Brain Teasers and Job Interviews

Once at a party a question arose about why companies ask brain teasers in job interviews. The question circled around to me, since at the time I ran a company that helped companies evaluate job candidates.

It’s funny, because I asked that problem years ago in my Hiring and Selection seminar in business school, and even the professor admitted he didn’t really get it. The professor did say he got one of his first (non-academic) jobs because he solved a problem about how many gas stations are in the U.S.

So why do companies ask brain teasers?

I’ve got a couple of unfortunate answers, and one somewhat (although only somewhat) admirable answer. Probably the true answer varies by case and maybe it’s a mix of these three.


  1. Interviewers are grasping at straws. Interviewers may not view interviewing as a key aspect of their job, and so they don’t think much about how to do it well. They mostly look around for some questions they can use to pass the time and give them some interaction with the candidate, even if the questions are not especially predictive of job success. There are lots of brain teasers lying around, so these are easy questions to use.
  2. Interviewers use the types of questions they have seen in the past. Again, many interviewers aren’t well-prepared, so they default to conducting the type of interviews they have seen in the past, instead of designing an interview or assessment from data or at least first principles. Some interviewers may even be particularly good at puzzles themselves, and believe that success on the job is correlated with puzzle-solving ability. Research indicates this is unlikely to be true, but, again, many interviewers are insufficiently well-prepared to know this.
  3. Interviewers use puzzles to measure intelligence. This is the most admirable of the answers I’ve outlined, but it still doesn’t pass muster. There is substantial empirical evidence that mental ability correlates well with job success, across a wide range of jobs. Certainly IQ is not a perfect predictor, or even a great predictor, but it’s one of the best predictors available. And it seems likely that puzzle-solving ability correlates with IQ, although my guess is that the puzzle-IQ correlation is weak. IQ is notoriously complicated to measure (even more so given legal constraints in the hiring process). A brain-teaser, especially one which has never been calibrated and cross-validated against IQ, is unlikely to be meaningfully predictive of job success. It’s a pretty wild double-bank shot. Interviewers would be better off using SAT scores, or better, relying on a vendor who specializes in cognitive tests for hiring.

Those are my hypotheses about why brain teasers show up on job interviews, even though they shouldn’t. Hopefully, over time people will get smarter about hiring and hone in on assessments that are more highly predictive of job success.