There's a guy I worked with years ago who drove me up the wall.
Every design review, he'd sit there with his arms folded and pick holes. Not helpful holes. Pedantic ones. He'd find the one edge case nobody had thought about and hammer it until the meeting ran out of road. I left those rooms with my jaw tight, thinking the same thing every time: he's obstructive. It's who he is. It's his personality.
I was wrong. Not a little wrong. Completely wrong.
The Question I Should Have Asked
Grace Judson writes about a reframe I wish I'd found twenty years earlier. When someone's behavior makes no sense to you, stop and ask one question:
Why would a reasonable person act this way?
She's blunt about how ridiculous it sounds. In her piece on empathizing with someone who's wrong she writes: "I know. Their behavior isn't reasonable ... from your perspective. But try to answer the question anyway."
The last bit is the whole trick. Answer it anyway. Not because they've earned the benefit of the doubt. Because you need better information, and you're not going to get it while you're busy diagnosing their character.
What I Found Out About the Guy With the Folded Arms
He'd been burned. Two years before I joined, he raised a concern about a data migration in a design review. Got told it was out of scope. Shipped anyway. It corrupted six weeks of customer records, and he spent the next three months of evenings unpicking it by hand. Nobody who made the call to ship was still at the company. He was.
So he wasn't obstructive. He was the only person in the room still paying the bill for being wrong.
Once I knew, everything changed. Not his behavior. Mine. I started giving him the edge cases before the meeting instead of during it. He stopped needing to ambush anyone. Same man, same personality, completely different meetings.
I hadn't fixed a difficult person. I'd stopped mistaking a situation for a person.
This Has a Name, and It's Older Than Your Career
Psychologists call it the fundamental attribution error. Lee Ross named it in 1977 in a paper with the wonderful title "The Intuitive Psychologist and His Shortcomings". The idea is simple and deeply annoying once you spot it in yourself.
When you do something stupid, you know why. Traffic was bad. You'd been up half the night with a sick kid. The brief changed on Friday and nobody told you.
When someone else does something stupid, you skip all of it and go straight to the person. They're lazy. They're difficult. They don't care.
Same behavior. Two completely different explanations. And you apply the generous one to yourself without noticing you're doing it.
Here's the part to worry about as a leader. You have less information about your people than they have about you. You see the output. You don't see the two-hour commute, the parent in hospital, the last boss who punished honesty. You see arms folded in a design review.
Why It Matters More When You're the Boss
Gallup estimates managers account for at least 70% of the variance in employee engagement scores across business units. Seventy percent. Not the perks. Not the mission statement. You.
So every time you label someone instead of understanding them, you're not making a private judgment. You're setting the weather for a person's working life.
And once you've labeled somebody, you stop collecting evidence. You've filed them. Everything they do afterwards gets read through the file. They're late because they're unreliable, never because daycare ran over. They pushed back because they're negative, never because they spotted something you missed.
My own research on bad bosses found 99.5% of people surveyed said they'd had one or more types of bad boss. Not "a rough patch with a manager." A bad boss. I don't think most of those bosses were monsters. I think most of them were people who filed their team early and never reopened the file.
The Bit Nobody Wants to Hear
This is not about being nice.
Asking why a reasonable person would act this way does not mean the behavior is fine. It doesn't mean you skip the hard conversation. Somebody rude in front of a client still needs telling. Somebody missing deadlines still has a problem to solve.
What changes is the problem you're solving. If you believe the problem is their character, your only lever is pressure ... push harder, escalate, performance-manage them out. If you believe the problem might be their situation, you suddenly have a dozen levers. Change the workload. Change the information flow. Change the deadline. Change yourself.
One of those approaches has a track record. The other is how you end up hiring a replacement in six months and watching the same thing happen all over again to the new person.
I've written before about how surviving a toxic workplace isn't a career plan. The flip side is true for leaders. Grinding people down until they quit isn't a management plan.
How to Run the Question for Real
Three steps. None of them take long.
Stop before you respond. Grace suggests three deep breaths. Sounds soft. Isn't. You're giving your slow brain a chance to catch up with your fast one, which has already decided the person is an idiot.
Ask the question and answer it. Properly, in your head, all the way through. "Why would a reasonable person fold their arms and pick this apart?" Come up with three answers. At least one of them will be uncomfortable, because at least one of them will involve you.
Go and check. This is the step people skip. The question isn't a guessing game you win alone at your desk. It's a prompt to go ask them. "You push hard on data integrity in these reviews. What have I not understood about why?"
Then shut up and listen. Listen in the way where you're prepared to be changed by the answer.
What It Costs You to Be Right
We skip all this because labeling people is fast and it feels good. "He's difficult, full stop" closes the loop. No further thought required. You get to be the reasonable one.
Curiosity is slower, and it costs you something. It costs you the certainty. Sometimes the answer is a reasonable person acted the way they did because their manager did something thoughtless. Horrible thing to find out about yourself on a Tuesday afternoon.
But you'd want to know. Wouldn't you?
There's someone on your team right now you've already filed. You know who. The name came to you two paragraphs ago.
Go and ask them the question this week.