PostHog Funnel Analyzer gives product teams a bounded, aggregate view of funnel friction without exposing person-level data or turning correlation into a product diagnosis.
It queries exact event names through PostHog's read-only Query API, normalizes ordered-funnel steps, compares two explicit date windows, and returns daily event trends. The companion brief identifies the largest observed numeric drop-off but labels it as a measurement, not a reason. It never requests user profiles, session recordings, request properties, or write permissions.
What you can do
- analyze_funnel — calculate aggregate counts, step conversion, and drop-off
- compare_funnels — compare the same event sequence across two explicit windows
- analyze_event_trends — read bounded daily totals for up to ten events
- build_friction_brief — preserve observations and generate safe validation steps
Who it's for
Product managers, growth teams, founders, analysts, developers validating instrumentation, and agencies improving onboarding or checkout flows.
How to use it
- Connect a personal API key with Query Read permission only
- Supply two to ten exact event names in their intended order
- Choose a bounded date range and conversion window
- Compare equal windows before interpreting movement
- Validate the largest observed drop with instrumentation, qualitative research, and release context
Getting started
Start with a two-to-five-step funnel over 30 days. PostHog answers where aggregate drop-off was observed; the tool deliberately does not claim why a user left or which product change should ship.
