⁠State of Paywalls 2026: see what works for 78,000+ paywalls
Get the free skill

Turn your paywall into a prioritized test plan

  • Upload a screenshot
  • Get 5-8 test hypotheses tailored to your paywall, ranked by what to try first, with expected impact for each

Get the Paywall Teardown Skill

By submitting the form, you are agreeing to the Terms of Use and Privacy Policy.

Built on real paywall analyses

  • 30K+apps
  • 105K+paywalls analyzed
  • 40+paywall patterns

What you get

A test plan for your paywall

Upload a screenshot. Get 5-8 test hypotheses tailored to your paywall, ranked by what to try first, with expected impact for each.

Get the free skill
Claude's paywall teardown table: test ideas with their priority, the pattern behind each, where it is seen and the expected CR and ARPU impact
Claude's paywall teardown table: test ideas with their priority, the pattern behind each, where it is seen and the expected CR and ARPU impact

Getting started

Install in a minute

No prompts or configuration required.

  1. Download the skill

    Get the paywall-teardown.zip file.

  2. Add it to Claude

    Go to Settings → Customize → Skills and upload the ZIP.

  3. Start a teardown

    Turn it on and drop in any paywall screenshot.

Your first paywall teardown is a screenshot away

By submitting the form, you are agreeing to the Terms of Use and Privacy Policy.

Frequently asked questions

What is a paywall teardown?

A paywall teardown is a structured review of a subscription app's paywall against known conversion patterns: plan layout, price framing, badges, social proof, trust copy, and the call to action. The output is a list of A/B test ideas ranked by expected impact on conversion rate (CR) and average revenue per user (ARPU).

What should I A/B test first on my paywall?

Start with contradictions and missing trust signals, then move to redesigns. Two common examples are a "Best value" badge on a plan that isn't the cheapest per week and a buy button with no price line under it. Both cost conversion on every paywall view. In Adapty's teardowns, fixing badge placement carries an expected CR lift of 5–12%, while copy tweaks usually land at 3–8%.

What makes a subscription paywall convert?

High-converting paywalls share five patterns:

  • One recommended plan, marked by both the badge and the layout
  • A per-week price on every plan
  • A savings figure on the annual plan
  • Social proof near the CTA
  • A line under the button with the price and cancellation terms

The benefit copy describes outcomes for the user instead of features of the app.

Where should the "Best value" badge go on a paywall?

Put it on the plan you want users to pick, usually the annual one, and on only one plan. The badge, the card highlight, and the per-week math should all point to the same plan. If the badge sits on one plan and the highlight on another, users see two recommendations and hesitate.

Should a paywall show weekly prices?

Yes, on any paywall with more than one plan. A per-week equivalent lets users compare monthly and annual plans at a glance and makes the annual discount visible. Pair it with a savings figure on the annual card, such as "Save 68% vs weekly." In Adapty's teardowns, visualizing annual savings carries an expected ARPU lift of 10–15%.

How much does social proof lift paywall conversion?

Social proof next to the purchase button, such as an App Store rating with a review count, carries an expected CR lift of 10–15% in Adapty's teardowns. The effect is strongest in categories where users need proof before they pay, such as finance or navigation apps.

Do countdown timers increase paywall conversion?

Only when a real discount is ending. On a full-price paywall, a timer creates fake urgency that costs more trust than it earns. Use timers for time-limited offers like holiday sales or win-back discounts, and leave them off standard paywalls.

Can I use AI to analyze my paywall?

Yes. Adapty's free Paywall Teardown Skill for Claude reads a paywall screenshot and checks it against patterns from Adapty's teardowns of top subscription apps. It returns 5–8 A/B test hypotheses ranked by expected impact. Without a pattern library, general-purpose AI tends to give generic advice like "add social proof" and doesn't check what the screen already has.

© 2026 Adapty Tech Inc.