UCL study: habit formation takes 18 to 254

A University College London study found that forming a new daily habit to near-automaticity can take anywhere from 18 to 254 days. The median time was 66 days, according to research published in the European Journal of Social Psychology.

This finding directly challenges the widespread belief that any habit can be solidified in just 21 days. The study, led by Phillippa Lally, followed 96 adults who each chose a new eating, drinking, or activity behaviour to perform once daily in a consistent context.

The study's design and definition

Participants selected their own behaviour, such as drinking water with lunch or running before dinner. The crucial requirement was a stable cue. For 12 weeks, or 84 days, they reported daily whether they performed the action and how automatic it felt.

The research defined a habit as an association built through repetition in a consistent setting. This means the cue itself begins to prompt the action with less conscious thought over time.

A wide range from a specific group

The headline figures of 18 to 254 days did not apply to all 96 participants. They came from a subset of 39 people whose self-reported progress fitted the researchers' mathematical model well. For this group, the estimated time to reach 95% of their projected automaticity plateau showed this wide variation.

The median of 66 days means half of these estimates were below that point and half were above. It is not a universal average for every person or behaviour.

Why estimates exceeded the study period

Automaticity increased quickly at first, then in smaller increments, following an asymptotic curve that approaches a ceiling. The researchers defined "near-automaticity" as reaching 95% of this projected ceiling for each individual.

Since the observation period was only 84 days, longer formation times like 254 days were extrapolations based on the fitted curves. The model estimated when a participant would hit the 95% threshold if their observed pattern continued.

The limits of the findings

The measure of automaticity was based on self-reported feelings, not neural readings. Participants rated statements about performing the behaviour automatically or without thinking.

This captures the subjective experience but is one dimension of habit. Furthermore, achieving automaticity does not guarantee the behaviour will last permanently, as motivation, schedule, or environment can change later.

The study also had a limited scope. It examined only eating, drinking, and physical-activity behaviours chosen by 96 people. The results suggested more complex routines might take longer, but the sample is too small to assign reliable times to every possible habit.

Practical insights on consistency

A key practical finding was that missing a single day did not reset the habit-formation curve. One lapse did not erase the automaticity already built.

However, the study's mechanism relied on repetition in a stable context. Highly inconsistent participants did not show the same successful formation pattern. Consistency still matters, but perfection is not required.

**The origin of the 21-day myth**

The popular 21-day rule did not come from controlled habit research. It is commonly traced to plastic surgeon Maxwell Maltz's 1960 book Psycho-Cybernetics.

Maltz observed patients needed at least about 21 days to adjust to a changed appearance. This was an observation about adjustment, not an experiment measuring daily automaticity. Over time, popular retellings removed the qualifiers, turning a modest observation into a rigid formula.

**Why 66 days is not the new rule**

Phillippa Lally has repeatedly stated that 66 days is not a replacement deadline. In a later interview, she clarified it was the average in her specific study, not a universal answer.

Subsequent research, including a 2024 systematic review cited by Silicon Canals, supports this variability. That review, covering 2,601 adults, found formation times ranging from 4 to 335 days across studies, with medians between 59 and 66 days and means much higher. The evidence points to person, behaviour, and context all influencing the timeline.

The most instructive part of the UCL study may be the shape of the curve itself. Early repetitions yield large gains in automaticity, with each subsequent repetition adding a smaller increment as the curve approaches its plateau.