NewThe guide: 11 paywall examples

11 minutes

Paywall design: 6 decisions before the visual

The Adapty 2026 data covers more than 16,000 apps: the LTV gap between the best and the worst paywall is 636%. It comes from six structural decisions, not from the color of the button.

Clément Parot

Paywall design: 6 decisions before the visual

Article summary

  • Hard paywalls convert about 5 times better from install: 10.7% against 2.1% in freemium.
  • 82% of trials start on Day 0, so the paywall sits right after the aha moment.
  • Trials of 17 to 32 days convert at 45.7%, against 26.8% for those of 3 to 7 days.
  • 50 experiments or more on the paywall produce median revenue 18.7 times higher than a single experiment.
  • The documented test order starts with pricing and plans, then trial length, then the visual.

Summarize this article with

Most founders arrive at their paywall with Figma open.

Wrong door. Design is the last thing to optimize, not the first. The Adapty 2026 data, built on 3 billion dollars of subscription revenue and 16,000+ apps, is clear: the LTV difference between the best and the worst paywall in their dataset is 636%. That difference does not come from the color of the button or the layout. It comes from 6 structural decisions that most apps have never deliberately tested.

firstapp works with founders on monetizing their apps. What we see every time: teams that optimize the design before locking the structure lose time and money. We opened eleven paywall examples with that in mind. Here are the 6 decisions to make, in order.

Decision 1: hard or soft paywall?

This is the first structural decision, and it conditions all the others.

A hard paywall blocks full access to the app until payment or the start of a trial. A soft paywall lets the user access part of the product for free before monetizing them. These are not two different designs. They are two different economic models, with radically different implications for the funnel, the metrics, and LTV.

The data: hard paywalls convert about 5 times better from install (10.7% vs 2.1% for freemium, according to RevenueCat SOSA 2026) and generate a 1-year LTV that is 21% higher (Adapty 2026). But soft paywalls convert better at the moment the paywall is shown (4.85% vs 3.34% according to Adapty), because they bring users who have already experienced the value.

To go further on this decision, read our hard paywall vs soft paywall comparison, with the full 2026 data.

What you need to lock before opening Figma: your model. Not your color palette.

Newsletter

Once a week.

Last issue

We opened 10 apps to see how they sell a subscription.

+534 people read us each week

Decision 2: where to place the paywall in the funnel?

When the paywall is shown is one of the highest-impact levers, and one of the least tested.

82% of trials start on Day 0, meaning the day of install (RevenueCat SOSA 2025). That figure says something essential: the user's motivation is at its peak in the first minutes. Every extra screen before the paywall is a chance to lose that motivation.

The basic rule: the paywall should arrive as early as possible after the user has had their click, their aha moment. Not before, because a user who has not yet understood the value will not pay. Not long after, because motivation drops as the onboarding gets longer.

There are several possible placements:

At the end of onboarding, after personalization and the aha moment. This is the most common sequence, and the most effective for the majority of B2C apps.

At the start of onboarding, before personalization. This works for high-intent apps (very niche apps, B2B tools) where the user knows exactly what they are looking for before they download.

In a contextual position, triggered by a specific action (an attempt to access a premium feature). Effective for soft paywalls that want to monetize usage rather than the start.

What does not work: a paywall buried at the end of a 15-screen onboarding, after the user has had time to get bored.

Decision 3: how many plans, and which ones?

Plan structure is an architecture decision, not a design decision.

The basic rule: two plans in most cases. A weekly and an annual. It is the most tested and best documented configuration in the industry. Weekly plans now account for 55.6% of mobile subscription revenue in 2025 (against 43.3% in 2023, according to RevenueCat): they convert better because they reduce the perceived risk of the commitment.

Beyond two plans, cognitive load goes up and conversion goes down. When a user has to compare three, four, or five options, their brain switches into analysis mode, which slows the decision. Tests documented by RevenueCat in 2026 show that an app where 80%+ of conversions were concentrated on two plans improved its conversion by removing the middle options.

On naming: avoid repeating your app or product name in every plan. "AI Assistant Weekly", "AI Assistant Annual", "AI Assistant Monthly": that is useless cognitive load. Users do not read, they scan. Short labels ("Weekly", "Annual", "Best Value") are enough and perform better according to Superwall data (4,500+ A/B tests).

If you want to offer a lifetime plan, place it as a discreet tertiary option, never as the main option.

Decision 4: what trial length?

Trial length is the most under-tested structural variable, and yet one of the most impactful.

Apps with trials of 17 to 32 days convert at 45.7% trial-to-paid. Apps with trials of 3 to 7 days convert at 26.8%. That is a 70% difference in conversion, produced by a single configuration decision (Adapty State of In-App Subscriptions 2026).

The logic: a longer trial gives the user time to fold the app into daily use. A habit that takes 10 days to form cannot be evaluated on a 3-day trial. When the trial ends, the user has not yet had time to see what the app would be like without them. That moment of friction is exactly what triggers conversion.

A few important nuances:

A long trial works better when the app's value shows up over time (fitness, wellness, productivity apps). For apps whose value is immediate and occasional, a shorter trial can be enough.

Trial length should be aligned with your billing cycle. A 7-day trial on a weekly plan can work. A 7-day trial on an annual plan means the user pays for 365 days after testing 7. The value-to-commitment ratio is unbalanced.

To go further on how to improve your paywall conversion, trial length is the first lever to test before touching the design.

Decision 5: what price level?

Price is a structural decision with an impact on LTV that most apps have never measured.

The data is counterintuitive: price tests have a 28.3% success rate on immediate conversion, but 45.5% on LTV (Adapty 2026). Raising your price does not necessarily bring more users, but it brings users who are worth more.

Apps with premium prices (the high end of their category) reach 9.8% trial-to-paid conversion. Apps with low prices reach 4.3% (ArpuBrothers 2025). The explanation: a price that is too low sends an implicit signal about quality. In a market where users pay €10 to €15 per month for premium apps, a price of €2.99/month is often read as suspicious rather than as a good deal.

Another documented point: in Europe, mobile subscription prices rose by 18% in 2025, and Europe now exceeds North America on annual, monthly, and weekly prices (Adapty 2026). The European market has more tolerance for premium pricing than people think.

The question to ask before setting your price: have you tested a price 30% higher? If not, you may be optimizing a design on an undervalued price base.

On how the price is presented, showing the annual plan as a monthly cost with pre-selection increases trial starts by 30% and annual-plan adoption by 10%, according to RevenueCat. That is a presentation decision, not a design decision. It is made before opening Figma.

Decision 6: how do you test systematically?

The paywall is not a screen to finish. It is a permanent experimentation program.

Apps that run 50 experiments or more on their paywall have median revenue 18.7 times higher than those that run only one (Adapty 2026). Apps that run regular A/B tests have 74% more average MRR than those that do not test.

These figures do not come from an app that found a magic formula. They come from teams that installed an experimentation discipline: testing in the right order, the right variables, with enough volume to be significant.

The recommended test order according to the Adapty data:

  1. Pricing and plan structure first (up to 80% potential LTV uplift)
  2. Trial length second
  3. Visual and copy optimization third (up to 30% uplift)
  4. Pricing by country fourth

What most teams do: the reverse. They start with the color of the button and the headline, and they never get to the structural variables that have 2 to 5 times more impact.

Technical rule: you need at least 200 subscriptions per variant to reach statistical significance. If your volume is low, focus on structural decisions with no documented downside rather than on formal tests.

What this changes for the design

Once the 6 decisions above are locked, paywall design has a precise role: to communicate clearly the structure you defined.

What the design must do:

  • Make your app's value immediately readable (benefits, not features)
  • Create a clear visual hierarchy between the plans
  • Reduce friction at the moment of the tap (native look, an obvious button, clear trial conditions)
  • Repeat the trial message at every point of visual tension

What the design cannot do: make up for a trial that is too short, badly structured plans, an undervalued price, or a paywall placed at the wrong moment in the funnel.

As Airbridge sums it up in its 2026 analysis: "If the structure is badly configured, visual iteration produces marginal gains at best." Design comes last. Not out of disregard for aesthetics, but because aesthetics have no impact if the structure is wrong.

FAQ

In what order should you optimize a paywall?

The order recommended by the Adapty 2026 data: pricing and plan structure first (LTV impact up to 80%), then trial length, then the visual and the copy (up to 30%), then pricing by country. Most teams do the reverse: they start with the design and never get to the variables with the most impact.

Does paywall design really affect conversion?

Yes, but less than the structural variables. Visual tests have the lowest success rate of every experiment category according to Adapty 2026, lower than price tests, lower than trial-length tests. That does not mean design does not matter: a badly presented paywall can hurt the conversion of a good model. But a beautiful design cannot save a model that is structurally wrong.

How many plans should you put on a paywall?

Two in most cases: a weekly and an annual. It is the best-performing configuration according to industry data. Beyond two plans, cognitive load goes up and conversion goes down. If you want to offer a lifetime plan, place it as a discreet tertiary option.

What is the best trial length for a mobile app?

Apps with trials of 17 to 32 days convert at 45.7% trial-to-paid, against 26.8% for trials of 3 to 7 days (Adapty 2026). The optimal length depends on your category and the time your app needs to generate perceived value. A fitness or wellness app will benefit from a longer trial than a productivity tool with immediate value. What matters: trial length should be aligned with your billing cycle and the average time to reach the aha moment.

Should you test your price before optimizing the design?

Yes. Price tests improve LTV in 45.5% of cases according to Adapty 2026, against 28.3% for immediate conversion. Testing a price 30% higher on a segment of your traffic before reworking the design is a priority that is often ignored. If your price is undervalued, you are optimizing a design on a fragile economic base.

Newsletter

Once a week.

Last issue

We opened 10 apps to see how they sell a subscription.

+534 people read us each week