Interviews don't work

You're fooling yourself if you think they do

Interviews are like classroom teaching. We’ve been doing both for millennia without really improving them much. We take them for granted. Want to teach something? Get some students together, stand up in front of them, and explain it to them. Want to hire someone? Bring them in for an interview.

Neither of these is effective. But they are cheap—or seem so until one looks at the bigger picture—and they are easy to do. All you need is a classroom or office. These days, for the interview all you need is a phone or the Internet.

But they simply do not work, even though we are convinced that they do. This is not to say that people can’t learn in a classroom environment or that interviews can’t produce good hires, but these outcomes happen despite our efforts, not because of them.

I’ll get into why classroom teaching sucks in another essay. Today, let’s talk about interviewing.

You are a shit judge of character and ability

The biggest problem with the interview is the interviewer. Humans are generally very bad at judging both character and ability. What they are relatively OK at is deciding whether they like someone.

So if your goal is to hire only people your interviewers like, then maybe interviews can work for you.

The reason you can’t judge character or ability well, despite your Dunning-Kruger-like insistence that oh, yes, you can, is because, frankly, you are biased as hell. Not only are you biased, you have no clue how biased or in what ways you are biased.

Note to readers: before you complain that this is only my “opinion”, permit me to point out: the fuck it is. It is scientifically-proven fact. It has been proven so thoroughly that it’s not even the least bit controversial.

Cognitive biases have been well understood—by everyone except interviewers and their apologists, it seems—for decades. And yet we keep pretending that they either don’t exist, or we can somehow “work around them”.

No. We. Can’t.

Here’s what Wikipedia has to say. (Granted, Wikipedia is biased as hell, but they’re usually pretty good about uncontroversial science such as this.)

A cognitive bias is a systematic pattern of deviation from norm or rationality in judgment. Individuals create their own "subjective reality" from their perception of the input. An individual's construction of reality, not the objective input, may dictate their behavior in the world. Thus, cognitive biases may sometimes lead to perceptual distortion, inaccurate judgment, illogical interpretation, or what is broadly called irrationality.

Ha, ha. Did you get that bit about irrationality? But not you, right? You can see clearly to interview fairly and well.

What arrogance and self-delusion!

And it’s not as if we have one or two or half dozen cognitive biases. Just take a look at the Cognitive Bias Codex:

Stolen fair and square from Designhacks.co via Wikipedia

Holy crap! And can you even name a cognitive bias? How then are you going to avoid bias in hiring?

Answer: you won’t.

Here’s another great quotation, this one from an article on feedback in the Harvard Business Review, but it works equally well for interviewing:

Over the past 40 years psychometricians have shown in study after study that people don’t have the objectivity to hold in their heads a stable definition of an abstract quality, such as business acumen or assertiveness, and then accurately evaluate someone else on it. Our evaluations are deeply colored by our own understanding of what we’re rating others on, our own sense of what good looks like for a particular competency, our harshness or leniency as raters, and our own inherent and unconscious biases. This phenomenon is called the idiosyncratic rater effect, and it’s large (more than half of your rating of someone else reflects your characteristics, not hers) and resilient (no training can lessen it).

Of course your interviewers would never fall prey to the idiosyncratic rater effect, would they?

But wait, there’s more!

You might be thinking that you can overcome these biases with sufficient training and practices. But no, you can’t. It’s just how humans work. Go back and reread the feedback quote above: no training can lessen it.

But that brings up another important point. Have you trained your interviewers at all?

Here’s another similarity between interviewing and teaching: virtually everyone thinks that they can do it, and do it well. It’s astonishing really. And incredibly insulting to the interviewers and teachers who have spent whole careers honing their craft.

Almost anyone can get up in front of a room of people and blab for an hour and the people in that room will teach themselves from the material presented. They will do this despite the inadequacies of the blabberer. And yet the blabberer will claim to be a teacher.

Ditto for interviewers. Anyone can get in a room and harass another person with questions and then make a yea or nay decision. And once the position is filled, we can pretend that the persons asking the questions were “interviewers”, but really they were just inquisitors.

Even if good interviewing were possible, your interviewers are incompetent. Worse, they don’t realize it.

Dunning & Kruger didn’t know the half of it.

It gets even worse

Even if you had the best interviewers. Even if they had no biases and could evaluate fairly and critically. Even if you trained them for years. Even if you invented a perfect AI to do the interviews for you.

Your interviews would still be unfair and counterproductive.

Why? Because unless you are hiring people to be interviewed, you are testing the wrong thing.

No doubt you have a job description for the position, right? And does the job description include sitting in an office with one or two strangers who are badgering you with stupid and arbitrary questions that even they don’t give a shit about? No?

Then you are testing the wrong thing.

What you are testing is this: is this candidate good at passing interviews?

You know who is really good at passing interviews? Sociopaths. No, really. They tend to be charming. They tend to know how to manipulate people. And they are so self-possessed that they don’t get nervous or flustered.

I actually knew a sociopath once who decided that for fun he would interview for a bunch of law schools. He had no intention of actually going to law school. He just wanted to see if he could game the system. Kind of like trying to fool a polygraph, I guess.

And he did. Fool them, that is. His interviews all went extremely well and the interviewers practically begged him to choose their school. He thought it was great fun. I remember him laughing while he told me how pathetic the interviewers were.

But non-sociopaths tend to get nervous. Most are not particularly eloquent under pressure. Most are just trying hard not to mess up: they need this job. How many interviewees actually take control of the interview, as my sociopathic friend did, and lead it to a successful conclusion for themselves? Virtually none.

But even if they did, that would still not ensure that the right candidate got the job.

The entire setting of an interview is utterly artificial. There is a time limit. The interviewee rarely knows the interviewer(s). Both parties are starting from scratch. The interviewee doesn’t know the questions ahead of time and has no time to prepare for them. Often, the whole point of the interview is the search for the perfect “gotcha” question.

Who works like that? No one.

And if you blow the interview—whether as interviewee or interviewer—the process stops, so you rarely get another chance. One bad day, one wrong or misunderstood comment, and you’ve just wasted your time. And the company’s time. And money. Lots of it.

And don’t even get me started on interviews that claim to test “how you think” or “how you respond to pressure”. They can only test how you think and how you respond to pressure in that interview. On that day. At that time. With those people. In that room. Which almost always has zero relevance to the world outside the interview.

More often than not, it is not the interviewee who failed the interview, but the interview that failed the interviewee. And the company.

Too often, great hires are overlooked because the interview didn’t go well or the interviewers couldn’t see past their own biases and prejudices. That’s a failure of the interview, not the interviewee.

And too often the wrong people are hired for a job, or people are hired for the wrong job, again because the interview failed.

And if we’re really honest, that’s such a common outcome that we’d almost certainly be better just using a random number generator to pick new hires once the candidate pool has been thinned to those who have the essential skills. I think the science actually backs me up on that.

There’s much more, but enough on interviewing for now. So what is the best way to hire then?

The best way to hire

Like it or not, the only truly effective way to hire for a role is to let the candidate do the role long enough to learn whether that candidate can be successful in it.

In doing so, we have eliminated the unfairness of the artificial world of the interview, and replaced it with reality. This is the job. You can do it, or you can’t.

There is still plenty of room for fucking up, of course. Just because someone is on the job doesn’t mean that they’re being treated well or fairly, or that the evaluation period won’t be utterly mishandled. It may well be. But that’s a different issue.

At least there’s a fighting chance that it will be fair. Interviews never are.

So take a few tasks required for the role, then offer the candidate a short-term contract with pay to perform those tasks. If they do well, then hire them full-time. If you’re still not sure, do another short-term gig. If they fail at any point, cut the contract short, pay them, and move on. No harm, no foul. It just didn’t work out.

I have had people say to me, But gosh! That’s too much work. It would be too expensive.

Seriously? You are paying them to do work. How can that be too expensive? Isn’t that what they’ll be doing—or trying to do—after you hire them? You know what is very expensive? Hiring the wrong person.

But the best part of this is that it gives both you and the candidate time to get to know each other and to decide if this really is the right role for this person. Perhaps you realize that it is not, but you have another role that would suit them better. Perhaps they discover that the role is nothing like what they thought it would be like. That might be bad, but it might be good, too. Or perhaps it just doesn’t work.

It is true that many companies claim to have a “probationary period”, but few actually put it to proper use. And why go through the whole “permanent” hiring process for a probationary period? It gives the wrong impression: to be let go at the end of it feels like failure, rather than just a mismatch. After all, you were supposed to be permanent.

In my experience, the best hires I’ve ever seen have all had one thing in common: they started as interns. That means that they were already doing the job for weeks or months before we “hired” them. They were a known quantity.

We can dance around the truth and wave our hands in the air all we like. The truth is that the only fair and effective way to determine whether someone is right for a role is for them actually to do it.

‘Nuff said.