How to Make a Decision You Won't Regret

I've sat in this room more times than I can count. A founder or exec leans back, says they already know what customers want, and says it with enough weight that the room goes quiet. Sometimes two leaders at the same company want opposite things, and the louder one wins. The call gets made in a minute. Everyone nods. And a few people walk out uneasy, even if nobody says so.

Months later the thing ships and underperforms. Here's the part that stings: almost nobody is surprised. The doubt was in the room the whole time. It just never had a way to speak.

Confidence isn't proof

We treat confidence like proof. It isn't. Feeling sure has almost nothing to do with being right. But in a fast meeting, the most certain voice sounds like the most credible one, so it wins.

There's a habit that makes this worse. We judge a decision only by how it turned out. If the bet pays off, we call it smart. If it flops, we call it dumb. But a reckless call can get lucky, and a careful call can hit bad luck. Judge the choice by the result and you learn the wrong lesson every time: you reward the loud guess that worked and punish the thoughtful one that didn't.

A better test than "did it work"

So here's a better test. A decision you won't regret is one you'd make again with the same information, even if it flopped. The goal isn't to be right every time. It's to make calls that were at least tested and challenged when you made them.

That takes the pressure off being certain and puts it on being honest about what you actually know and don’t know. Which is where research comes in.

Research lowers the cost of being wrong

People think research slows a decision down. It's the opposite. Research lowers the cost of being wrong. It's the difference between finding out in a week that an idea doesn't land and finding out after two quarters of engineering and a launch built on top of it.

And it doesn't have to be big. Five or six real conversations with the right users, a rough prototype people can react to, a small test before anyone commits. Enough to turn "I think" into "we checked."

One habit to start with

If you want one habit, try this. Before you commit to a big call, ask the room: what would we need to see to know this is wrong? Then go find that, cheaply, before you build.

Instead of nodding along to the loudest person, the team agrees on what evidence would settle it, then goes and gets it. The decision stops being a contest of who's most sure and becomes a question anyone can help answer. That's not slower. It just catches the expensive mistakes while they're still cheap.

What this looks like in practice

I see a version of this on a lot of engagements. Two leaders are certain about opposite directions, and the argument gets settled by conviction, not evidence. The times it goes well are the times we stop and put the disagreement in front of real users. It's striking how fast a real reaction takes the heat out of the room. The debate stops being about who's more senior and starts being about what we just watched someone do.

Strong opinions are useful, and someone still has to make the call. The shift is simple: stop trying to be the most certain person in the room, and start trying to make a decision you'd still respect a year from now, whether it worked or not.