Improved activation and subscriptions
in the UGC app
Role: Product Designer (mobile direction)
Period: 3 weeks
Format: A/B experiment, UX research, product analytics
Period: 3 weeks
Format: A/B experiment, UX research, product analytics
Context
UGC entertainment app with feed, subscriptions, and saves.
Goal: to increase the Activation Rate (D1 Active) and Trial→Paid conversion, while maintaining user trust and not increasing churn.
Problem
UGC entertainment app with feed, subscriptions, and saves.
Goal: to increase the Activation Rate (D1 Active) and Trial→Paid conversion, while maintaining user trust and not increasing churn.
Problem
- Research and analytics have shown:
- The bounce in the first 60 seconds exceeds 35 %
- The session before the first save is too short (TTV ≈ 2 min)
- The cold feed — without personalization
- The paywall is shown too early, without revealing the value of the product
- Hypotheses
- H1. Quick personalization of interests at the start + a "warm" feed → ↑ activation
- H2. Paywall at the time of “value discovery" (after 3 relevant cards) → ↑ trial-conversion
- H3. Social proof (ratings, number of active users, reviews) → ↓ doubts when paying
Decisions
- New onboarding: Choosing 3+ interests before the first launch
- Dynamic feed: relevant content with a hint "Save the first post — recommendations will become more accurate"
- Paywall after value moment — showing when 3 interactions are reached (save/like)
- Social proof and transparent value-prop ("7 days free • cancellation in 1 tap")
Additionally tested the paywall in stories — a soft upsell after viewing the content
Added paywall to stories
Результаты
| Metrics | Before | After | Δ | Comment |
| Activation Rate (D1 active) | 41 % | 53 % | +12 п.п. | retention growth on the first day |
| Trial Conversion | 4 % | 6.1 % | +75% rel | the best moment of the show |
| Time-to-Value (первое сохранение) | 2:10 мин | 1:34 мин | −28 % | they reach the “value moment” faster |
| Day-7 Retention | 17 % | 21 % | +4 п.п. | The trend has been stable for 4 weeks |
The key conclusion
Optimization of the paywall moment and value-oriented UX increased activation and conversion without increasing negativity and falling reviews.
The A/B test is completed with a p-value of < 0.03 and a statistical power of > 0.8.
Further steps
Check the stability of the effect on new traffic sources.
Expand personalization based on actual behavior (implicit signals).
A/B/n test of various paywall layouts (tariff matrix, video offer).
Optimization of the paywall moment and value-oriented UX increased activation and conversion without increasing negativity and falling reviews.
The A/B test is completed with a p-value of < 0.03 and a statistical power of > 0.8.
Further steps
Check the stability of the effect on new traffic sources.
Expand personalization based on actual behavior (implicit signals).
A/B/n test of various paywall layouts (tariff matrix, video offer).