Metrics question

You are working as PM of a social media app. You noticed that DAU of the app fell by 50%. How would you figure out what might have happned?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Root cause diagnosis under a sudden, large metric drop, distinguishing measurement issues from real user behavior change.

First rule out a measurement or tracking bug, since a 50 percent drop is often instrumentation error, not real behavior.
Check the timing: sudden versus gradual, since sudden drops point to a technical or tracking cause.
Segment the drop: uniform across iOS, Android, and web, or concentrated on one, pointing to a platform specific bug.
Check for external causes: an app store outage, a third party login dependency down, or a competitor launch.
Check for internal causes: a recent release, an A/B test gone wrong, or a notification policy change.
Once the cause is isolated, define the fix and a recovery metric, like DAU returning to baseline within a set number of days.

Leads with ruling out a tracking bug, since real 50 percent single day DAU drops are rare and usually not organic.
Segments systematically by platform and geography instead of guessing at causes.
Ties the investigation to a timeline, correlating the drop with recent releases or external events.
Gives an illustrative number, for example assuming the drop is isolated to iOS only, pointing to an app store issue.

Jumping straight to user behavior explanations before ruling out technical causes.
Not segmenting the drop by platform, geography, or cohort before proposing a fix.

Chapters of the AI PM course that teach what this question tests.