To make better decisions under uncertainty, get the read right before you weigh the options. Most bad decisions aren't poor choices between alternatives. They're sound choices made on a wrong picture of what's happening.
That sounds obvious. The research says almost nobody does it well.
The problem isn't the decision
In surveys of 106 C-suite executives from 91 companies in 17 countries, Thomas Wedell-Wedellsborg found that 85% agreed their organizations were bad at problem diagnosis, and 87% agreed it carried significant costs (Harvard Business Review, 2017). His conclusion: organizations are generally good at solving problems. What they struggle with is figuring out what the problems are.
Most decision frameworks skip that step. They help you compare options, weigh risks and assign owners. They all assume you've framed the situation correctly. When you haven't, the framework just helps you make the wrong decision more efficiently.
Why experience makes it harder
Gary Klein's research on firefighters, military commanders and intensive-care nurses found that experts rarely deliberate between options. They recognize a situation as similar to one they've seen before, and act. He calls it recognition-primed decision making, and it's why experienced leaders are fast.
It's also why they're wrong in new situations that only resemble old ones. The pattern fires before you've checked whether it fits.
Here's what that looks like. Say you worked with a team two years ago that stalled at every decision point. You diagnosed analysis paralysis, intervened, and it worked. Now you're with a new team that slows down at decision points too. The read comes instantly. But this team isn't paralyzed. The stakes just went up, and they're being careful. Careful is appropriate. Pushing them to decide faster is exactly the wrong move, and you can't see it, because your history is doing the seeing.
Small errors compound
Navigators know this one. In dead reckoning, a one-degree heading error doesn't fix itself. It grows with every mile you travel. Over a hundred miles, it puts you more than a mile from your target. Over a thousand, about seventeen.
A misread decision works the same way. At any single step, nothing feels wrong. You're just executing. But each decision built on the first one carries the error further, and by the time it shows up, it's expensive to unwind.
What to do instead
In cave diving, the rule when something goes wrong is stop, breathe, think, act. In that order. The whole sequence takes about ten seconds. In those ten seconds you decide whether you're solving the real problem or the panic version of it.
Leadership decisions deserve the same order. Before you choose, run three questions:
- What do I see? The facts you could point to, separate from the story you've attached to them.
- What is driving it? At least three possible explanations, not one. Ask what you'd expect to see if each were true.
- What does this moment need? Sometimes it's a decision. Sometimes it's one small move that tests your read first.
Then add one more question before you commit: What would change my read? If you can't name anything that would, you're not holding a hypothesis. You're defending a conclusion.
The questions to watch for
Some situations look like decision problems and aren't. A team that keeps asking for more data might need information. Or it might need clarity about who's allowed to decide. A room that agrees quickly might be aligned. Or it might be reluctant to disagree with you. Each wrong read sends you to the wrong fix: more analysis, more alignment meetings, more pressure.
Uncertainty doesn't go away with a better read. That's not the goal. The goal is to be wrong less often about what's in front of you, and to find out sooner when you are. Hold the read loosely.
That's the discipline at the center of Diagnostic Leadership: reading what's actually happening before deciding what to do.