Metrics question

You are a PM of Lime Scooters. Users are signing up but are not using the scooters. What do 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

Whether you can structure an activation funnel to diagnose a drop-off and target the right fix.

Define the funnel: download, sign up, first ride, repeat ride, and find exactly where users stall.
Segment non-activators by city, time since signup, and whether a scooter was actually nearby when they opened the app.
Generate hypotheses: no scooters nearby, price or deposit friction, a confusing unlock flow, or safety concerns.
Test each hypothesis against data, comparing distance to the nearest scooter at signup for riders versus non-riders.
Prioritize the highest-volume, most fixable driver, most likely scooter availability near the signup location.
Test a low-cost fix, such as a nearby-scooter push notification, and measure the lift in first-ride rate.

Frames this as an activation problem with a concrete metric, like percent of signups completing a ride within seven days.
Uses a plausible number, assuming 40 percent of non-riders had no scooter within a five minute walk at signup.
Separates a supply-side cause, no scooters, from a product-side cause, confusing unlock, before proposing a fix.
Proposes testing a cheap fix, a nearby-scooter notification, before a costly one, like rebalancing the fleet.

Assuming the cause is pricing without first checking availability data.
Proposing broad marketing pushes instead of diagnosing the actual activation funnel.

How would you separate a supply shortage from a confusing unlock flow using data alone?
What would you do if the drop-off were concentrated in a single city?