Diagnose why YouTube engagement fell by half and decide how to respond

You are a product manager at Google working on YouTube. Your team reports that YouTube usage is down by half. Walk through how you would respond: what you would check first, how you would find the cause, and what you would do about it. ```hint Verify before you diagnose A drop this large is extraordinary. Before hunting for product causes, check whether the number itself is right, and pin down exactly when, where and for whom it fell. ``` ### Clarifying Questions - What does "usage" mean here: watch time, daily active viewers, sessions or video starts, and is it measured the same way as in the comparison period? - Over what window is the 50 percent drop measured, and did it happen suddenly or decline gradually? - Is the drop global and on every surface (mobile apps, web, TV apps, Shorts, YouTube Music), or concentrated in certain platforms or regions? - Did viewing fall alone, or have creator uploads fallen too? ### What a Strong Answer Covers - Validating the metric definition and the data pipeline before treating the drop as real - Segmenting the drop by time, platform, app version, region, product surface and user type to localize it - Internal and external hypotheses, ordered by likelihood given the drop's shape - Concrete checks that confirm or rule out each hypothesis, and immediate mitigation once the cause is found - Communication, prevention and monitoring after recovery ### Follow-up Questions - The drop is real and limited to the newest app version on one platform. What are your next three actions? - Watch time is down by half while daily active viewers are flat. How does that change your hypotheses? - How would you design alerting so a drop like this is caught within an hour?