Number of words: 239

A good interview puzzle should be easy enough that you’re willing to reject anyone who doesn’t solve it That’s probably the best way of describing the optimal difficulty level for a question with a definite right answer. You want a puzzle such that many people will fail to solve it, but practically all “acceptable” candidates will get the answer. Not solving the puzzle should raise a red flag.

There is a trade- off. People’s reactions to puzzles are personal. You should expect that there will be capable candidates who fail to solve the puzzle and who will be rejected unnecessarily. Remember, the guiding principle should be that it’s better to lose some good people in order to avoid hiring unsuitable people.

One weakness of interviewing is that smart people generally come off well. Selective companies tend to hire smart people and then are baffled when some turn out to be disastrous employees. Just like everyone else, smart people can lack motivation.

Puzzles, design questions, and impossible questions are miniprojects. It is not enough to have good, intelligent insights. You also have to weave them together and bring things to a conclusion. Now, sure it’s a lot easier to solve a puzzle. than to ship a new product at a big company. For that very reason, leaving puzzles and questions unresolved should raise a red flag.

Excerpted from page numbers137 of ‘How Would You Move Mount Fuji?’ by William Poundstone.