Android’s subscription conversion rate, compared to iOS, looks like a platform problem. When RevenueCat analyzed over 115,000 apps and $16 billion in revenue for the 2026 State of Subscription Apps report, the numbers were clear: on Android, the median download to paid conversion at day 35 sits at 0.9%, while iOS lands at 2.6%. That’s a nearly threefold gap. The instinct is to blame the trial, or the audience, or some fundamental difference in how Android users behave. The data says otherwise.
Looking one level deeper in the same dataset, trial to paid conversion on Android is 32.5%. On iOS, it’s 32.6%. Once a user starts a trial on either platform, they convert at statistically the same rate. There is an important caveat: Android’s trial-starter pool is likely more filtered. Because Android surfaces fewer trials overall, the users who do start one tend to be higher-intent. That selection effect partly explains why the rates equalize. Even so, the primary lever is clear: Android apps are sending far fewer users into that first stage. Closing the gap starts with getting more users to begin a trial in the first place.
In this article, you’ll explore the two-stage Android paywall funnel and where Android apps lose users, how Google Play’s offer and tag system controls which subscription option is surfaced to users, how RevenueCat’s SubscriptionOptions selection works and where silent misconfigurations occur, what the data shows about hard paywalls versus freemium models, and how to use RevenueCat Experiments to close the gap systematically.
The fundamental problem: Two stages, one broken
The download to paid journey has two distinct stages, and they behave differently.
The first stage is download to trial start: the user installs your app, reaches your paywall, and decides whether to begin a free trial. The second stage is trial start to paid: the trial ends, and the user decides whether to continue as a subscriber.
The RevenueCat data shows these two stages behave very differently on Android. A user who starts a trial converts at 32.5%, close to the 32.6% on iOS. But Android apps are sending far fewer users into that first stage. The bulk of the conversion gap lives in stage one.
One data point makes this concrete: 89.4% of all trial starts happen on the day of install. Users who download with high intent act immediately. Users who do not start a trial on install day rarely return to do so later. That makes the first paywall impression on Android the moment that determines most subscription revenue. Everything downstream from that moment, including your trial experience, your onboarding, your product, performs about as well on Android as on iOS. The question is whether users reach that moment at all.
Stage one: What determines whether users see a trial
Two things control whether a user is presented with a free trial on Android: what you show (the paywall type) and when you show it (the timing). Both are fully within your control.
The paywall type gap
The RevenueCat data breaks down paywall models by D35 download to paid conversion, RevenueCat’s measurement window for capturing conversion across trial lengths up to one month. Hard paywalls, where users must interact with a subscription offer before accessing core features, achieve a median D35 conversion of 10.7%. The top 10% of hard paywall apps reach 38.7%. Freemium models, where users get some access without paying, convert at a median of 2.1%.
That’s a fivefold difference in conversion with nearly identical annual retention. Hard paywalls retain 27% of subscribers at 12 months. Freemium retains 28%. For most app categories, the hard paywall numbers are substantially better. If your product delivers clear, immediate value in a single session, a hard paywall is almost certainly the right model. Categories where freemium remains appropriate are those with network effects or long value-discovery cycles, such as social apps and community tools, where acquiring a broad user base matters before monetization.
| Paywall model | Median D35 conversion | Top 10% D35 conversion | 12 month retention |
|---|---|---|---|
| Hard paywall | 10.7% | 38.7% | 27% |
| Freemium | 2.1% | — | 28% |
There is one case where freemium shows a