Home / Three types / Cumulative

Cumulative retention

Similar to bracket retention anchored at Day 1. How it is counted, a worked example, and what it can and cannot tell you.

Short answer

Cumulative retention is the share of an install cohort that opened the app at least once between Day 1 and Day N. Similar to bracket retention anchored at Day 1. In the sample app, the 1 Sep cohort's cumulative Day 7 is 64%.

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.

A user counts on Day N if they opened the app at least once on any day from Day 1 to Day N. Day 0 is the install day and never counts as a return; several opens on one day count once.

cumulative(N) = users active at least once from Day 1 to Day N / installs x 100

The same definition is used in Linkrunner's glossary entry on cumulative retention, so the numbers here line up with tools that follow it.

Figure 1

Worked example: ten users

Ten made-up users over 14 days, counted with the cumulative rule at Day 7. Shaded columns are the days that count.

	D0	1	2	3	4	5	6	7	8	9	10	11	12	13	14
User 1																counted
User 2																counted
User 3																counted
User 4																counted
User 5																not counted
User 6																counted
User 7																counted
User 8																counted
User 9																counted
User 10																not counted

Cumulative Day 7: 8 of 10 = 80%

The same ten users at Day 7 under the other rules: classic 30%; rolling 60%. Try other days in the ten-users view.

Source: Sample data: a made-up app to show the method, not a benchmark.

How does cumulative retention behave across a cohort?

The 1 Sep cohort of the sample app, the only one observed to Day 14.

Day N	Users counted	Cumulative retention
Day 1	51 of 148	34%
Day 3	79 of 148	53%
Day 7	94 of 148	64%
Day 14	108 of 148	73%

Figure 2

Every cohort, counted as cumulative

Rows are install days, columns are days since install.

Install cohort	D1	D2	D3	D4	D5	D6	D7	D8	D9	D10	D11	D12	D13	D14
1 Sep Tue148 installs	34	45	53	55	57	61	64	64	69	69	69	72	72	73
2 Sep Wed152 installs	34	52	59	66	68	70	71	78	79	80	81	82	82
3 Sep Thu154 installs	41	51	58	60	68	71	75	77	77	77	78	79
4 Sep Fri142 installs	38	51	57	61	64	71	72	73	76	77	77
5 Sep Sat148 installs	36	49	54	59	67	68	69	70	70	71
6 Sep Sun249 installs	29	39	46	55	58	59	62	63	64
7 Sep Mon262 installs	29	38	52	55	58	60	61	61
8 Sep Tue278 installs	32	53	57	61	63	65	67
9 Sep Wed258 installs	52	58	62	63	66	67
10 Sep Thu266 installs	31	39	45	49	51

0%85%+not reached yet

Source: Sample data: a made-up app to show the method, not a benchmark.

Strengths

Bounded: fixed once Day N has passed, so it is safe to report.
Never falls as N grows, which makes it a clean activation or onboarding measure.
Works for apps used on irregular days, because one return anywhere in the window counts.

Traps

A user who came back once on Day 1 and never again still counts for cumulative Day 30.
It says nothing about whether users are active now; pair it with classic.
Easy to mislabel as plain "retention" in exports, so check which column you have.

What can cumulative retention tell you?

What you can conclude

What share of installs you brought back at least once in the first N days.
Whether an onboarding change made more people return early.
Activation quality by channel, once cohorts have passed Day N.

What you can't

How many users are still active on Day N.
Long-term stickiness: the curve only goes up.
Comparison with a classic benchmark.

Compare with the other two

Questions

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.

Can cumulative retention go down?

Not for a given cohort. Moving from Day N to Day N+1 only adds a day to the window, so the share of users who came back at least once can stay the same or rise, never fall.

Is cumulative retention the same as bracket retention?

They are close. Bracket retention counts users active at least once inside a range such as Days 1 to 7; cumulative is that idea always anchored at Day 1 and ending on Day N.

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.

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.