median closer to 66 days, with wide variation — so 21 days is realistic only for the easiest, newest habits.
Because the underlying research shows enormous individual variation — the same study spanned 18 to 254 days. How ingrained the habit is, how often you do it, how automatic it feels, and whether you've planned a replacement all shift the timeline. A range is more honest than a single false-precision number.
Generally, yes. Habits run on a cue → routine → reward loop, and the cue doesn't disappear just because you want to stop. Giving that cue a planned substitute response — what James Clear calls an 'inversion' of habit-building — is one of the most consistent accelerators in the behaviour-change literature, which is why this tool shortens the estimate when you have one.
Lapses are normal and, in the Lally data, a single missed day did not meaningfully reset progress toward automaticity. What matters is getting back to your replacement response quickly. Tracking a streak helps — not to punish slips, but to make the next rep the obvious move.
An estimate only matters if you show up daily. HabitBox lets you log a one-tap check-in for your replacement habit, watch the streak grow through the tough early weeks, and keep going — free, no account, data stays on your device.
Free · Local-only data · No account required