Only 20% of users are affected — why does the metric look flat?

Experiment designMedium

Problem. A new feature only fires for users who hit a specific screen — about 20% of users. Your whole-population OEC barely moved. Is the feature a failure?

Before you reveal: say your answer out loud, as if you were in the real interview — get your reasoning across clearly first. There is no single correct answer: reading what the interviewer is really after and defending your own thinking is what makes an answer strong.

← Your feature spreads through the friend graph — is the A/B biased?Does the app being slower actually cause churn? →