How Long Does It Take to Build a New Phone Habit? What the 66-Day Study Found

The famous number is 66 days, but it is an average of a wide spread, it comes from a study of eating, drinking and exercise, and no one tested phones. Here is what it does tell you.

Matthew KFounder, FellaBlog

Short answer

Expect weeks to months, not 21 days. The study's average was 66 days, and individuals ranged from 18 days to a predicted 254.

The study is Lally, van Jaarsveld, Potts and Wardle (2010), published in the European Journal of Social Psychology. It measured how long a new daily behaviour takes to feel automatic. It did not look at phone use, so applying it to habits like leaving your phone alone is a reasonable guess, not a finding.

What the researchers measured

Per the British Psychological Society's summary, 96 volunteers (mean age 27) each chose a health-related behaviour to repeat once a day in a consistent context, such as eating a piece of fruit with lunch, a 15-minute run before dinner or 50 sit-ups after morning coffee. The study ran for 84 days, and 82 people saw it through to the end.

Every day, participants went online, reported whether they had done the behaviour and rated how automatic it felt, with statements like "I do it automatically," "I do it without thinking" and "I'd find it hard not to do." That automaticity rating is the measure of a habit. It is entirely self-reported, which the researchers acknowledge as a shortcoming.

What they found

  • The curve flattens. Early repetitions produced the biggest gains in automaticity, then gains shrank until they levelled off at a plateau.
  • 66 days on average. The average time to reach the plateau was 66 days. The summaries we read give an average, not a median.
  • An enormous range. The quickest person took 18 days. At the other end, one participant had not plateaued by day 84, and the 254 day figure is a prediction from their still-rising curve. The BPS summary adds that about half of participants had still not reached a high enough automaticity score after 84 days.
  • Simpler behaviours were faster. BPS reports that exercise took about 1.5 times as long as eating or drinking behaviours.
  • One missed day barely mattered. A single lapse had little effect on later gains. Many missed repetitions, even spread out, lowered the automaticity people reached.

A UCL behavioural science blog by Ben Gardner also addresses the "21 days" figure. It traces it to a 1960 book by Maxwell Maltz and calls it unfounded as a basis for habit formation, concluding that habit formation typically takes longer than that.

Limits to keep in mind

  • Not about phones. The behaviours were eating, drinking and activity habits.
  • Building a habit is not breaking one. Participants added a new behaviour. Stopping an automatic action, like opening an app, is a different task.
  • Self-reported. The only measure was how automatic people said it felt.
  • The average hides the spread. A group average of 66 days says little about how long you will take.
  • We read summaries. We did not read the paper itself. The details above come from the BPS Research Digest and the UCL blog, which agree on the shared figures, and from the paper's published citation.

What it means for your phone

These are our suggestions, not findings from the study.

  • Plan for months, and judge progress by how automatic it feels, not by a day count. Expect the early weeks to show the largest change.
  • Pair the new behaviour with a fixed cue. The study's habits were tied to a daily cue, like lunch or a morning coffee. Putting the phone in the same spot after you get home is the same idea.
  • A slip is not a reset. One missed day barely mattered in the study, so do not abandon a plan after one lapse.
  • Reduce what the habit has to fight. Habits form by repetition in a context. Our habit loop page explains the cue, routine and reward pattern, and a scheduled app block can remove the cue while the new habit builds.

Habit formation FAQ

In Lally and colleagues' study, the average was 66 days to reach maximum automaticity, with individuals ranging from 18 days to a predicted 254 days.

No. A UCL blog traces the figure to a 1960 book and calls it unfounded as a basis for habit formation.

A single missed day had little effect on later gains, while many missed repetitions lowered the automaticity people reached.

It did not test phones. Participants chose eating, drinking or activity behaviours, so applying it to phone use is a reasonable guess, not a finding.

References

  • Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998-1009. DOI. We could not read the full paper or its abstract.
  • British Psychological Society Research Digest, How long does it take to form a habit? (October 2010): sample sizes, measures, behaviours, the range and the missed-day finding.
  • UCL Behavioural Science and Health blog, Busting the 21 days habit formation myth (Ben Gardner, 2012): the 66 and 18 to 254 day figures and the origin of the 21 day claim.