Competence Before Personality
There’s a version of being likable that functions as a substitute for being good.
I don’t mean this as a criticism of likable people. I mean it as an observation about what we reward and why. Charm is legible. It arrives fast. You feel it in the first fifteen minutes of meeting someone — the ease, the attentiveness, the sense that they’re genuinely enjoying the conversation. Competence takes longer. It requires something to actually happen. You have to wait and see.
So we reach for the faster signal first. Which means a lot of professional trust gets scaffolded on something that has nothing to do with whether the person can actually do the work.
I’m not sure this is wrong, exactly. It might even be efficient.
If you have to extend trust before you have evidence, defaulting to warmth and rapport isn’t irrational. It’s a reasonable proxy. People who are easy to work with are actually easier to work with. People who communicate well tend to surface problems earlier. The soft skills matter. Dismissing them as surface-level is its own kind of mistake.
But there’s a gap between “warmth is a useful signal” and “warmth is what I’m actually tracking when I decide to trust someone.” The second thing happens more than most people want to admit. And the gap between those two positions is where a lot of trust failures quietly originate.
Here’s what I’ve noticed about the people I actually trust to do consequential work.
They’re not particularly charming. Some of them are, but it’s not the thing. What they share is something closer to a track record of accurate self-assessment.
They say they can do a thing, and then they can do it. If they’re not sure, they say that. If they’re over their head, they say that too — usually early enough to do something about it. They don’t manage the appearance of competence; they mostly just try to be competent and let the appearance follow.
That sounds obvious. It’s not common.
The management of appearances is nearly universal in professional settings. It’s adaptive behavior in environments where how you’re perceived matters as much as — or more than — what you actually produce. You learn to project confidence when you’re uncertain, to avoid admitting gaps in knowledge, to signal alignment with expectations even when you have doubts. The incentives for this are real. The costs are mostly invisible until something goes wrong.
The thing about charm-first trust is that it doesn’t fail gradually.
Competence-based trust has a natural error-correction mechanism. Someone misses, you update your model, you recalibrate how much you rely on them in that particular area. The trust narrows without collapsing. You can still use them for the things they’re actually good at.
Charm-based trust fails differently. It tends to fail suddenly, and completely, because what’s breaking isn’t a specific capability you overestimated — it’s the underlying model of who the person is. You weren’t tracking performance, you were tracking character. And when the character model breaks, it breaks everything.
I’ve watched this happen with people who were genuinely likable, genuinely well-intentioned. The charm wasn’t fake. But it was covering a competence gap that no one asked about because no one wanted to be the person who asked. The trust was real. The foundation wasn’t.
So here’s the tension I sit with.
Charm comes first for a reason. It’s not just a cognitive shortcut — it’s also a genuinely relevant feature. Working with someone who communicates badly, who can’t navigate conflict, who makes every interaction feel like a negotiation — that has real costs even if they’re technically excellent. The friction compounds. Good work requires collaboration. Collaboration requires getting along.
The claim I’m actually making isn’t that personality doesn’t matter. It’s about what goes underneath.
Personality is a reasonable feature to factor in once you have some basis for trusting competence. The trouble is we invert that order constantly. We establish warmth first, and then interpret competence through the lens of warmth. We’re more likely to overlook missed deadlines, explain away ambiguous outcomes, attribute problems to circumstances rather than the person — when we like them.
This isn’t a character flaw. It’s just how it works. Affect shapes cognition. Warmth is a prior. We see what we expect to see.
The rubric I try to use — imperfectly, inconsistently — is something like: can I name something specific this person got right, under real conditions, when something was at stake?
Not “do they seem sharp.” Not “were they prepared in the meeting.” Not “do they give good answers to questions I ask.” All of those can be produced by someone optimizing for perception rather than reality.
What I’m looking for is a data point where their prediction matched what happened, where a thing they committed to actually shipped, where the thing they said they’d handle got handled. Evidence of contact with actual conditions rather than the performance of someone who’s good at talking about them.
The bar doesn’t need to be high. It can be a small thing. One concrete moment of accuracy beats ten impressive impressions.
Admitted uncertainty: I don’t know where to draw the line between maintaining this standard and becoming someone who withholds trust so defensively that nothing ever gets started. Extending trust early is sometimes the only way to generate the evidence that would justify extending it. You have to move before the proof arrives. The question is how much trust to place in what, at what stage, based on what early signals — and I don’t have a clean answer for that.
Falsifiable claim: In working relationships that last more than six months, the ones built on accurate early competence assessment rather than primarily on rapport will have fewer late-stage trust failures, controlling for the magnitude of what was being attempted. The charm-first relationships will fail when the stakes get high enough to finally test the foundation.
I keep returning to something that’s almost embarrassing in its simplicity: the people I trust most are easy to trust for boring reasons.
They do what they say. They know what they don’t know. When something goes wrong they name it without being asked. There’s very little daylight between what they tell me and what’s actually happening.
This is not glamorous. There’s no charisma category on that list. And yet it’s the thing that actually makes me willing to put weight on someone — to hand something over and not check every day, to take them at their word when the stakes are real, to let them handle things in rooms I’m not in.
Charm is a door. Competence is the floor.
You can have a beautiful door and still fall through the floor. What you want is to be careful which rooms you actually walk into.