The people I wrote off in the first five minutes

A habit that feels like discernment and works like a cage.

You’ve done this. I’ve certainly done this.

Someone new joins, they talk for a few minutes in a meeting, and somewhere in your head a file opens and closes. A bit slow. A bit uptight. A bit much. Not someone you’d get a coffee with. You don’t say any of it out loud and you may not fully notice you did it, but from that point on you are a slightly different person around them, and you will be for as long as you both work there.

It reads awful written down. It’s also completely ordinary, everyone does it, and I’ve come to think it’s one of the more expensive habits we carry around.

What it costs, specifically

The obvious cost is that it makes you cynical, and cynicism is corrosive in a way that’s hard to reverse once it sets.

The cost that took me longer to see is structural. Every time you close that file on somebody, you are choosing to stay exactly where you are. You’ve made a decision about what a person has to offer, in five minutes, on a sample size that would be laughed out of any analysis you’ve ever produced at work. And you’ve made it based on surface qualities that correlate with almost nothing.

The person who seems slow may be careful in ways you aren’t. The one who seems uptight may be seeing a risk nobody else has clocked yet, and you’ll find out in about four months. The one who seems like too much may be the only person in the building who can hold a room together when things go badly, which is a rarer skill than being pleasant in a status update.

Everyone you meet has lived a story you couldn’t invent. I don’t mean that as a sentiment. It’s closer to arithmetic. They have been in rooms you’ve never been in and made calls you’ve never had to make, and some meaningful fraction of that is transferable to you for the price of a conversation you keep declining to have.

Why it feels like a skill

The reason this is hard to shift is that snap judgment isn’t a flaw in the system. It’s the system working as designed.

Reading people quickly is useful. It’s how you work out who to escalate to, who needs context before a decision, whose “sounds good” means yes. Most of us who are decent at our jobs have that instrument running constantly, and it’s usually right about small things. Which is exactly why we trust it on large things, where it isn’t.

The instrument is good at predicting behaviour in the next ten minutes. It is very bad at predicting what somebody knows, what they’ve survived, or what they’d be like to work with over a year. We use one reading for both, because it’s the only reading we have.

So this isn’t really about being nicer. It’s about knowing the range of an instrument you’re already using, and not pointing it at questions it can’t answer.

The new version of the problem

The thing that used to correct us was friction.

You got stuck, and the fastest route out was turning around and asking the person two desks over. Even if you had privately filed them away, you needed something from them, so you talked. And often that conversation was what overwrote your first impression, without either of you noticing a correction had occurred.

That friction has largely disappeared. When you get stuck now, you ask a model and you get a decent answer in four seconds, without ever engaging with a human you’d already dismissed. I use it constantly and I’m not going to pretend I don’t. But the accidental mechanism that used to force us back into contact with each other is no longer running by default, and I don’t think there’s an obvious replacement for it.

Which means the correction now has to be deliberate, and deliberate things mostly don’t happen.

The caveat that matters

None of this means you owe everyone unlimited patience. Some people at work are genuinely difficult in ways that will cost you, and the right response is distance. Protect yourself. I’d rather say that plainly than leave someone reading this feeling obliged to keep opening themselves up to a person who has already shown them who they are.

What I’m talking about is narrower. It’s the judgment formed in the first five minutes, before any evidence exists at all. That one is nearly always wrong, and it’s expensive in a way you’ll never receive an invoice for, because the cost is entirely made of things that didn’t happen.

I’ve learned an enormous amount from people I wouldn’t have bothered speaking to ten years ago. About work, about myself, about how to be a person in a room. I’d have missed all of it if I’d kept trusting my first read.

Everyone is worth more than your first impression. That’s not generosity. It’s just accuracy.