Metrics question

There has been a drop in the number of Facebook groups created. Why?

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

Structured root-cause diagnosis of a metric decline, testing hypothesis generation and prioritization.

Confirm the drop is real by ruling out a tracking change or a definition change in what counts as a 'group created' event.
Segment by platform, region, and user tenure to see if the drop is broad or concentrated in a specific cohort.
Check for a recent product change: a UI change to the create-group flow, a new content policy, or increased friction like added verification steps.
Check upstream funnel metrics: are fewer users reaching the create-group entry point, or are they abandoning the form itself.
Consider external factors: seasonality, a competing platform's growth, or a broader shift toward private messaging over group-based interaction.
Propose next steps based on the most likely cause, such as an A/B test reverting a recent flow change.

Distinguishes a top-of-funnel drop (fewer people starting) from a form-abandonment drop (fewer people finishing), which need different fixes.
Names a plausible concrete cause, like added verification friction from a recent policy change, as a first hypothesis.
Gives an illustrative finding, e.g. the drop concentrated on Android after a recent app update, marked as an assumption.
Considers the broader behavioral shift toward private messaging as a genuine external hypothesis, not just an internal bug.

Jumping to one cause without segmenting the data to see where the drop concentrates.
Ignoring broader platform or behavioral trends outside the immediate feature.

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