Metrics question

If a large number of drivers are dropping out of a particular city, why would it be?

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 for driver side supply churn using segmented hypothesis testing.

Confirm the scope, verify the drop is real and specific to this city, not a broader trend or a data or tracking issue.
Hypothesize earnings related causes, lower fares, increased competition from another company, or higher local gas or vehicle costs eating into take home pay.
Hypothesize experience related causes, poor rider behavior, unsafe pickup zones, or app issues specific to that city's conditions like poor GPS in dense areas.
Hypothesize external causes, a local regulatory change or a seasonal event ending that temporarily drew in extra drivers.
Investigate using driver exit surveys and comparing earnings per hour in this city against similar cities to narrow down the real cause.

Structures hypotheses into clear categories, earnings, experience, and external or regulatory, rather than guessing a single cause.
Proposes a specific comparative check, earnings per hour in this city versus similar sized cities, to see if pay is genuinely the outlier factor.
Proposes driver exit surveys as the direct way to validate the real reason rather than only inferring from aggregate data.

Naming only one hypothesis instead of considering earnings, experience, and external factors separately.
Not comparing against similar cities to confirm this is a real local anomaly, not a broader trend.
No plan to directly ask departing drivers why they left.

What would you do if the exit survey pointed to competitor pay being higher?
How would you design a retention offer without overspending on drivers who would have stayed anyway?