Metrics question

As a PM of Uber, how would you solve the problem of sudden increase in demand of cabs when an event is over?

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 operational problem-solving for a predictable but sharp supply-demand mismatch, balancing pricing tools with proactive supply planning.

Clarify this is a forecastable spike, since event end times are known in advance, unlike random demand surges.
Use historical data from similar past events to predict the expected demand spike's size, location and timing.
Proactively pre-position drivers near the venue before the event ends, using incentives to guarantee driver presence at the predicted spike time.
Apply dynamic pricing as a secondary lever, balancing rider fairness concerns against the real need to pull in more driver supply quickly.
Define success as average wait time immediately after the event and rider complaint rate about pricing or availability.

Recognizes this spike is predictable and shifts focus to proactive pre-positioning rather than only reactive dynamic pricing.
Uses historical event data to forecast supply needs, a concrete and specific solution.
Explicitly weighs rider fairness concerns around surge pricing rather than treating it as a purely mechanical lever.

Relying on dynamic pricing alone without any proactive driver pre-positioning.
Treating this like an unpredictable random spike rather than using historical event data to forecast it.

How would you incentivize drivers to pre-position without guaranteed pickups?
How would you handle rider backlash against high surge pricing after a big event?