Day 2 is not Day 7 when the clock is sunset

Advertising reports love Day 1, Day 7, and Day 30 because campaigns are bought in those buckets. A fasting-window product in Malaysia does not live in those buckets. The meaningful clock is often sunset to sunset, or the first opening of a fast, and the calendar is interrupted by Ramadan, by travel home, and by weeks when a person is sick and must not fast.

If you still cut “Day 2” at 48 hours after install, you mix people who installed on a Thursday afternoon with people who installed twenty minutes before Maghrib. One of those people may have already closed a window. The other has not had a sunset yet. Calling both of them Day 2 is not precision. It is a habit borrowed from a different industry.

When we reconstruct these products, we ask for a local-time stamp and, where the product already stores it, a sunset or window-close stamp. Week zero becomes the first logged close. Return marks sit on subsequent closes, not on midnights. A person who logs five closes in a week and then travels is not “Day 7 churn”; they are a travel-week cohort, which you can only see if you named travel or a missing-sunset gap in the first place.

This is slower than copying a template from a North American consumer product. It is also the only way we have found to stop a team from “fixing” Ramadan every year as if it were an outage.

If your fasting product is used outside Malaysia as well, do not average the clocks. Cut a Malaysia-time cohort and a separate cohort for the other region. Sunset in Ipoh and sunset in another longitude are not a rounding error.

If this note names the argument in your team, send the ritual and the week people vanish.