Are You Smart Enough to Work at Google?
Summary
Poundstone examines the hiring practices of technology companies, particularly the puzzle and brainteaser interview questions Google and others became known for, and assesses whether they work.
The evidence he presents is largely negative. Google's own internal research, which the company published, found that brainteaser interview performance had no correlation with subsequent job performance, and the company substantially abandoned the practice. Poundstone covers the wider literature on selection methods, which consistently finds that unstructured interviews predict performance poorly, that structured interviews with consistent questions and scoring do better, that work sample tests and general cognitive ability tests do better still, and that interviewer confidence in their own judgement is unrelated to its accuracy.
The book also contains a large collection of the actual questions and worked solutions, covering estimation problems, logic puzzles, algorithmic questions and the deliberately strange ones. Poundstone treats these as intellectually interesting independent of their validity as selection instruments.
His broader argument is that hiring practices spread by imitation of prestigious firms rather than by evidence, which is why a method Google abandoned continued to be used elsewhere for years afterward.
The book straddles two purposes and does neither completely. As a puzzle collection it is enjoyable, as an analysis of hiring it is a survey rather than original work, and the two halves interrupt each other. It is from 2012 and technology hiring has changed substantially since.
Key takeaways 7
- Google's own research found that brainteaser interview performance had no correlation with subsequent job performance, and the company abandoned the practice.
- Unstructured interviews predict job performance poorly, while structured interviews with consistent questions and scoring predict substantially better.
- Work sample tests and general cognitive ability measures predict performance better than any interview format.
- Interviewer confidence in their own judgement is unrelated to the accuracy of that judgement, which is why experienced interviewers do not outperform.
- Hiring practices spread by imitation of prestigious firms rather than by evidence, which is why abandoned methods persist elsewhere.
- Estimation questions test a real and useful skill, which is decomposing an unknown quantity into components that can be bounded.
- Selection methods should be validated against actual performance data, which almost no organisation does even when the data exists.
Who should read it
Operators designing hiring processes who want the evidence on interview validity, alongside a collection of the puzzles themselves.
About William Poundstone
American science and business writer covering game theory, probability, pricing psychology and decision-making.
Fortune's Formula traces the Kelly criterion from Bell Labs information theory to Ed Thorp's blackjack and hedge-fund bet sizing, and it remains the standard entry point to position sizing under an edge. Priceless collects the experimental work on how buyers anchor and compare prices, which transfers straight into pricing decisions.
Who recommended it 1
Similar books 6
Chosen by shared topics, the entries that hold them, and the people who recommended both.
A fable of a failing executive team and a five-level model of what goes wrong.
Recommended by: Alfred Lin, Chris Powers, Sarah Pritula and 1 other
Recommended by: Mohnish Pabrai
Recommended by: Mohnish Pabrai
Building an ownership culture in which employees think and act like owners.
Recommended by: Tony Hsieh
An owner who converted his own firm, on employee ownership in practice.
Where it sits in the reference 3
Topics