Habit formation calculator

How to use the calculator

Pick a Start date: the first day you did the new thing, or the day you plan to start. If it’s an eating, drinking or exercise habit, choose that under Kind of habit, because those were the three groups in the study; anything else uses the overall figures. The result gives three dates (earliest, average and slowest), a timeline from your start date and the study’s average curve. You can copy the dates or a shareable link.

If your start date is in the past, the result also says which day you’re on and where that falls in the study’s range.

How the dates are worked out

Each date is your start date plus a number of days from Lally and colleagues’ 2010 study: 18 days (the fastest anyone reached their plateau), 66 days (the median) and 254 days (the slowest). The middle half of people took 39 to 102 days, which the timeline shades darker. Choosing a habit type adds that group’s median: 65 days for eating, 59 for drinking and 91 for exercise. Dates are whole calendar days, so a change to or from daylight saving time can’t shift them.

Starting on Sunday, November 1, 2026, the calculator gives:

The curve under the timeline is the study’s own model, automaticity = a − b × e^(−c × day), drawn with the median value of each parameter (a = 35, b = 30, c = 0.042). On that curve, a habit is at 36% of its plateau on day 7, 76% on day 30 and 95% on day 68. Day 68 and the 66-day median differ slightly. A curve made from median parameters isn’t the same thing as the median of everyone’s own curve.

What the study measured

Phillippa Lally and colleagues at University College London recruited 96 volunteers, mostly postgraduate students with an average age of 27. Each chose one healthy eating, drinking or exercise behavior to do once a day after the same daily cue, such as eating a piece of fruit with lunch or running for 15 minutes before dinner. They tried to do it every day for 84 days. Each day they rated how automatic it felt, using seven questions from the Self-Report Habit Index (scored 0 to 42).

Of the 96, 82 gave enough data. The researchers fitted a curve to each person’s ratings and judged it a good fit for 39 of them. The 66 days is the median time those 39 took to reach 95% of their own plateau; a 2012 summary by Gardner, Lally and Wardle calls it “an average of 66 days”. The study lasted 84 days, so every time beyond that, up to 254 days, is where the person’s curve was heading, not something observed.

Three more limits apply. Automaticity was self-rated. The habits were simple and health-related. And more than half the people who gave enough data didn’t fit the curve well, so their times aren’t in the figures at all.

Why exercise took longer

Among the 39 good fits, the 13 people who chose exercise took a median of 91 days, against 65 days for the 10 who chose eating and 59 days for the 15 who chose drinking. The difference wasn’t statistically significant, and the authors point out that the study wasn’t designed to compare small groups. Still, exercise took about one and a half times as long. The authors suggest that more complex behaviors may take longer to become automatic.

Missing a day

Lally and colleagues looked at 140 occasions when someone skipped their habit after doing it three days running. On the next day their automaticity score was an average of 0.29 points lower, on a scale of 0 to 42. That’s a small dip. When they did it again the day after, their scores were higher than before the missed day. Over two days, scores rose by 0.79 points when the behavior was done both days and by 0.55 when a day was missed in between. The authors concluded that a missed opportunity “did not materially affect” habit formation, while noting that an earlier study found missing a whole week of exercise did hurt.

What a 2024 review added

In 2024, Singh and colleagues reviewed 20 studies of health habits with 2,601 participants. Only four reported how long habits took, and those gave medians of 59 to 66 days, means of 106 to 154 days and individual times from 4 to 335 days. Across the studies, habits done in the morning and habits people chose for themselves tended to be stronger. Eleven of the 20 studies were rated at high risk of bias, so treat these as rough guides.

For a one-minute daily habit like reading a lesson, we haven’t found a study that measured how long it takes; the nearest evidence is the simple eating and drinking habits above. Scroll counts a streak for each day you read a lesson, which makes the days visible. We haven’t tested whether that changes how quickly anything becomes automatic.

What the result doesn’t tell you

These times come from small studies of simple daily health habits, mostly in students. Yours could be shorter or much longer. The dates are the study’s figures laid on your calendar, not a forecast for you, and nothing on this page can tell you how automatic your own habit feels. The study measured forming new habits; it says nothing about breaking old ones. For the longer story of the 66 days and what helps, see the guide how long does it take to form a habit.