I have been running a small business for more than twenty years. I have met thousands of people. I have hired employees, contractors and suppliers. I have sold products and services to countless customers. You would think by now I would have a good sense of when someone is telling me the truth. You would be very wrong. Even to this day, I'm still terrible at telling whether that person sitting across my desk is being straightforward with me or not. You just never know. I can talk to someone for an hour and, in the end, my decision to do business with him is based on a gut feeling.

That's why job interviews are a waste of time. I've always thought this. Now there's research to back me up.

Jason Dana, an assistant professor of management and marketing at the Yale School of Management, has validated my experiences. In this New York Times piece, he goes so far as to call job interviews, particularly the free-form ones that we all mostly have, as "useless." And he cites actual results from actual data to support his position. Examples:

  • When the University of Texas Medical School at Houston in 1979 was ordered to increase its incoming class size, it admitted more than 50 students that year who were previously rejected at the interview stage. Turns out those students did just as well as the rest.
  • Dana's own studies found that when his students were asked to predict the future grade point average of people they interviewed, as well as people they didn't interview using instead the candidates' historical performance and course schedules, his students more accurately predicted the G.P.A. for those they didn't interview. In other words, the interviews they did conduct were counterproductive.
  • In those same studies, half of the interviewees were told in advance to give honest answers to the questions while the other half were told to answer randomly based on a made-up algorithm. Not only did the interviewers not figure this out, but the students who answered questions randomly also received overall higher scores.

Amazingly still, Dana shared these results with another group of students and, even knowing this, "a majority felt they would rather base their predictions on an interview they knew to be random than to have to base their predictions on background information alone." Humans will just be humans.

The takeaway: Interviews are pretty useless.

Dana thinks that a more structured and formal interview using the same set of questions for each candidate would be better than just the free-form, unstructured conversation that most of us do. However, he hasn't yet provided the data to back up his theory. Perhaps soon.

In the meantime, I say go ahead, sit down, have fun and interview that next candidate. It's always good to get to know someone, and maybe there's an obvious red flag that you'll uncover. But don't put too much faith in your gut, because if it's like mine, it's probably wrong half of the time. Instead, look at past performance, talk to references and former employers and do a background search using tools like Checkr and maybe even a few pre-tests with services like Hire Success.

Then (and most importantly), consider bringing that person on for a 90-day trial period before making them a bona-fide full-time employee. Interviews are nice, but what the heck do you and I know? Trusting our gut is fine, but truly good workers will demonstrate their abilities only when we can see them on the job.