Metrics question

The number of matches on Tinder are down. How do you identify the problem and rectify it?

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

Tests metrics diagnosis for a core marketplace metric: isolating whether the drop in matches is a supply, demand, or algorithm problem.

Define matches precisely: mutual right swipes, distinct from messages sent or dates arranged, to scope the metric correctly.
Break potential causes into supply side, fewer active users or a gender ratio imbalance, and product side, an algorithm or UI change.
Check for a recent change correlated with the drop, such as a matching algorithm update or a UI change to the swipe flow.
Segment the data by geography, age, and gender to see if the decline is broad or concentrated in a specific segment.
Once isolated, propose a targeted fix matched to the diagnosis rather than a generic re engagement campaign.
Define the recovery metric: matches per active user per week, tracked by the affected segment.

Insists on diagnosing before fixing, first checking whether swipe volume, match rate, or both declined, since each points to a different cause.
Checks for a recent product change correlated with the drop, such as an algorithm tweak or UI redesign to the swipe card stack.
Segments by geography and gender ratio, since a local supply demand imbalance often explains match declines concentrated in specific cities.
Proposes a targeted fix based on the finding, for example reverting or tuning a recent algorithm change if that is the correlated cause.

Assuming the cause is a supply problem without first checking for a recent product or algorithm change.
Proposing a broad re engagement campaign before isolating whether the issue is swipe volume or match conversion.

How would you distinguish a seasonal dip from a genuine structural decline?
What would you do if the drop was concentrated in one specific city or age group?