Estimation question
How many drivers does Uber have in a medium sized US city?
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
Estimation: can you build a supply-side Fermi estimate anchored on population, ride demand, and driver utilization for a specific city size.
How to approach it
- Define 'medium sized US city': assume a metro population of roughly 1 million (state as assumption).
- Estimate daily Uber rides in that city based on rider penetration and ride frequency, e.g. 5% of population takes 1 ride every few days.
- Estimate rides per driver per day based on typical shift length and average ride duration plus dead time between rides.
- Divide total daily rides by rides-per-driver to get an estimate of drivers needed to serve demand, then add a buffer for uneven demand (peak hours needing more supply than average).
- Sanity check the final driver count against a plausible range for a city that size.
- State what real data (Uber's public city-level statistics, if any) would refine the estimate.
What a strong answer includes
- Builds the estimate from a demand-side, then converts to supply using a rides-per-driver assumption, rather than guessing total drivers directly.
- Explicitly adds a peak-demand buffer, recognizing driver supply must cover peak hours, not just the daily average.
- States a clear population assumption for 'medium sized' since the term is undefined, and reasons from there.
- Lands on a defensible range with assumptions labeled rather than a single falsely-precise figure.
Common mistakes
- Not defining what 'medium sized city' means numerically before starting the estimate.
- Ignoring peak-hour demand and only estimating for an average, understating needed driver supply.
Likely follow-up questions
- How would your estimate change during a weekend night versus a weekday morning?
- How would you refine this with actual Uber trip data if you had access?
More estimation questions
- You have a grocery delivery service that delivers food within 24 hours. How many trucks would you need to operate it?Google · Estimation · Medium
- What is the market for driverless cars in 2020?Google · Estimation · Easy
- How would you determine how many trips are taken on the London Underground each year?Lyft · Estimation · Easy
- How would you price a Tier 1 and Tier 2 Uber subscription?Lyft · Estimation · Hard
- How many ride-share trips occur in the US daily?Google · Estimation · Easy
- How much more or less should Lyft pay drivers per trip (by changing Lyft’s take) to maximize net revenue for the next 12 months on this route in Toledo, Ohio? The goal is to find the optimal balance between Lyft's take and driver pay to improve the match rate and reduce failed rides while maximizing net revenue.Lyft · Estimation · Hard
More questions from these companies
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.
- Chapter 2: Data fluency: SQL, logs, and reading the truth yourself
- Chapter 9: Prove it paid off: outcomes, economics, and pricing
- Chapter 14: Get the job: the AI PM interview loop