Metrics question

You have seen a 10% decrease in total bookings on Airbnb this month. What will you do?

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 ambiguity, a core PM skill for responding to a sudden, unexplained metric decline with structured investigation.

Confirm the scope of the decline: is it global or concentrated in specific regions, listing types, or user segments.
Check for external explanations first, such as seasonality, a major news event, or a competitor promotion.
Check for internal explanations: a recent app change, pricing update, or a bug in the booking flow.
Segment the data by funnel stage to see if the drop is in search, listing views, or final checkout.
Once the cause is isolated, propose a targeted fix and a plan to monitor recovery week over week.

Checks external causes, seasonality or a competitor event, before assuming an internal product bug, avoiding a wrong first guess.
Segments the decline by funnel stage, distinguishing a demand problem from a conversion problem.
Proposes a monitoring plan after the fix, showing the investigation does not stop at diagnosis alone.
Assumes an illustrative comparison, such as checking if the 10 percent drop matches the same period last year.

Jumps straight to a fix without first isolating whether the cause is external or internal.
Looks only at the aggregate number and never segments by funnel stage, region, or listing type.

What would you check first if the drop turned out to be seasonal?
How would you rule out a tracking or data pipeline error before concluding it's a real decline?
How would you communicate this investigation's findings to leadership under time pressure?