Metrics question
Uber could use two strategies to notify drivers when a ride is requested: broadcast to all drivers in the vicinity or sequentially notify drivers. As an Uber PM, explain which one you would choose and why.
- Uber
- Metrics
- Hard
Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.
Start a mock interview on this question · Mock interview from a job description
What this question tests
Tests structured trade-off analysis on a two-sided marketplace: can you weigh driver experience against rider wait time using concrete mechanisms, not just a gut pick.
How to approach it
- Restate the options: broadcast (all nearby drivers alerted, first to accept wins) versus sequential (offer to the best-matched driver, then the next on decline).
- List the metrics each moves: driver acceptance rate, rider ETA, earnings fairness, and idle miles per trip.
- Work through broadcast's failure mode: drivers race toward one rider, wasting time for the losers and rewarding fast reflexes over best match.
- Work through sequential's failure mode: each decline adds latency, raising rider wait time during surge.
- State the assumption: dense areas absorb sequential's latency easily, sparse areas lose too much time per hop.
- Recommend a hybrid: sequential to the top match first, falling back to a small-radius broadcast if no acceptance within seconds.
What a strong answer includes
- Names broadcast's cost concretely: if 5 drivers race to one rider, 4 trips worth of driving time are wasted, lowering drivers-per-hour.
- Ties the choice to density: sequential for dense zones, broadcast-leaning hybrid for sparse suburbs.
- Proposes an A/B test on match rate, wait time, and cancellation rate before rollout.
- Flags fairness: pure broadcast rewards fast reflexes, sequential rotates who gets first offer.
Common mistakes
- Picking one option outright without naming the metric trade-off between rider wait time and driver efficiency.
- Ignoring that both approaches can be blended rather than treated as mutually exclusive.
- Not mentioning surge or low-supply conditions, where the right answer flips.
Likely follow-up questions
- How would you measure whether the hybrid approach is working within the first two weeks?
- What happens to this trade-off during a surge event like a concert letting out?
- How would driver trust and platform reputation be affected by each option?
More metrics questions
- Late deliveries lead to customer churn. What data we should look at to prove this hypothesis for a food delivery app?PayPal · Metrics · Medium
- Which metrics would you track for Uber Pool during the first 6 months after product launch?Uber · Metrics · Medium
- Uber has seen a dip in the ride confirmation rate. As a PM at Uber, you are responsible to investigate and fix the issue. How will you go about it?Uber · Metrics · Hard
- How would you describe and assess how the Uber Marketplace (the matching platform) is doing, for UberX?Uber · Metrics · Hard
- How would reduce cancellations for Uber?Shopify · Metrics · Medium
- What is the most important metric for Uber Eats? Why?Uber · Metrics · Medium
More questions from Uber
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.
- Chapter 9: Prove it paid off: outcomes, economics, and pricing
- Chapter 2: Data fluency: SQL, logs, and reading the truth yourself
- Chapter 14: Get the job: the AI PM interview loop