Cohort Rebuild Study

Retention charts move for reasons that have nothing to do with product quality: SDK upgrades, timezone mishandling, anonymous-to-known identity merges, or an onboarding rename that breaks day-one definitions.

This study rebuilds cohort windows from raw event history, verifies identity continuity across platforms, and tests early-session behaviours against return rates. We state sample limits plainly and refuse to dress pipeline artefacts as insights.

You receive a shortlist of behaviours worth protecting, a list of artefacts to stop celebrating, and a reusable cohort outline your analysts can maintain after we leave.