Root-Causing a 20% Monthly Active User Drop on the Instagram Feed

You are the product manager for the Instagram feed, and you see that the feed's monthly active users (MAU) have fallen 20%. Explain how you would figure out why. ```hint Split the total into flows A monthly active count is made up of people who stayed, people who are new and people who came back. Find which of those shrank, and where, before you list reasons. ``` ### Clarifying Questions - How is a feed monthly active user defined: anyone who opens Instagram, or only an account that views at least one post in the home feed during the month? - Is the 20% month over month or year over year, and did daily feed users fall suddenly on one date or slide over several weeks? - Is the decline in every country and on every platform (iOS, Android, web), or concentrated in some of them? - Did total Instagram MAU fall too, or only the number of users reaching the feed? ### What a Strong Answer Covers - Confirming the metric definition and data integrity, including any change in what counts as the feed, before any product explanation. - Timing and shape read from daily data, allowing for the lag in a rolling monthly count and for seasonality. - Separating a feed-specific drop from an Instagram-wide one, and decomposing MAU into new, retained and resurrected users and their entry paths. - Segmentation by geography, platform, app version and user cohort to localize the decline. - Ranked internal and external hypotheses, including attention moving to other parts of the app, each with the evidence that would confirm it. ### Follow-up Questions - Total Instagram MAU is flat and only feed MAU fell. What does that suggest, and is it bad for the business? - You trace the drop to a ranking change that showed neutral results in its experiment. How could that happen, and what would you do? - How would you set up monitoring so a drop like this is caught within days rather than at the monthly review?