Metrics question

You have to launch an Uber Rider pass in your city. What should be the price of the pass?

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

Pricing strategy grounded in unit economics, checking whether you can size the discount against ride frequency and rider willingness to pay.

Clarify the pass structure: is it a monthly unlimited-rides pass, a ride-credit bundle, or a discount-percentage pass, since pricing differs by type.
Segment target users: frequent commuters taking 20 plus rides a month get the most value, so anchor pricing to that segment.
Estimate average per-ride cost in the city, say 8 dollars, and current monthly spend for a frequent rider, roughly 160 dollars for 20 rides.
Price the pass to offer a visible discount, for example 120 dollars for 20 rides, a 25 percent discount, enough to feel valuable without eroding margin too much.
Model the trade-off: lower per-ride revenue against higher ride frequency and reduced churn to competitors like Lyft.
Test with a pilot in one city, tracking rides per pass holder versus a control group, before a full launch.

Anchors the price to actual per-ride spend data, not a guess, making the number defensible.
States the trade-off explicitly: a discount that is too small won't drive adoption, one too large erodes margin, and picks a number, 20 to 30 percent off, that balances both.
Proposes a pilot with a clear control group instead of committing to a citywide launch immediately.

Picking a price with no reference to actual per-ride economics, making the number arbitrary.
Ignoring the risk that a pass mostly attracts riders who would have ridden anyway, inflating apparent success.

How would you detect if the pass is mostly used by riders who were already frequent, not net-new demand?
How would pricing differ between a high-demand city and a low-demand one?