This is a poor way to interpret the reality. Interviews have relatively weak validity, but the relationship is still positive. Of course that assumes a structured interview from competent interviewers, which is more the exception than the norm.

What interviews may lack in utility for predicting good performance they can potentially make-up for in determining cultural/supervisor-candidate fit. But this also requires the interviewer to 1) accurately represent the culture and 2) to not implicitly discriminate with things that might correlate with fit, which many things (age, religion, gender) might do. No simple ask there.

I think the primary value of interviews it that it makes the process seem more fair. Everyone had a chance to put their thumbprint on the experience, including the interviewee. In the absence of other selection measures (e.g., work samples) I understand its use, but if you have better tools then you should use those. For me, an interview's primary outcome question should really only be "Is this person completely unqualified or terribly unprofessional?" If the answer is "no", rely on the other indicators to make a decision.

I wouldn't want someone to get the impression I completely write off interviews as a way to assess the potential in someone. It's been around forever, and I don't see it going away anytime soon.

Also I don't think my comment implies that outgoing / social people are bad at what they do.

I wouldn't think too hard about the correlation I'm trying to make. Really my point is that the skillset you need in order to "pass" an interview is far different skillset in many cases than what the actual job requires.

Same with politics. The irony is the skillset necessary to "win" an election is far from what is necessary to do a good job once you're elected.

EDIT: Also, I would clarify that the context of the original thread is about hiring a software engineer at Google. The interview(s) necessary for hiring for that position I imagine could not possibly be conducted in the proposed "2 minutes at the beginning of the interview" I was originally replying to.

I have no doubt that there are plenty of positions that an interview can be conducted in 2 minutes, but software engineer at Google I doubt is one of them.

I'm replying to my own comment here just because it's a different point.

I posit, interviews are unlikely candidates to truly separate the wheat from the chaff in these 2 distinct but probably not uncommon scenarios.

  1)  ambitious person with limited skills
  2)  highly skilled person with limited ambition

With the first you wonder if their ambition will be enough to get them over many technical hurdles in the course of the job. When bringing someone like this onto your team you almost need to take personal stake in their success for fear that by hiring them you're potentially taking them away from another opportunity where their immediate qualifications may be a better fit.

With the second person I imagine it's a different kind of challenge to try to uncover this kind of personality trait during an interview process.

A good interviewer isn't looking for somebody that's suave and comfortable in the interview. They understand awkwardness, nervousness and that it's easy to make stupid technical and social mistakes under pressure -- and they may even relate to it.

But they still probably make a 95%-accurate assessment in those first couple minutes.

Or at least, think they do. They're human, and subject to human biases which no amount of awareness and understanding will completely remove.

I don't really agree with that. Some people are naturally more outgoing and comfortable in interview situations than others. It doesn't mean they're bad people.