PM estimation: size the yearly revenue of Amazon's cloud computing business

Quick Overview

A product manager estimation question that asks you to size the annual revenue of Amazon Web Services. It tests writing a clear revenue equation, segmenting customers by spend, keeping units and arithmetic consistent, and cross-checking the result with an independent method.

PM estimation: size the yearly revenue of Amazon's cloud computing business

Company: Microsoft

Role: Product Manager

Category: Product / Decision Making

Difficulty: medium

Interview Round: Other

Estimate how much revenue Amazon Web Services (AWS), Amazon's cloud computing business, brings in over one year. Explain the approach and the assumptions behind your number. ```hint Customers are far from equal Cloud spending is heavily concentrated: a small number of very large accounts spend orders of magnitude more than the long tail. Build segments that reflect that, keep your units consistent, and then check the result with a second, independent method. ``` ### Clarifying Questions - Which year should the estimate describe: the latest full year, the current year, or a forecast? - Should revenue cover only what outside customers pay, or also the cloud capacity Amazon's own businesses use? - Is a worldwide figure across all AWS services expected, or a specific region or product line? - Do you want a bottom-up build from customers, or is a top-down estimate from the size of the cloud market acceptable? ### What a Strong Answer Covers - A defined revenue scope and a written equation before any numbers. - Customer segments that reflect spend concentration, each assumption labeled and rounded, and segments that add up to a plausible customer count without double counting. - Clean arithmetic with consistent monthly versus annual units. - An independent cross-check, such as market size times share, and a reconciliation of the two results. - The assumptions the answer is most sensitive to and how to tighten them. ### Follow-up Questions - Which single assumption moves your estimate the most, and how would you narrow it? - How would you split the total across compute, storage, databases and other services, and why might that view matter? - How would you turn your estimate into a forecast for the next three years?

Overview: A product manager estimation question that asks you to size the annual revenue of Amazon Web Services. It tests writing a clear revenue equation, segmenting customers by spend, keeping units and arithmetic consistent, and cross-checking the result with an independent method.

Microsoft logo
Microsoft
Oct 21, 2025
mediumProduct ManagerOtherProduct / Decision Making
0
0

Estimate how much revenue Amazon Web Services (AWS), Amazon's cloud computing business, brings in over one year. Explain the approach and the assumptions behind your number.

Clarifying Questions Guidance

  • Which year should the estimate describe: the latest full year, the current year, or a forecast?
  • Should revenue cover only what outside customers pay, or also the cloud capacity Amazon's own businesses use?
  • Is a worldwide figure across all AWS services expected, or a specific region or product line?
  • Do you want a bottom-up build from customers, or is a top-down estimate from the size of the cloud market acceptable?

What a Strong Answer Covers Guidance

  • A defined revenue scope and a written equation before any numbers.
  • Customer segments that reflect spend concentration, each assumption labeled and rounded, and segments that add up to a plausible customer count without double counting.
  • Clean arithmetic with consistent monthly versus annual units.
  • An independent cross-check, such as market size times share, and a reconciliation of the two results.
  • The assumptions the answer is most sensitive to and how to tighten them.

Follow-up Questions Guidance

  • Which single assumption moves your estimate the most, and how would you narrow it?
  • How would you split the total across compute, storage, databases and other services, and why might that view matter?
  • How would you turn your estimate into a forecast for the next three years?