Estimate China's Red Car Population and How You Would Measure It

Quick Overview

A product manager estimation prompt about working out how many red cars there are in China. It tests defining what counts as a car and as red, sizing the passenger fleet across urban and rural households, choosing a defensible color share, and naming real data sources that could confirm the figure.

Estimate China's Red Car Population and How You Would Measure It

Company: Google

Role: Product Manager

Category: Product / Decision Making

Difficulty: medium

Interview Round: Other

How would you figure out how many red cars there are in China? Explain the approach you would take. Reason your way to a number with each assumption labeled, and describe what data you could use to confirm or refine it. ```hint Break it into pieces you can reason about Split the question into quantities you can estimate separately, and keep in mind that car ownership in China differs greatly between big cities and the countryside. ``` ### Clarifying Questions - Which vehicles count as cars: private passenger cars only, or also taxis, ride-hailing and rental fleets, and vans or trucks? - Which shades count as red: does maroon, burgundy or orange-red count, and how should two-tone cars be treated? - Does China mean mainland China only? - Should the count include every registered car, or only cars still in use? ### What a Strong Answer Covers - Definitions of a car, of red and of the geography before any numbers appear - A structured estimate of the car fleet that reflects the urban–rural gap in ownership and includes cars not owned by households - A reasoned, labeled assumption for the share of cars that are red, with a range - Clean arithmetic, a final range and a sanity check against a known benchmark - Real data sources that could confirm the number, with the biases of each ### Follow-up Questions - If you counted cars with roadside cameras, why might the share of red cars you see differ from the share that exists, and how would you correct for it? - Registration records seem to answer the question directly. What could make them wrong? - Which assumption drives your answer most, and how would you narrow it?

Overview: A product manager estimation prompt about working out how many red cars there are in China. It tests defining what counts as a car and as red, sizing the passenger fleet across urban and rural households, choosing a defensible color share, and naming real data sources that could confirm the figure.

Google logo
Google
Aug 31, 2024
mediumProduct ManagerOtherProduct / Decision Making
0
0

How would you figure out how many red cars there are in China?

Explain the approach you would take. Reason your way to a number with each assumption labeled, and describe what data you could use to confirm or refine it.

Clarifying Questions Guidance

  • Which vehicles count as cars: private passenger cars only, or also taxis, ride-hailing and rental fleets, and vans or trucks?
  • Which shades count as red: does maroon, burgundy or orange-red count, and how should two-tone cars be treated?
  • Does China mean mainland China only?
  • Should the count include every registered car, or only cars still in use?

What a Strong Answer Covers Guidance

  • Definitions of a car, of red and of the geography before any numbers appear
  • A structured estimate of the car fleet that reflects the urban–rural gap in ownership and includes cars not owned by households
  • A reasoned, labeled assumption for the share of cars that are red, with a range
  • Clean arithmetic, a final range and a sanity check against a known benchmark
  • Real data sources that could confirm the number, with the biases of each

Follow-up Questions Guidance

  • If you counted cars with roadside cameras, why might the share of red cars you see differ from the share that exists, and how would you correct for it?
  • Registration records seem to answer the question directly. What could make them wrong?
  • Which assumption drives your answer most, and how would you narrow it?