We talk about people as though our assessments of them are fixed facts rather than working hypotheses.

“She’s not quite ready”. “He’s solid and dependable, you know”. “She’s not really leadership material”.

These aren’t unusual things to hear when I talk to leaders about their people. They’re the normal language of management.

What’s less normal is asking when our view was last updated, or whether it ever has been.

The Pygmalion effect in management has been studied and replicated many times since the mid-1960s, when psychologist Robert Rosenthal began testing a simple but uncomfortable idea: that what we expect of people shapes how they perform.

The most well-known test of this was in a California school, where teachers were told, falsely, that certain students had been identified as likely to show significant intellectual growth. Those students – chosen at random – did exactly that. The teachers hadn’t changed the curriculum but they had changed what they expected from those students.

The finding has been debated, tested and replicated many times since but what has held up consistently across the research is the behavioural finding – that teachers treated the students they believed in differently, and those students felt it. When the same experiment was run with adult soldiers in Israeli Defence Force training programmes in the 1980s, the results were the same. The instructors who were told their trainees had exceptional potential produced trainees who outperformed the control groups on objective tests and the instructors had no idea they were behaving any differently.

It isn’t the finding that intrigues me about the Pygmalion effect, it’s the mechanism. The teachers didn’t share their higher expectations. They expressed them without really knowing they were doing it by giving the students more patience, slightly longer explanations and a willingness to wait for an answer. The students felt it and they responded to it.

The Pygmalion effect has a darker version.  Where high expectations lift performance, low expectations tend to hold it down. That version is called the Golem effect, and it runs in the same way. Managers may express frustrations with team members in private but they don’t tend to share low expectations.

Instead, they express them in opportunities not offered, in giving more challenging work to someone else, in less time, less patience, less investment. The person on the receiving end recognises it.  They may not know or understand why but they feel it.

Why we don’t question it

Labels don’t feel like labels to the people applying them. They feel like accurate assessments because the evidence has built up to support them. This is confirmation bias in its workplace form. Once we’ve placed someone in a category – high potential, difficult, solid but not exceptional – we notice what confirms it and we discount contradictory evidence. Our opinion solidifies, and the person it’s about rarely gets the chance to challenge it.

There’s a related finding called the idiosyncratic rater effect. Studies suggest that performance ratings tell us as much about the person doing the rating as about the person being rated – that our assessments reflect our own standards, our own experience, our own ways of seeing. That’s not a comfortable finding if you’re responsible for decisions that impact people’s careers.

What it does to the people around you

Being labelled doesn’t just affect someone’s next promotion. It changes how they show up each day.

The person who’s been quietly written off stops raising their hand. They don’t put themselves forward for the project, don’t share the idea in the meeting, don’t speak up when they probably should. Not because they’ve decided against it but because they’ve learned, without anyone telling them, that it probably isn’t worth it. The label and the behaviour start to confirm each other, and the original assessment ends up looking more accurate than it ever really was.

Think about what that means in a team of twelve where three or four people sit in that category. You’re not looking at a marginal performance issue. You’re looking at a significant amount of potential that never gets used.  Gallup’s research consistently finds that teams where managers hold fixed, undiscussed views about their people show lower engagement and less discretionary effort.

What you can do differently

The antidote to labelling isn’t relentless positivity or a new performance framework. It’s simpler and harder than either of those things: being willing to change your mind about someone, and then doing something about it.

That means updating your view of what someone can do and then acting on it before they feel 100% ready. It means giving them the task or project that will stretch them, rather than the one that feels safe. It means speaking up for them when they’re not in the room, and expecting more from them than they’re currently showing.

Carol Dweck’s work on mindsets is usually applied inward – whether you believe your own abilities are fixed or can grow, but it works just as directly the other way. A fixed view held by someone with power over your career is harder to shift than any fixed view you hold about yourself. You can decide to think differently about your own potential. It’s much harder to change what your manager has quietly decided about yours.

None of this requires a huge change. It starts with a more honest question about your own assumptions.

Think about the people in your team you’d describe as solid but not exceptional.

When did you last genuinely revisit that view? What would it take to shift it – and have you created the conditions where you’re likely to see what you’d need to see?

And if your view turned out to be wrong, what has that cost them?