paywall
Also called: paywalls, paywall screen, upgrade screen
The screen that asks a user to pay or start a trial to use a feature. Its conversion depends on what triggered it as much as on its design.
A paywall is the screen where a free user is asked to subscribe or start a trial. In a subscription app it's the single busiest point of monetization, and small changes to it move revenue noticeably. Paywalls differ along a few lines:
- Hard or soft. A hard paywall blocks the product until the user pays or starts a trial; a soft (freemium) paywall leaves the core free and gates extra features.
- Where it appears. An onboarding paywall is shown to every new user right after signup; a contextual (feature-gated) paywall appears when a user reaches for a paid feature, such as scanning a receipt; a metered paywall appears after a free allowance runs out.
- What it offers. Plans (monthly and yearly, often with the yearly price shown per month), a trial or not, the price, and the reasons to pay.
The basic metric is paywall conversion: trial starts (or purchases) divided by paywall views. Decide whether you count views or unique viewers, because one user can see the paywall many times; per-viewer conversion is usually the more honest number. Then follow the funnel further: view → trial start → first payment → first renewal. A paywall change that lifts trial starts but attracts people who cancel before paying hasn't helped.
The most common reading error is ignoring the trigger mix. Users who hit the paywall while trying to do something valuable convert much better than those who wander into it from settings. If the share of high-intent triggers grows (a new season, a new feature, a change in onboarding), overall paywall conversion rises even though the paywall itself didn't change. Always break conversion down by trigger before giving the paywall credit.
Example
Halves shows its paywall from four triggers. Conversion to trial start in May and July (the paywall didn't change):
Overall conversion rose from 560 ÷ 10,000 = 5.6% to 800 ÷ 10,000 = 8.0%, yet every trigger converts exactly as before. The whole rise comes from the mix: in July travel season, multi-currency (12%) grew from 10% to 40% of views. Maya shouldn't credit the new paywall copy that shipped in June; the result to report is "no change per trigger, more high-intent traffic".
Common mistakes
- Crediting the paywall for a mix shift. Break conversion down by trigger, platform and plan before comparing periods.
- Views instead of viewers. Repeated views inflate the denominator and hide how many people actually decided.
- Stopping at trial starts. A variant that wins on trials can lose on first payments and renewals.
- Testing price and design at once. Change one thing per test, or you won't know which one worked.
- Ignoring the users who never see it. A paywall shown later reaches fewer people; judge revenue per new user, not only conversion per view.