The three retention types, side by side
One table with the rule, how the number behaves and what each one is for.
Classic counts users active on exactly Day N, rolling counts users active on Day N or later, and cumulative counts users active at least once from Day 1 to Day N. Classic is always the lowest; rolling and cumulative can be in either order. Classic and cumulative are fixed once Day N passes; rolling keeps rising.
How do classic, rolling and cumulative retention differ?
| Type | Counts a user if active... | Fixed once Day N passes? | Sample Day 1 / Day 7 | Best for |
|---|---|---|---|---|
| Classic | On exactly Day N. Activity on any other day does not count. | Yes | 35% / 18% | Benchmarks, day-specific changes, forecasts |
| Rolling | On Day N or any later day up to the last day in the data. | No, keeps rising | 67% / 38% | Is the user base still around? State the as-of date |
| Cumulative | At least once on any day from Day 1 to Day N. | Yes | 35% / 67% | Activation and onboarding |
Source: Sample data: a made-up app to show the method, not a benchmark. Day 1 pools all ten cohorts; Day 7 pools the eight that reached it.
Rules that always hold
- Classic is the floor. Every user counted by classic is counted by the other two.
- Cumulative only goes up as N grows: each extra day widens the window.
- Rolling only goes down as N grows, on a fixed as-of date, but rises for the same cell as time passes.
- Two identities to check your tool: rolling on a cohort's last observed day equals classic, and rolling Day 1 equals cumulative on the last observed day. In the sample, the 1 Sep cohort's rolling Day 1 and cumulative Day 14 are both 73%.
- Rolling and cumulative can cross. Rolling usually leads early, cumulative overtakes later.
One page per type
- 01Classic retentionClassic retention is the share of an install cohort that opened the app on exactly Day N after installing.
- 02Rolling retentionRolling retention is the share of an install cohort that opened the app on Day N or on any later day observed so far.
- 03Cumulative retentionCumulative retention is the share of an install cohort that opened the app at least once between Day 1 and Day N.
Questions
Which is always the lowest: classic, rolling or cumulative?
Classic. Rolling and cumulative both include Day N itself plus other days, so neither can be lower than classic for the same cohort and day.
Which retention types are bounded?
Classic and cumulative are bounded: once Day N has passed, their value never changes. Rolling is unbounded and keeps rising as late returns arrive.
What is the difference between classic, rolling and cumulative retention?
Classic counts users active on exactly Day N. Rolling counts users active on Day N or any later day, and cumulative counts users active at least once from Day 1 to Day N. They use the same raw activity but answer different questions.
Why is my rolling retention higher than classic?
Rolling also counts users who skipped Day N but came back later, so it can never be lower than classic. The gap grows the longer a cohort has been observed, because late returns keep adding to it.
What is cumulative retention?
Cumulative retention is the share of an install cohort that opened the app at least once between Day 1 and Day N. It never falls as N grows and it is fixed once Day N has passed, which makes it a good activation measure.
Which retention number should I compare with published benchmarks?
Classic retention, unless the benchmark says otherwise, because most published retention benchmarks count activity on an exact day. Comparing a rolling or cumulative number with a classic benchmark will flatter you.