Personal Stats
Always-available numbers about the user's own behaviour, on their terms.
The problem it solves
Users want to know how they are doing and often have no way to find out. Withholding their own data forces them to guess, and a guess is easier to walk away from than a record.
Personal stats is the standing version of the annual recap. A page the user can open whenever they want, showing what they have done over time. No event, no campaign, no prompt. It sits there and it is honest.
The value comes from being available rather than pushed. Nobody resents a stats page. They resent being told their numbers are down. Same data, different consent, and the difference decides whether the feature builds trust or spends it.
Rules for the page
- Show outcomes the user cares about, not sessions and time in app.
- Default to a long window. Short windows exaggerate normal variation into a trend.
- Present low periods neutrally, without colour coding them as failure.
- Let the user pick which stats appear, and let them hide ones that do not help.
- Offer export in a format that opens elsewhere.
That fourth rule matters more than it looks. Someone tracking recovery from injury may want distance hidden. Someone managing an eating disorder may need calorie totals gone. Giving people control over which of their own numbers they see is cheap to build and occasionally protects someone from real harm.
Worked example
Pull, not push
A finance app could send a monthly push saying spending is up 30 percent. Or it could keep a trends page the user opens when they want to know. The push version wins the session and costs some goodwill on a month when the user already knew and did not need telling. The pull version is there on the day they are ready to look.
When it fits
- The user is tracking something they chose to track, like training load, spending, or writing output.
- There is enough history for patterns to be real rather than noise.
- You can report outcomes rather than engagement proxies.
- The user can configure or hide individual metrics.
When it backfires
- Stats are pushed rather than made available. An unrequested notification about a decline is a judgement the user did not ask for.
- The metrics measure your engagement. A page telling someone how many hours they spent in your app is a strange thing to present as a service.
- Sensitive metrics cannot be turned off, which puts weight, calories, or spending in front of someone every time they open the page.
- Short windows create false trends, so a normal quiet fortnight is presented as a decline and prompts behaviour the user did not need to change.
Products using it
Strava
Keeps per-sport training and fitness trend views the athlete opens on demand, with configurable date ranges.
Apple Health
Aggregates metrics into a single view where the user chooses which categories are pinned and which are hidden.
YNAB
Provides spending and net worth reports over user-selected ranges rather than pushing periodic verdicts.
The psychology under it
The takeaway
Make the user's numbers available rather than announced, and let them hide the ones that do not help them.
Finished the teardown? Bank it and the day counts toward your run.