The quiet tragedy of behavioral interviews

You know these questions: "Tell me about a time you had to handle a conflict with a coworker." "Describe a situation where you had to show leadership." "What's the biggest obstacle you've faced in your career?"

If you've been involved in any recruitment process in the past 5 years, chances are high that you've had to answer those questions yourself.

Now, do you also remember the old interview questions which are now laughed out or actively belittled such as "give me 3 qualities and 3 flaws", "what's your biggest fear" or "what's your spiritual animal and why".

Well, I intend to show you that the new type of behavioral questions that we now see everywhere, all the time, are just the upgraded fancy version of those old ones, with very little evidence of them being correlated with great hirings.

The Promise That Started It All

These behavioral questions—cornerstone of the structured interview movement that swept through HR departments in the 2000s—were supposed to revolutionize hiring.

The promise was backed by data: structured interviews predicted job performance at 0.50-0.60 correlation versus 0.20 for traditional interviews. Google validated it. Amazon standardized it. McKinsey swore by it. Companies reported 40% better hiring decisions and dramatic reductions in bias.

The logic was bulletproof: past behavior predicts future performance. Same questions for everyone. Same scoring rubric. No more hiring on "gut feel" or because someone went to your college. By 2020, virtually every major corporation had adopted behavioral interviewing as gospel.

And it's TRUE. Past behavior is a great indicator of future performance. I'm not challenging this fact, but I want to address how many - or most - companies started from there and turned this proven assumption into a theatrical nonsense, unable to effectively assess great candidates from bad ones.

The Reality Check

From my personal experience as a corporate and freelance recruiter (both running interviews and being interviewed), discussions with peers, feedback from hundreds of candidates about their interview processes, here is the trend I see emerging:

ATS and all hiring tools made it so easy now for any interviewer (recruiter, hiring manager, team member) to walk into an interview with a structured set of questions to follow. I mean, in theory, that's awesome.

In practice, what I see is that overly prepared interviews result in the interviewer shooting each one of their questions one after the other - even though they're not related at all - just to check or uncheck a list of skills that has been pre-defined.

Let's list a few as examples:

Describe a time when you disagreed with a team member. How did you resolve the problem?
Tell me about a time when you failed, and how you handled it.
What is the most difficult/ challenging situation you've ever had to resolve in the workplace?
Tell me about a time when you disagreed with a supervisor.
Describe a time when you were able to motivate unmotivated team members.
Have you ever been lied to? How did you handle that situation?

Well you get it, there's an infinite number of questions like those, or variations of them. Most companies I come across today have implemented behavioral questions like this.

So what's wrong with them?

Problem #1: They Are Overwhelmingly Vague

Let's take one as example: "Tell me about a time when you failed, and how you handled it."

Let's imagine you're in an interview right now, and let's say you have 7 years of experience, and you're asked to answer one of those. You literally have 5 seconds to run your entire 7 years career in your head, in search for the ONE occurrence which will perfectly illustrate the question asked and how you, as a person, handled the situation.

Then, you have to turn this recollection of events into a nice and neat 1 minute answer, while applying the STAR method or one of its equivalents.

For someone who has not prepared the question beforehand, 5 seconds is a short window to come up with a solid answer to a question as vague as this.

We could argue: "Well you should prepare your answers and examples before the interview". I'll come back to this a bit later, but as we'll see, preparing beforehand defeats the whole purpose of those questions.

Problem #2: They Focus on the Negative

They are most of the time "negative" (they involve a difficulty, a problem, an obstacle, something that you had to overcome).

It is true that assessing candidates requires understanding how they would act and react when facing challenges, rather than just focusing on the "easy stuff". But it adds a new layer of difficulty for candidates.

On top of finding the best example in their working life to answer the question, they now have to be extremely careful about HOW they present the story. They need to describe the problem (with our example: "A time you failed") in a way that shows they:

Made a mistake (but not too big!)
Showed exceptional maturity as to how they handled it emotionally
Never did it again
Used their learnings to turn their next iteration into a big WIN

All that, in 5 seconds of thinking after the interviewer asked the question. Easy, right?

The Candidate's Dilemma

Many people would argue that it's not like that, that interviewers are not looking for a perfect answer. Well I would argue that it's exactly like that, 95% of the time.

Let's dive just a bit deeper. Let's say you either try to be honest, at least partially in your answer, OR that you're not a professional in rhetoric and your answer is only average.

You admit to a mistake too big? You may be assessed as someone reckless, unreliable, or just bad at your job.
You explained that you did not handle the mistake too well at first? You may be someone immature and incapable to act under pressure.
You did not draw an invaluable lesson from that mistake and turn it into a big win? You're not a big learner, and you probably don't have a great potential.

It's not that interviewers are bad -not all of them at least-, but the questions are. They're just red-flags generating machines, unless you've prepared and rehearsed them like a paid actor.

And since those questions are part of a structured interview process, they are likely to be asked to all candidates. So your answers will unavoidably be compared with dozens of other answers to the same questions.

A Tale of Three Candidates

Let's compare 3 candidates, and let's say they have the same technical level and expertise. They are all asked the same question: "Tell me about a time you failed, and how you handled it"

Candidate #1: A great candidate, with exceptional soft skills. However, she did not rehearse this specific question, and although her answer is somewhat ok, this question made her look less good than she actually is. If this scheme repeats itself over a bunch of similar questions, she'll be assessed as a "good but not great" candidate.

Candidate #2: A good candidate, with solid soft skills. But she's also well aware that she has flaws, and she answers this question with a bit too much honesty - let's say she admits that she handled a conflict poorly. Self-awareness and honesty should normally be considered prime soft skills, but they are not considered at all here - because they do not enter in the assessment prism of this specific question. She is extremely likely to be assessed as a poor candidate choice.

Candidate #3: Also good at her job, but she carries around a significant amount of flaws and poor behavioral habits, and has been part of many conflicts in her past jobs. However, she's particularly good at expressing herself and at using persuasion techniques. She did a good job at rehearsing behavioral questions and preparing concrete examples that are partially fabricated, for the purpose of acing all those interview questions - and she does.

I'd say that only 10% of interviewers are trained enough to spot the best candidate among those 3. Again, the other 90% are not at fault here. The poor question choices are.

The Preparation Paradox

So, should all candidates just rehearse those questions beforehand?

I remember 10 years ago when advice on the web was to train yourself for interviews by preparing a 5 or 10 min speech recap about your whole professional journey. I mean the idea was good, the underlying goal was to get better at talking about yourself in a clear and articulated manner.

But today good recruiters gently cut you off after 15 seconds when you start your rehearsed monologue, because they know they won't learn anything interesting from a rehearsed speech.

And somehow all those new generic behavioral questions are bringing us back into a - slightly different but kinda same - loop.

You have tons of websites giving you examples of behavioral questions you might encounter in interviews, and cheat codes about how to craft solid answers. And if people who give the best answers to those questions in interviews are only candidates who prepared them with great attention (and most likely twist the reality in their favor) or rhetorical geniuses, then what are they good for?

They would be in no way whatsoever a genuine indication about how a person would act and react at work. They only measure interview performance, not job performance.

A Better Way Forward

So what should we do instead?

Well, there are plenty of ways to assess candidates (and AI will bring even much more options to the table in the upcoming years).

Let's stick to the skills we are usually trying to assess with behavioral questions, namely the soft skills, such as collaboration, humility, curiosity, proactivity, emotional intelligence, resilience, adaptability… etc.

Oddly enough, I think the techniques shared by Geoff Smart - 17 years ago - in "Who: The A method for Hiring" are still the best ones you could ever implement.

The Project Deep-Dive Method

Instead of shooting a vague and generic question about a random challenge the candidate may have faced somewhere, somewhen in her lifetime; dive in with her on a specific project she's worked on, on a definite part of her journey.

And instead of shooting one question and waiting for the well rehearsed answer, ask her to go through the project with you, and dig with 20 mini follow-up questions:

"Who did you work for / with?"
"How did it go?"
"What were the results at the end?"
"Oh it did not work out, can you explain why?"
"Oh you exceeded expectations, well done! How do you explain this great success?"
"What do you think you could have done differently for that project to have gone better?"
"What did you learn from it?"

And then do the same with another specific project, during the same or another role the candidate had.

Why This Works

The advantages of this are multiple:

✓ You're jogging the candidate's memory about a specific project

✓ You're giving her time to answer each question more easily, therefore decreasing the pressure

✓ You're choosing the direction each time you ask a new question so it limits the candidate's ability to regurgitate rehearsed answers, and it increases authenticity

✓ You're building a better discussion flow

The Bottom Line

It's a bit harder to do, but I guarantee that this process allows you to assess any candidate in a much more efficient way.

And it's nicer for you as an interviewer because you don't get preheated answers, and it's nicer for the candidate because it feels more genuine.

Win win win situation.

Try it. Now. And hopefully you'll recruit better talent.