User retention is the percentage of users who come back and perform a meaningful action after their first visit or signup. It’s the antidote to vanity growth: instead of counting raw signups or unique visitors, you measure whether people stick around, engage, and generate value over time. In analytics, retention is usually tracked by cohorts—groups of users who started in the same period—see cohort analysis.
How to calculate user retention
Formula:
Retention rate (%) = (Retained users at period end ÷ Users at period start) × 100
- Users at period start: the cohort size (e.g., all signups on Week 1).
- Retained users: members of that cohort who returned and did a qualifying action (session, purchase, feature use—define it explicitly).
- Period: fixed window (D1, D7, D30; M1, M3, M6). Pick one that matches your product cycle.
Mini-example:
You acquired 200 users on July 1. By July 31, 56 of those exact users were active.
Retention D30 = 56 ÷ 200 = 28%.
Sample cohort table (fixed 30-day windows)
| Month since signup | Retained users (of 100) | Retention % |
|---|---|---|
| 0 | 100 | 100% |
| 1 | 38 | 38% |
| 2 | 26 | 26% |
| 3 | 19 | 19% |
Why user retention matters
- Compounds growth: Strong retention multiplies the impact of acquisition and lowers blended CAC.
- Signals product–market fit: Flat or rising retention curves suggest the core loop works.
- Drives LTV: Higher retention → higher lifetime value and healthier unit economics.
Practical notes & pitfalls
- Define “active” once, reuse everywhere. Tie retention to the same action you use for engagement rate or activation.
- Mind cohort hygiene. Don’t mix acquisition sources or platforms if behavior differs.
- Watch churn alongside retention. Churn rate = 100% − retention; track both to spot leakage early. See churn rate.
- Beware seasonality and one-off promos. Spikes in conversion rate without matching retention often mean low-intent traffic.