Stickiness (DAU/MAU Ratio) DAU/MAU
Of the people who used the product this month, what fraction use it on a typical day -- a rough read on habit.
The fork, why two teams get different numbers
The numerator is the argument. Some teams put a SINGLE day's DAU over the
month's MAU (spiky, and literally "today's actives as a share of monthly
actives"). Others put the PERIOD-AVERAGE DAU (mean daily actives across
the month) over MAU, which reads as "the average user was active this
fraction of days" and is the interpretation most people think they are
getting. The denominator inherits the MAU fork (calendar vs rolling-28).
A growth team quoting 0.20 average-DAU/MAU and an exec quoting a good-day
0.31 single-DAU/MAU are both "stickiness" and disagree by half.
The trap
Only the period-average form supports the "active X% of days" reading: average-DAU/MAU of 0.20 against a 28-day MAU means the typical monthly user showed up on roughly 0.20 * 28 = ~5.6 days.
- The full trap, worked on a real export
- Every formula variant, spelled out
- The reconciliation anchor, what to tie it to and when to refuse
Reference: Product-analytics convention (no governing body)