Trials and retention

Trial → paid

The best categories convert a little under half of trials: 48.7% (median, Travel — the highest category) RevenueCat 2025 and 43.8% (median, Media & Entertainment) RevenueCat 2025. Longer trials do better — Trial → paid by trial length: 45.7% (median for trials of 17–32 days — the best-converting length) RevenueCat 2025. Judge your own rate against thirty to fifty percent, not against a hundred; and judge it late, because a trial started near the end of any window has not had time to convert. Broctic labels its store-side trial → paid provisional for exactly that reason.

RevenueCat 2025: RevenueCat sells subscription infrastructure, so the sample is apps that chose that tooling — better-resourced than the App Store as a whole. Medians and percentiles are reported; exact app counts per figure are not.

The first month is the retention problem

Annual subscriptions cancelled in the first month: ~30% RevenueCat 2025. Not month six — month one. People who paid for a year and cancelled in thirty days had the outcome promised and did not see it delivered in their first few sessions. Everything after that month retains far better, which is why the fix is in the product's first week, not in a win-back email.

• Deliver the core outcome in the first two sessions. If the onboarding answers were used to promise something, the first screen after the paywall should start on it.
• Notify around the outcome, not around the app. A reminder that references the user's own goal is the only kind that earns a re-open.
• Measure it: a custom funnel from subscription_started to subscription_cancelled over a 30-day window, split by the onboarding answers in props, shows which promise is not being kept.

Refunds

Refunds run 3.4% (freemium apps) RevenueCat 2025 for freemium and 5.8% (hard-paywall apps) RevenueCat 2025 for hard-paywall apps; Health & Fitness and Education sit at 4.71% (Health & Fitness) RevenueCat 2025 and 4.86% (Education) RevenueCat 2025. Two causes dominate every practitioner account: a charge the user did not see coming, and a first session that did not deliver what the listing promised.

• Write the trial length, the price after it and the charge date on the paywall, and send a reminder the day before the charge. A reminder reduces refunds by more than it reduces conversion — a refunded subscriber counts as paid, then as nothing.
• Make cancellation obvious inside the app. A hidden cancel does not keep the subscriber; it turns a cancellation into a refund request, and Apple grants most of them.
• Refunds are netted out of proceeds in Apple's report and in Broctic, so a rising refund rate shows up as proceeds that do not match paid units.

Renewal and win-back

Year-one renewal is where plan choice is decided: Renewal rate at year 1: 3.4% (median, weekly plans) RevenueCat 2025, 17.0% (median, monthly plans) RevenueCat 2025, and annual plans around half. Win-back offers — a reduced price to a lapsed subscriber, which Apple supports natively — have no published conversion figure; the consensus is that they work on people who cancelled for price and not on people who cancelled for lack of use, which is the majority. The better investment is the first month.