Estimation question

Estimate the market share of Google Translator as an API.

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

Tests structured market sizing under ambiguity, decomposing a share question into a total market estimate and a company-specific estimate.

State the assumption that market means paid machine-translation API spend globally, since consumer usage isn't a share of revenue.
Estimate total market size top-down: number of businesses needing localization, assume 2M globally with digital products, times average annual spend, assume $5,000 a year, roughly $10B.
Estimate Google's piece bottom-up by naming known competitors, AWS, Azure, DeepL and Google, and splitting share, assuming Google captures 20 to 25% given scale and a free tier.
Sanity-check the estimate against Google Cloud's overall revenue so the segment isn't implausibly large relative to total Cloud revenue.
State the final estimate as a range, not a point number, given the many assumptions.

Labels every number as an assumption and shows the math rather than guessing a percentage outright.
Cross-checks the estimate two ways, top-down market size and bottom-up competitor share, and reconciles them.
States a defensible range, such as 15 to 25%, instead of false precision.
Notes DeepL's outsized reputation despite smaller scale as a reason Google's share might be lower than usage alone suggests.

Giving a single number with no shown work.
Not defining which market is being measured, consumer versus API, translation versus all NLP.
Ignoring competitors entirely.

Chapters of the AI PM course that teach what this question tests.