Monetization
Paywall A/B Testing
Test paywalls, pricing, and offers against each other in production — no app release, no engineering queue.
Most monetization ideas die in the backlog. You want to try a new paywall headline, a different price, a quarterly plan, a free-trial variant — but every one of those changes means a code change, a build, an app review, and a wait. By the time it ships, the idea is stale and you still can't tell whether it actually moved revenue. Paywall A/B testing on Superwall removes that whole loop. You build variants in the editor, split traffic between them, and read the results — all from the dashboard, with no new app release.
Test anything, fast, without shipping a build
A Superwall experiment is just two or more paywalls shown to the same audience. You pick the audience, open the campaign's Paywalls tab, and add the variants you want to compare. Because paywalls are configured remotely and preloaded and cached on-device, a new test goes live for real users instantly — no SDK upgrade, no store submission, no engineering ticket to coordinate.
See how experiments go liveThat means the people closest to the funnel can run the test themselves:
Copy and design. Headlines, layout, imagery, social proof — anything you can build in the editor is something you can test.
Pricing. Compare price points, monthly vs. annual framing, and trial-vs-no-trial without touching product configuration in code.
Offers. Pit intro offers, discounts, and promotional paths against each other to see which actually converts.
Placement. Test which moment in the app triggers the paywall, since experiments run per-audience inside a campaign.
How it works
Inside a campaign, you define an audience and attach your paywall variants to it. You then set a presentation percentage for each variant — say 60/40, or an even three-way split. Those percentages normally total 100%, and every user who matches the audience is given a sticky assignment to one variant, so they keep seeing the same paywall for the life of the test and your measurement stays clean.
If you set the percentages to total less than 100%, the remaining share becomes a holdout — a control group that sees no paywall at all. That lets you measure the lift of showing a paywall in the first place, not just which design wins. You can adjust allocations, drop a losing variant by setting it to 0%, or reset assignments at any time from the same screen.
Read results you can act on
Superwall reports each experiment on real revenue, not just taps. For every variant you see conversions, trial starts, subscription starts, proceeds, and users, with the default graph showing proceeds per user. A per-placement breakdown shows which trigger points are driving the numbers.
Read experiment resultsTo keep you from chasing noise, the dashboard surfaces confidence intervals on metrics like conversion rate and proceeds per user, so you can tell a real difference from random variance before you commit. And because the results are framed around proceeds — a high-conversion paywall selling a cheaper plan can still lose to a lower-conversion paywall selling a richer one — you're judging variants on the outcome that matters, not a vanity metric.
Why growth teams reach for it
Paywall experimentation is the single most common reason teams come to Superwall — it's the job buyers describe first and most often. The pattern is always the same: the team has a backlog of monetization hypotheses and no fast, safe way to put them in front of real users. Hard-coded paywalls turn every idea into an engineering project, and a slow release cycle means you learn the answer weeks late, if at all.
With remote paywalls and built-in experiments, the loop tightens to days. Growth and marketing can ship a test in the morning, watch it accrue revenue data through the dashboard, and roll the winner out to 100% — all without pulling an engineer off their roadmap. The experiments don't stop at the paywall, either: the same audience-and-variant model lets you test onboarding flows and the handoff into the paywall as one continuous funnel.
Get started
Drop in the SDK once, then run every future experiment from the dashboard. Start with the guide to starting an experiment, review the pre-launch checklist before you go live, and create a free Superwall account to build your first variant.
Start your first experiment