Does Habit Tracking Work? What 138 Experiments Found

Yes, habit tracking works, but less than apps promise and in a more specific way than most articles say. The largest review of the evidence pooled 138 experiments with nearly 20,000 people and found that tracking progress helped people reach their goals, by a small-to-medium margin. It helped most when people wrote the record down and when someone else saw it.

That review, and a handful of others, also say what tracking does not do. Both halves are below, with a link to each study so you can check the numbers yourself.

Does habit tracking work? The short answer

Tracking works on average, and the effect is real but modest. In a 2016 meta-analysis of 138 randomised experiments, people prompted to monitor their progress reached their goals more often than people who were not, with an average effect the authors call small-to-medium (d = 0.40). Tracking made people check in far more often, and the extra checking explained the gain.

The review is Harkin and colleagues, Psychological Bulletin, 2016. The authors pulled together every experiment they could find that randomly split people into a group asked to track their progress and a group that was not, then checked who got closer to their goal. The goals were ordinary ones: exercise, diet, weight, blood pressure, study, saving energy.

Two numbers carry the result. Tracking prompts raised how often people checked their progress by a very large margin (d = 1.98). And they raised goal attainment by a smaller one (d = 0.40). The authors then showed that the first explained the second: the more a prompt got people checking, the more it moved the goal.

What does d = 0.40 mean in plain terms? A standard way to read it: the average person in a tracking group finished ahead of about two in three people in the matching group that did not track. That is a real edge. It is not a transformation, and it varied a lot from study to study.

For scale, the authors set it beside the effect of raising how strongly people intend to act, which an earlier review put at d = 0.36. Tracking helps about as much as more motivation does, and it works on a different step: not wanting the goal, but checking how you are doing against it.

Tracking a habit helps the habit, not always the result

What you track is what moves. In the same review, tracking a behaviour, such as steps or workouts, had a large effect on that behaviour (d = 0.79) but no reliable effect on outcomes like weight (d = 0.14). Tracking the outcome did the reverse: it moved the outcome (d = 0.62) and not any one behaviour. Pick the record that matches the thing you want to change.

This is the most useful finding in the paper for anyone with a habit tracker, and almost nobody quotes it. The authors split the studies by what people were asked to watch:

Monitoring a behaviour (did I walk, did I floss, how long was my shower) changed that behaviour strongly, and did not reliably change the outcome it was meant to serve.
Monitoring an outcome (my weight, my blood pressure, my bill) changed the outcome, and did not reliably change any one behaviour, because people found other routes to the result.

Their explanation is that most outcomes have many paths. Watching your energy bill might lead you to shorter showers, or new bulbs, or a warmer jumper. Watching your shower length will shorten your showers and may or may not shift the bill.

So a row of ticks for "walked 20 minutes" is good evidence that you will keep walking. It is weaker evidence that you will lose weight, sleep better or feel calmer, even if that is why you started. If the outcome is the point, track it too, less often: the habit daily, the outcome weekly.

Written down and shared: what made tracking work better

Two conditions made tracking noticeably stronger. When people physically recorded their progress rather than just noting it in their heads, the effect rose from d = 0.29 to d = 0.43, and from 0.23 to 0.57 when the goal was measured objectively. When progress was reported to someone or made public, the effect was 0.47 to 0.55, against 0.19 for tracking kept private.

Both results come from the Harkin review, and the authors offer reasons rather than proof. A written record is easier to remember and harder to ignore when it says something unwelcome. Private checking lets you look away; a mark on a page, or a number a friend can see, does not.

Two other things the authors tested made no difference:

How long the tracking lasted. Longer programmes did not produce bigger effects.
What you compare against. Comparing today with a target in the future worked as well as comparing it with where you were in the past.

This fits an older finding. A 2009 review of 122 diet and exercise programmes, covering 44,747 people, found that self-monitoring explained more of the difference between programmes than any other technique, and that programmes pairing it with other self-regulation steps, such as setting a goal and reviewing it, did better (effect 0.42 against 0.26).

If you want someone to report to, accountability partners for habits covers how to set one up without turning it into a chore.

Why tracking helps

The best-supported reason is simple: a record shows the gap between what you meant to do and what you did, and seeing a gap is what prompts people to close it. Psychologists call this control theory. Tracking also turns a vague sense of "I have been doing all right" into a count you cannot argue with, in either direction.

The Harkin authors root their review in this idea: goals only guide behaviour if people compare where they are with where they want to be, and most of us skip that step unless something makes us take it.

A second reason is less studied but worth knowing. People keep going with long-term goals mainly because of rewards they feel now, not later. In studies of New Year's resolutions, studying and exercise, immediate enjoyment predicted who kept going; delayed benefits did not (Woolley and Fishbach, 2017). A tick at the end of a walk is a small, immediate result where the real benefit of the walk is months away. That link between the finding and tracking is our reading, not something the study tested.

When habit tracking backfires

Tracking has two documented costs. Measuring an activity can make it less enjoyable: in six experiments, counting steps or pages made people do more but enjoy it less. And a broken streak, once it shows in a log, lowers the chance you do the habit again, even when your actual record is just as good. Both costs can be managed.

Measurement can drain the fun. In six experiments by Jordan Etkin (2016), people who tracked walking or reading did more of it and enjoyed it less, because counting made the activity feel like work. This matters most for things you already like. If you read for pleasure, a page count may be the wrong thing to track; a tick for "read tonight" asks less of the book.

A broken streak changes behaviour on its own. In seven studies, Silverman and Barasch (2023) found that people shown an intact streak in their log did the behaviour again more often than people shown a broken one, even when the real behaviour behind the two logs was the same. The effect was stronger when people blamed themselves for the break, and weaker when they could repair the streak.

The practical point is that the way a record is drawn matters as much as the record. A missed day that looks like a wiped slate hurts more than a missed day that looks like one gap in a long run. What to do when you break a streak goes through the day after.

How long to track before a habit runs on its own

Plan on about two months and expect wide variation. In the best-known study, Lally and colleagues (2010) followed 96 volunteers doing one daily behaviour for 12 weeks; the median time for it to feel automatic was 66 days, with individual estimates from 18 to 254. Missing one day did not undo the progress.

Later work lands in the same place. A randomised trial of a daily eating habit found a median of 59 days to peak automaticity among people who formed the habit (Keller and colleagues, 2021). A 2024 review of 20 studies put the medians at 59 to 66 days, the means at 106 to 154, and the individual range at 4 to 335 days.

Two cautions. The 254-day top of Lally's range is a projection beyond the 84 days the study ran. And about half of Lally's volunteers had not reached a habit-level score by the end, a result the British Psychological Society's summary of the study also reports. The full story, including where the 21-day figure came from, is in how long it really takes to form a habit.

How to track habits so the tracking helps

Keep it small and physical, and make it visible. From the studies above: record each day rather than just noticing it, track the behaviour you control and check the outcome less often, show the record to someone, and use a format where one missed day reads as a gap, not a reset. None of this needs a particular app.

A short version, each point tied to the finding it comes from:

Write it down. A tick on paper or a tap in an app, not a mental note. Recording beat not recording (Harkin).
Track the habit daily, the result weekly. Behaviour tracking moves behaviour; outcome tracking moves outcomes (Harkin).
Let one person see it. Reported or public progress beat private progress (Harkin).
Draw misses as gaps. A calendar or grid that shows every day keeps a miss in proportion; a single streak number does not (Silverman and Barasch).
Do not count what you already enjoy. Tick it done, skip the page or mile count (Etkin).
Start with one to three habits. No study sets this number. It is our advice, because every extra row is another thing to record.

Paper works for all six, and habit tracking methods compares the formats side by side. If you want an app, Aura was built around the same points: a one-tap check-off, a completion grid that shows every day rather than only the current streak, friends who can see and cheer your progress, and a streak freeze in Aura Pro for the day that goes wrong. The best apps for habit tracking compares it with the others honestly.

What the research does not say

It does not say tracking will change your life, or that any one app works better than paper. The effect is an average across very different studies, most of them short and many about health goals set by researchers. No trial here tested a consumer habit app against no app. And the tracking prompts in these studies often came with other help.

A few limits worth stating plainly:

It is an average with wide spread. The Harkin authors found large differences between studies. Some people and some goals gained a lot; others gained nothing.
Many goals were not habits. The review covers weight, blood pressure, exercise, diet and study targets. Habits of the kind people track at home, such as reading, journaling or flossing, are only a small part of it.
Weight-loss evidence is weaker than it looks. A systematic review of 22 weight-loss studies found a consistent link between self-monitoring and weight loss, but rated the evidence weak: small, mostly white, mostly female samples, and a reliance on people's own reports.
A tick is not a habit. A checkmark records that you did something, not that it has become automatic. Researchers measure that separately, with short questionnaires such as the four-item Self-Report Behavioural Automaticity Index.
Nobody has tested the numbers apps love. There is no research behind "track exactly three habits" or "never miss twice". Treat those as advice, ours included.

Frequently Asked Questions

Does habit tracking actually work?

Yes, on average. A 2016 review of 138 randomised experiments with nearly 20,000 people found that prompting people to track their progress helped them reach their goals, with a small-to-medium effect. Tracking worked best when people wrote the record down and when they reported it to someone else.

Do habit tracking apps work?

No trial has tested a consumer habit app against no app, so nobody can say that a given app works. What the research supports is the act of recording progress, whatever you record it on. An app helps if it makes recording quick and the record visible; paper helps for the same reasons.

Is it better to track habits on paper or in an app?

The research does not separate the two. It found that physically recording progress beat keeping it in your head, and a tap in an app counts as recording. Choose the one you will use every day, and keep it where you will see it.

Can tracking habits make things worse?

Sometimes. Counting an activity you already enjoy, such as reading or walking, can make you do more of it but enjoy it less. And seeing a broken streak in a log can make people less likely to carry on. Tracking the done-or-not-done fact, and viewing misses as single gaps, reduces both risks.

Should I track the habit or the result?

Track the habit daily and the result less often. In the largest review, tracking a behaviour changed that behaviour strongly but did not reliably change outcomes such as weight, while tracking an outcome changed the outcome. If you care about both, record both.

How long should I track a habit?

Plan on about two months, and do not stop the day a habit starts to feel easy. Studies put the median time for a daily habit to feel automatic at 59 to 66 days, with some people far faster and others far slower. Missing a single day does not undo the progress.