October 26, 2026
Does tracking your progress help? What 138 experiments found
In 138 randomised studies, prompting people to track a goal raised how often they checked by a lot and goal attainment by a small to medium amount. What to track, and what the paper does not show.

Na tej stronie
In 138 experiments, adults were randomly given something that prompted them to track a goal, or put in a comparison group that was prompted less or not at all. Being prompted helped: across the range of estimates, the average prompted person did better on the goal than about 6 in 10 people in the comparison group. The gain is small to medium, and its size depends on how you correct for small studies that may never have been published (pp. 198, 213). It is much easier to get people to track than to get tracking to pay off, and almost all of these studies were about health.
Tracking is easy to get people to do
Have you ever worked on something for weeks and still had no idea whether you were getting anywhere? A team led from the University of Sheffield collected every experiment they could find on whether keeping track helps. Each randomly allocated adults to a condition designed to get them monitoring their progress, or to a control, and measured how often people then actually monitored1 A meta-analysis pools the results of many studies into one estimate, so it can say more than any single study can. ↩ (pp. 198, 203). Of 22,054 records, 138 studies qualified (Figure 1, p. 204).
People were asked to keep a food diary, wear a pedometer, weigh themselves daily, check blood sugar at home, or fill in a website (p. 202; Table 2, pp. 205 to 207). In 67 of the 138 studies the comparison group was not prompted to monitor at all (Table 3, p. 211).
Prompted people tracked far more, by more than twice the size of a conventional large effect, the authors say (pp. 208, 216). Goal attainment rose by much less, by about 0.2 to 0.4 standard deviations2 A standard deviation is a typical spread in people's scores. 0.4 means the average prompted person scored 0.4 of that spread above the average person in the comparison group. ↩ across 18,398 people, counting both groups (pp. 208, 213).
Looking is not the same as moving
In five pedometer studies, how often people checked rose by more than with almost any other method, and goal attainment rose by 0.02, with an interval that includes zero (Table 3, p. 210; p. 214). Five studies is a warning, not a verdict, and it compares studies, not people: in those studies, much more checking came with no better result.
The better lead is in what was tracked, again comparing groups of studies. Prompts to monitor behaviour moved behaviour (0.79, 17 studies) but had no reliable effect on outcomes (0.14 across 8, interval includes zero); prompts to monitor outcomes moved outcomes (0.62, 30 studies) but not behaviour reliably (0.17 across 4, interval includes zero) (p. 214). Track the thing you actually want to change.
Writing it down, telling one person
Progress written down: 0.43 across 106 studies, against 0.29 across 32 where it was not. Reported to at least one person: 0.47 across 95, against 0.19 across 14 where it stayed private (Table 3, p. 210; p. 214).
Nobody was randomly assigned to write or to tell. Those are comparisons between groups of studies, run without correcting for multiple tests, and the authors warn that study features like these travel together: studies that ask people to track outcomes may also be more likely to ask them to write things down (p. 218). For writing, they offer possible reasons, one being that "it is not enough merely to monitor progress" and "the person must also face up to what the information shows", which a written record may make harder to avoid (p. 217). For telling, they name commitment and accountability, and also that participants may have done what they thought the researchers wanted, and they call for an experiment that compares telling with not telling directly (p. 217). For company during the work itself, see our post on body doubling.
What it does not say about streaks
This paper never tested streaks, unbroken chains of days, or rewards for consistency. How long people were asked to monitor did not predict how well they did (Table 4, p. 215), and the pedometer studies were among those that raised checking most while gaining least. What it supports is that being prompted to look at your progress helps; that writing it down and telling someone add to that is suggested, not shown.
What it suggests trying
Our reading for work and study, which the paper did not test:
- Pick the thing you want to change, not the thing that is easy to count. If the goal is finishing a chapter, log what got written, and treat hours as the means.
- Write the number down where you will see it again, on paper or in a notes file, rather than keeping it in your head.
- Tell one person what it says, and ask for nothing back. And plan when and where you will work: programmes that included action planning came out at 0.50 across 43 studies, against 0.34 across 95 without, again a comparison between groups of studies (Table 5, p. 215).
Our example, not the paper's: each study day, write one line with the pages you finished, and send it to a friend on Fridays.
Where Yappie fits
We make Yappie, a focus app with a pet beside the timer, so read this as an interested party. It records your focus time in charts, and its study groups share goals and a leaderboard. That is time focused, and the matching result is a reason to also write down what the time produced. Yappie also has streaks, and this paper gives them no support.
References
- Harkin, B., Webb, T. L., Chang, B. P. I., Prestwich, A., Conner, M., Kellar, I., Benn, Y., & Sheeran, P. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142(2), 198 to 229. https://doi.org/10.1037/bul0000025↩