So 21 days really is a myth?

Yes. Maltz wrote about adapting to a new face after surgery, not about habits. The 2010 Lally study reported a 66-day average with a wide range depending on habit difficulty.

How many habits at once?

Lally suggests one. Fogg allows several if they are tiny (2 push-ups, not 20). After one is automatic, add a second — in sequence, not in parallel.

What counts as a done day?

The minimal version of the habit. If you planned 20 push-ups and did 5 — it counts. Goal: do not break the streak, even on a weak day.

Does streak length matter?

Symbolically yes, behaviourally less than it seems. Long-term frequency matters more (60% of days for a year beats 100% for a month then giving up).

Habit tracker — without the 21-day myth

The 21-day figure comes from plastic surgeon Maxwell Maltz in 1960 and was never a habit study — Maltz observed that patients needed at least 21 days to get used to a new face. It got repeated so many times it passes for fact. Actual research shows a range from 18 to 254 days, with a mean of 66.

How it works

1
Pick one habit
Not five. Specific, measurable, daily or a few times a week. “Read more” is not a habit, “10 pages before sleep” is.
2
Anchor it to an existing routine
BJ Fogg calls it anchoring. After morning coffee, 5 push-ups. The existing habit is a ready-made trigger — no remembering needed.
3
Log every day
The act of marking it itself works — it is evidence for the brain that this matters. A streak in the app is not gamification, it is feedback.
4
Allow one missed day, not two
Lally observed that a single miss does not break formation, two in a row begins to.

Why it works

Lally and colleagues (2010) followed 96 people for 84 days — the average time for a behaviour to feel automatic was 66 days, range 18–254. Smaller habits (drinking water) automated faster than complex ones (exercise).

Wood and Neal (2007) describe a habit as a triangle: context → response → reward. A habit tracker builds an artificial context (a daily “did you?”) and an artificial reward (the streak, the visualisation). Medicine for the first months, before the brain builds its own.

Sources

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).
Wood, W., Neal, D. T. (2007). A new look at habits and the habit-goal interface, Psychological Review, 114(4).
Fogg, B. J. (2019). Tiny Habits: The Small Changes That Change Everything, Houghton Mifflin Harcourt.

When this method does NOT work

Trackers work for additive habits (start doing X) and weakly for subtractive ones (stop doing Y) — removing a behaviour is a different mechanism, closer to therapy. Trackers can also turn into a trap: people start hunting the streak instead of the outcome (you read one page just to avoid breaking it). If you catch yourself doing it “for the box”, take a week off.

In TaskAura

The Habits view shows a list with the daily streak and a backward heatmap. You can pick weekdays (e.g. Mon–Fri only) and set reminders. No rewards, badges or paid “streak freezes” — on purpose. Only the data counts.

Try it free

Frequently asked questions

So 21 days really is a myth?: Yes. Maltz wrote about adapting to a new face after surgery, not about habits. The 2010 Lally study reported a 66-day average with a wide range depending on habit difficulty.
How many habits at once?: Lally suggests one. Fogg allows several if they are tiny (2 push-ups, not 20). After one is automatic, add a second — in sequence, not in parallel.
What counts as a done day?: The minimal version of the habit. If you planned 20 push-ups and did 5 — it counts. Goal: do not break the streak, even on a weak day.
Does streak length matter?: Symbolically yes, behaviourally less than it seems. Long-term frequency matters more (60% of days for a year beats 100% for a month then giving up).