Field Notes
Reading onboarding drop-off by first-session intent
Aggregate onboarding curves hid that “browse first” users and “start lesson” users left at different screens in a language-learning app.
A language-learning product booked a Drop-off Diagnosis Session on their free-trial gate. Their chart showed a smooth decline from install to trial start. Designers had already rewritten the paywall headline twice.
When we split first-session intent using an existing survey prompt — “browse courses” versus “start a lesson now” — the cliff moved. Browse-first users abandoned on a forced account wall before any lesson sample. Lesson-first users reached the paywall and hesitated on the annual plan default.
Workshop takeaway
One onboarding funnel was doing two jobs. The session left the team with separate hypotheses: delay the account wall for browsers, and default monthly for lesson-starters who had not yet tasted value. Aggregate curves alone would have kept them polishing the wrong screen.
When you examine in-app conversion funnel analysis on onboarding, ask whether intent was captured early enough to segment the path.