Metrics to track for a mobile app user retention strategy fall into four categories: retention curve metrics (D1, D7, D30, D90), engagement depth metrics (actions per session, core action completion rate), churn prediction signals (session frequency decline, notification opt-out rate), and re-engagement effectiveness metrics.
Most mobile apps track D30 retention as their primary retention metric and miss the three earlier signals that predict whether D30 will be healthy or catastrophic. By the time you see D30 decline, the cohort that caused it is 30 days gone.
Category 1: Retention Curve Metrics
The retention curve shows the percentage of users from each acquisition cohort who return on a given day after install.
Standard retention intervals to track:
| Interval | Benchmark (Casual Mobile App) | Benchmark (Utility App) | What It Measures | |---|---|---|---| | D1 | 25–35% | 40–50% | First-impression quality | | D7 | 10–20% | 20–30% | Core loop engagement | | D30 | 5–10% | 10–15% | Habit formation | | D90 | 2–5% | 5–8% | Long-term retention |
How to read the retention curve: A steep drop from D1 to D7 (more than 50% loss) indicates an onboarding problem — users tried the app and didn't see enough value to return. A steep drop from D7 to D30 indicates a core loop problem — users understood the app but the loop wasn't engaging enough to sustain a habit.
According to Lenny Rachitsky's writing on mobile engagement, the D1-to-D7 ratio is the leading indicator of your onboarding effectiveness — if you're losing more than half of D1 users by D7, every acquisition dollar you spend is funding a leaky bucket.
Category 2: Engagement Depth Metrics
Retention tells you users came back. Engagement depth tells you what they did when they returned.
Key engagement depth metrics:
- Core action completion rate: % of sessions that include the defining core action (for a fitness app, logging a workout; for a finance app, checking account balance or making a transaction)
- Actions per session: Average number of distinct interactions per session
- Session length: Average time per session (compare against your target engagement model — a 30-second utility app session is healthy; a 30-second social app session is not)
- Feature breadth: Average number of distinct features used per active user per month (low feature breadth indicates users are stuck in one workflow)
Category 3: Churn Prediction Signals
These are the leading indicators of churn that appear 7–14 days before a user goes dormant.
Early churn signals to monitor:
- Session frequency decline: User who averaged 5 sessions/week drops to 1 session over two consecutive weeks
- Notification opt-out rate: Rising opt-out rate by cohort or OS version
- Core action skip rate: User uses the app but does not complete the core action for 3+ consecutive sessions
- Error encounter rate: User who encountered an error in the past 7 days has 3x the churn rate of error-free users
According to Shreyas Doshi on Lenny's Podcast, the most powerful retention intervention in mobile is a session-frequency-based trigger that fires a re-engagement message (push, email, or in-app) when a previously active user's session frequency drops by 50% or more over a 7-day window — this signal predicts dormancy with 70%+ accuracy and the intervention window is roughly 5–7 days before the user fully disengages.
Category 4: Re-engagement Metrics
- Push notification open rate by message type: Benchmarks vary — transactional push achieves 10–15% open rate; promotional push achieves 3–7%
- Re-engagement campaign conversion rate: % of dormant users (0 sessions in 14 days) who return to the app within 7 days of receiving a re-engagement message
- Reactivation-to-retention rate: % of reactivated users who remain active 30 days after reactivation (this measures whether reactivation campaigns attract users who actually stick)
Cohort-Based Analysis
Track all retention metrics by acquisition cohort, not just overall DAU. Overall DAU can stay flat while cohort retention deteriorates — this is the "zombie metrics" trap where new user acquisition masks the fact that older cohorts are churning faster.
According to Annie Pearl on Lenny's Podcast, the retention metric that saved Calendly's product team from a false positive growth signal was cohort-level D30 retention by acquisition month — overall DAU was growing but D30 retention for newer cohorts was declining, indicating that acquisition quality was degrading even though the top-line numbers looked healthy.
FAQ
Q: What metrics should you track for a mobile app retention strategy? A: Track four categories: retention curve metrics (D1, D7, D30, D90), engagement depth metrics (core action completion rate, actions per session), churn prediction signals (session frequency decline, notification opt-out rate), and re-engagement effectiveness metrics.
Q: What are good D7 retention benchmarks for mobile apps? A: Casual mobile apps should target 10-20% D7 retention. Utility apps (productivity, finance, health) should target 20-30% D7 retention. Below these benchmarks indicates a core loop engagement problem.
Q: What is the D1-to-D7 ratio and why does it matter? A: The D1-to-D7 ratio measures what percentage of D1 users return by day 7. Losing more than 50% between D1 and D7 indicates an onboarding problem — users aren't seeing enough value to form a return habit.
Q: What are the early warning signals of mobile app churn? A: Session frequency declining by 50% or more over 7 days, notification opt-out rate increasing by cohort, core action skip rate for 3+ consecutive sessions, and encountering an error in the past 7 days (which triples churn rate).
Q: How do you measure re-engagement campaign effectiveness? A: Track re-engagement campaign conversion rate (dormant users returning within 7 days) and reactivation-to-30-day-retention rate. The second metric reveals whether campaigns attract users who actually stick or just generate a one-session spike.
HowTo: Track Metrics for a Mobile App User Retention Strategy
- Instrument cohort-based retention curves tracking D1, D7, D30, and D90 separately for each acquisition month to avoid the zombie metrics trap
- Define your core action and track core action completion rate per session as the primary engagement depth metric
- Build session frequency monitoring to detect when active users drop 50 percent or more below their baseline frequency over a 7-day window as a churn prediction trigger
- Track notification opt-out rate by cohort and OS version as an early warning signal of deteriorating push strategy effectiveness
- Measure re-engagement campaign effectiveness with both conversion rate (dormant users returning) and 30-day retention rate post-reactivation to distinguish true re-engagement from one-session spikes