Which Events to Track Besides Purchases in Mobile Subscription Experiments

Activation events mark the moment users experience genuine value from your app. These might include completing a first meaningful action, reaching a usage threshold like three sessions, or successfully using your core feature. Unlike purchases, activation events predict long-term retention because they measure actual product-market fit at the individual level. Users who activate retain at dramatically higher rates regardless of whether they've paid yet.

Define activation based on behavioral cohort analysis—identify which early actions correlate most strongly with week-8 retention among your current subscribers. Common activation events include creating content, completing onboarding tutorials that teach core features, inviting others, or achieving a first successful outcome. Track time-to-activation across variants, since faster activation typically predicts both higher conversion and better retention.

Paywall view rates, time spent on paywall screens, and paywall dismissal patterns reveal how onboarding variants affect purchase consideration. A variant might increase purchases simply by showing the paywall more aggressively or earlier, without actually improving perceived value. Tracking these interactions separately lets you distinguish between variants that create more purchase-ready users versus those that simply apply more pressure.

Monitor which paywall users see first, how many times users encounter paywalls before converting, and whether users dismiss paywalls to continue exploring or abandon the app entirely. Segment paywall metrics by session number—purchasing on session one often indicates different quality than purchasing after experiencing the product across multiple sessions. These patterns help identify whether conversion gains come from better value communication or just friction changes.

Track how many features users engage with and how deeply they use each one during onboarding. Shallow engagement with many features often indicates confusion or random exploration, while deep engagement with fewer features suggests focused value discovery. Variants that increase breadth without depth may create an illusion of engagement without building genuine habits or understanding.

Measure session length, actions per session, and return visit patterns within the first week. Users who return multiple times before seeing a paywall typically convert at higher rates and retain better than those who purchase immediately. Feature adoption sequences also matter—users who follow your intended discovery path often activate more reliably than those who encounter features randomly, even if total usage appears similar.

For apps offering free trials, tracking trial starts separately from completed purchases reveals important onboarding quality signals. High trial start rates with low trial-to-paid conversion suggest users feel pressured to commit before understanding value. Low trial starts with high conversion rates often indicate better value communication that makes users confident enough to commit to a trial period.

Monitor trial engagement patterns—users who actively use the product during trials convert to paid at much higher rates than those who start trials but remain dormant. Track which onboarding variants produce engaged trial users versus those who sign up but never return. Trial cancellation timing also provides insight, as immediate cancellations suggest user experience problems while late cancellations may indicate value communication issues.

Build a measurement framework that connects early behavioral signals to downstream revenue through cohort analysis. Identify which combinations of activation, feature adoption, and engagement metrics predict 30-day and 90-day revenue per user. This lets you evaluate onboarding variants on leading indicators while maintaining confidence they'll drive actual business outcomes, rather than waiting months for revenue data.

Tools like Spinosis that attribute RevenueCat purchases when App User IDs match experiment user IDs let you track both immediate behavioral metrics and eventual revenue outcomes within a single experiment. Configure your tracking to capture the full funnel—from first session events through activation milestones, paywall interactions, trial behavior, and finally attributed purchases. This complete view prevents optimizing for metrics that don't actually predict revenue or retention.