Guess What’s Real: How HR Made Intuition a Profession

There is a question that has been asked billions of times across the world. It sounds innocent — almost routine — and every single time it is asked, neither side of the conversation can answer it honestly. The candidate does not know what will happen. The interviewer never checks whether it did. And yet the question continues to exist as a polished, professional, supposedly science-backed tool for evaluating a human being.

The question is this: "Where do you see yourself in five years?"

Stop for a moment. Can you answer it yourself? Not with the kind of answer that lands well in an interview, but with something that actually corresponds to reality. Most likely, no. And that is not because you are bad at planning your life. It is because no one can predict the future with the precision this question demands. No one ever could. Science has known this for a long time.

What Science Says

In 1998, psychologists Frank Schmidt and John Hunter published a study that became the single most influential piece of research in personnel selection over the past eighty-five years. They analyzed thousands of individual studies conducted by different methods in different countries and asked one simple question: of everything we use when hiring, what actually correlates with how a person performs on the job?

The answer was uncomfortable for the entire industry. Education correlates with job performance at just 0.10. Work experience, at 0.18. Statisticians call these coefficients "unlikely to be useful." In other words, all the information in a resume — the thing HR professionals spend months scrutinizing — is almost entirely disconnected from how a person will actually perform in a new role.

Unstructured interviews — the kind where you feel a "spark" and try to read someone’s character from their eyes and the way they speak — account for less than fifteen percent of the variation in their subsequent performance. Fifteen percent. That means that in eighty-five out of a hundred cases, your "right" candidate could have been the wrong one, and the person you rejected could have been flawless.

Education accounts for ten percent of the outcome. Experience for eighteen. And the "inner world" of the candidate — the thing we try to read during the interview — for almost nothing.

But Schmidt and Hunter’s findings are only the beginning. That same year, Daniel Kahneman and Amos Tversky described a phenomenon they called the "planning fallacy." People are systematically incapable of predicting even their own future behavior — even in situations where past experience is practically screaming at them about what will actually happen. We build plans based on the ideal scenario, carefully excluding every obstacle, every delay, every catastrophe that has happened to us before in exactly the same circumstances. And we do it again and again. Not out of bad faith. Not out of poor memory. It is a structure of thought. It is wired into how our brains work, and it cannot be overridden by willpower alone.

Schmidt, F. & Hunter, J. (1998). The Validity and Utility of Selection Methods in Personnel Psychology. Psychological Bulletin, 124(2), 262–274. | Kahneman, D. & Tversky, A. (1979). Intuitive prediction: Biases and corrective procedures. TIMS Studies in Management Science, 12, 313–327.

Three Rituals Nobody Questions

If you want to see how an industry turns guesswork into a profession, look at three practices so deeply entrenched that no one questions them anymore. They exist not because they have proven their effectiveness. They exist because someone once decided they should exist — and after that, no one thought to check whether they actually work.

The first is the attempt to read a candidate’s motivation. "What attracts you to this role?" "Why do you want to work with us specifically?" The idea is that if you ask the right question in the right tone, the person will reveal their inner world — and you will understand how serious they are, how genuine their intentions are, how much they are "one of us." But there is a fundamental problem this logic ignores: people do not understand themselves well enough to describe their own motives with the precision this kind of assessment requires. Research by Koehler and Poon revealed something important: the strength of intention — that feeling of "I want to do this" that a person experiences when answering the question — has almost no bearing on whether they actually follow through. Between what a person feels now and what happens later lies a gap that no interview can bridge.

The second practice is loyalty assessment. Companies invest serious budgets and effort into determining, before a person even starts working, whether they will stay. Over the past several decades, an entire ecosystem of tools has been built for this purpose: questionnaires, specialized tests, multi-stage procedures where every question is supposed to somehow reveal the candidate’s future devotion.

But research published in Frontiers in Psychology in 2022 states with rare candor what practitioners have long suspected: "loyalty research is still at an experimental stage." Over forty years of studying this phenomenon, dozens of measurement models have been developed — and not a single one produces a reliable forecast. And here is why that makes sense: loyalty is not a property of a person that can be measured like height or weight. It is the state of their relationship with a specific company at a specific moment in time. It changes every day. You cannot capture it on a Monday and expect it to remain the same on Thursday.

The third practice — and perhaps the most telling of all — is that very question about five years. It exists as a professional tool with one single purpose: to determine how serious the candidate is and whether their time horizon aligns with the company’s. The logic seems airtight — until you start asking uncomfortable questions. In five years, neither the company nor the candidate will be the same. The market will have shifted. The strategy will have changed. The person’s circumstances may have transformed so fundamentally that their priorities become unrecognizable. And despite all of this, we ask them to tell us now what will happen then. And we evaluate them based on that answer. But why exactly this is impossible — why no answer to this question can ever be correct by definition — deserves to be examined on its own.

What No One Can Foresee

There is a risk that exists in every hire, but that almost no one speaks about directly — as though acknowledging it would itself be a professional failing. This is the risk of unpredictability. Not rare unpredictability. Not exceptional unpredictability. Fundamental, unavoidable unpredictability, present in every single decision about whether to bring someone on board.

Consider it from the candidate’s side. A person joins a company fully engaged. They are serious, genuinely committed, ready to work. And then, six months later, something happens — the kind of thing that happens in every person’s life. A sharp, painful shift across every coordinate of their existence. A loved one is diagnosed with cancer. Or they themselves receive a diagnosis that changes everything. Or someone they are closest to leaves. Or a quiet, gradual, but inexorable realization sets in: the things that seemed important when they were sitting across from the interviewer no longer are. Their priorities have shifted. Not out of bad faith. Not out of any lack of professionalism. Something simply happened — something that did not yet exist at the moment the hiring decision was made.

None of the questions asked during the interview could have warned about this — simply because what needed warning about did not yet exist.

Now consider the same situation from the company’s side. It hired someone for a specific task, for a specific reality that existed at that moment. Three months later, the market turned in a different direction. Or the strategy changed under the influence of events that no one in leadership foresaw. And the task for which the hire was made ceased to be that task. A person who was an absolutely perfect fit for yesterday’s objectives may become entirely irrelevant to today’s — not because they changed, but because the entire reality around them did.

This risk is not an exception. It is a business risk that exists in every hire, on both sides simultaneously, and that cannot be eliminated by any procedure, any test, or any hours-long assessment. It exists because the future — that specific future in which this specific person will work — simply does not exist at the moment we decide to hire them. At the moment of hiring, only the present exists. And the only thing that can genuinely be assessed in this present is a person’s skills and their past experience. Everything else is speculation about something that has not yet materialized.

Why We Continue

The question is not why the HR industry exists as such. The question is why it exists in exactly the form it takes today. Why do we invest significant resources in tools whose effectiveness has not been rigorously demonstrated — and still feel like thoroughly professional, competent people?

The answer is simple, but deeply uncomfortable for anyone who has built a career inside this system. People cannot exist in a space of total uncertainty. They need the feeling of control over a situation — even if that control is largely illusory.

"Cultural fit," "growth potential," "red flags," "intrinsic motivation" — these are not scientific categories in the way they exist in psychology or statistics. They are categories of confidence. Professional language creates the impression that every hiring decision is made rationally, on the basis of data, with a deep understanding of human nature. But behind each of these terms lies exactly the same thing that lies behind any prediction about the future: intuition. Simply wrapped in professional terminology.

The industry evolves not because the tools become more precise. It evolves because the need for the illusion of control never disappears — it exists in every organization, at every level. Each new personality test, each new loyalty assessment model, each new interview methodology is a new, more sophisticated way of convincing ourselves that we can know what is fundamentally unknowable.

What Actually Works

If everything above sounds like a verdict on the industry, it is not. Science does offer an answer to the question of how to hire effectively. It simply looks different from what we have grown accustomed to.

Schmidt and Hunter, in the very same study where they exposed the limitations of traditional methods, also pointed to what does work. Structured interviews with a validity of 0.51 — where the same questions are asked of every candidate and evaluation is conducted on a single, unified scale. Cognitive tests with the same validity. And — particularly important for practice — work samples: tasks that a person completes right there during the assessment, doing exactly what they would do on the job. Not "Tell me about yourself." Not "What is your dream?" A concrete task. A concrete result. A concrete measurement.

In 2022, Sackett and colleagues conducted a new large-scale meta-analysis — the largest in twenty years — and confirmed, and in many ways revised, these findings. Most previous validity estimates turned out to be inflated due to errors in statistical corrections.

But one conclusion only grew stronger: structured interviewing proved to be the single most powerful predictor of job performance — stronger than cognitive tests, stronger than personality assessments, stronger than all the instruments that form the backbone of modern HR practice.

But only under one condition: the interview must be conducted strictly according to protocol — the same questions for every candidate, a unified system for evaluating answers, and every question must connect directly to the specific requirements of the role. Nothing about the soul. Nothing about five years. Only what can be measured right now.

The Only Thing That Can Be Assessed

When we hire someone, there are exactly two layers of reality.

The first is what exists right now: a person’s skills, their experience, their demonstrated ability to perform certain tasks — confirmed by concrete results.

The second is everything else: their motivation tomorrow, their loyalty a year from now, the way their life will unfold over the next five years. The first layer can be assessed. The second cannot. Not because our diagnostics are insufficiently refined, or because better methods exist that we have not yet discovered. But because the second layer simply does not exist at the moment of assessment. It is made up of events that have not yet happened, decisions that have not yet been made, circumstances that no one yet knows about — not the candidate, not the company, not anyone.

It is more honest to acknowledge this than to keep pretending. The best hiring decisions are made not when we finally "read the soul" of a candidate, but when we stop trying to do so. And begin assessing the one thing that can genuinely be assessed: what a person can do, and what they have already done.

The rest is guesswork. Simply very professionally dressed up.