A common, hidden vulnerability affecting the decisions owners make.

I once ran a small study on a group of clients: about 40 business owners, with a mean income of about $700K (25th percentile about $450K, 75th percentile about $900K). This wasn't a random sample of people, it was a rather similar group of small business owners: roughly similar incomes, ambitions, and values. Basically saying, alike enough that if any set of people on earth would give matching answers, it's this one. Why am I going out of my way to point this out? It changes the findings from being interesting to surprising and worth taking a deeper look.

I went in expecting it to be a little funny. My theory at the time was that people diverge in what they assume, more than they assume: that a room which feels like-minded would show maybe 20-30% disagreement on a handful of questions they'd all have sworn were common ground. A mild, entertaining gap, and a reminder to be careful about making assumptions. That was the prediction.

I was calibrated on the wrong axis entirely.

One question asked, in effect: most people would agree on most things. Agree or disagree?

It came back exactly even.

Sit with what that does. A group was asked whether people broadly agree with each other, and split down the middle on the answer. They couldn't agree on whether people agree.

And the split settles itself. The 50/50 is the disagreement, which means the half who answered "yes, people mostly agree" held a read the group itself didn't confirm, demonstrated on the spot by the room they were sitting in. None of them could tell they were in that half. That's not a mild result. It's the signature of a group with no shared read on itself: a collective sense of "how aligned are we" that was, functionally, noise.

But agreement is the surface of it. Set aside whether the group matched, and look at what was actually on the table. Nobody treated these as trivia. Each answer is a belief someone genuinely holds, and beliefs like these are what a person reasons from when they make a call. What just showed up is that a stack of decision-grade beliefs was far less settled, and far less shared, than the people holding them assumed.

Two questions, asked separately. Same trait: asked about yourself, then about everyone else.

Before the numbers land, be clear what "reality" is standing in for here. This isn't philosophy-class reality. It's your market. Your customers. Your numbers. Whether the problem is actually where you think it is. Your family, your health, the economy you're placing bets on. "A grasp on reality" means a grasp on that. So the answers below aren't an abstract curiosity.

The answers don't just diverge. They invert, almost to the point. 82% say they see reality clearly. The same 82% say most other people don't. Flip the subject of the sentence and the result mirrors.

That can't be true as stated. You can't have a group where nearly everyone is the exception. What these two questions demonstrate is the exact blind spot that's invisible from the inside: the strong, default, near-universal sense that my view is the accurate one and other people are the distorted ones. Every person here was running it. Including me, the one asking.

And it wasn't a one-off. Asked whether most people's beliefs are limiting, 71% agreed. Asked whether their own beliefs are limiting, only 24% did. Same shape, a second time: the distortion is always in someone else.

There's a practical edge here that's easy to step past. It isn't only that people are mistaken about others. It's that they don't know they are, so they act on the assumption of agreement. You price, you hire, you negotiate, you partner as if the person across from you reads the situation the way you do. On the things you'd never think to check, a good share of the time they don't. And the gap doesn't announce itself. It just makes the move land differently than you expected.

One question I nearly cut. It asked about wanting to be happy, in three slightly different framings.

Same underlying want. Three framings. The answers swing from unanimous, to a real split, to near-total rejection.

Here's the part that matters, and it isn't about happiness. At the moment each question was answered, every answer felt right: each one judged on its own, in its own moment, each one holding together. Nobody felt a conflict, because you never hold three framings of the same thing side by side in real time; you answer the one in front of you and move on. The contradiction didn't seem to exist.

It only appeared when the answers were pulled out and set next to each other. And once they were, the clients agreed: those answers don't reconcile. Which means the contradiction was there the whole time. It wasn't created by lining the answers up, it was revealed by it. From the inside, judging each in its own moment, it stayed invisible. It took an outside view to surface a conflict that was already present.

And it isn't only soft topics. Asked flat out, the same group said making money is hard (44% agreed) and, reworded, that making money is straightforward (79% agreed). The same underlying question, the wording alone moving where they landed.

Set the individual questions aside and just watch the pattern. A fast pass through six more, each one the kind of thing a like-minded group might assume it broadly agrees on:

Split, split, split. Not on obscure questions. On the sort of thing you'd expect a similar group to land on together.

Now a word directly to you, because if you're the kind of person reading this, these were people a lot like you: similar drive, similar success, similar enough to each other that you'd expect consensus. And you might be a little surprised how divided they came out.

If a thought is forming right about now, something like "that was probably a more mixed group than he's letting on; the people around me would mostly agree," hold there for a second. That is exactly what they thought about their own group. And honestly, when I set this up, I gave it a fair chance it would come back that way too. I was still surprised how divergent it was.

So the objection isn't a way out. It's the position nearly everyone stands in right before the spread proves them wrong.

Here's the part that makes the rest hold up.

On some questions the group was unanimous. 100% agreed they want to be happy. 100% agreed it's worth questioning your own beliefs. 100% agreed that what you believe plays a big role in your outcomes. When something was genuinely universal, it showed up as total agreement.

That's the control. It means the divergence everywhere else wasn't a broken survey or random noise: when there was real consensus, the survey caught it cleanly. So the splits are signal. These owners genuinely didn't share the views they'd have assumed were common ground.

Mixed in were questions that seemed almost silly:

They weren't filler. They were the most useful questions in the survey, for two reasons.

First, they're checkable. Nobody can really say who's right about raising $5M in two weeks, so you could wave that divergence off as "who knows." But baking a cake this weekend isn't debatable. Nearly a third (30%) said they couldn't. When someone answers no to a thing that's plainly doable, they aren't reporting a fact about cake. They've surfaced a belief about their own capability that's demonstrably out of step with reality, and they can't see that it is.

Second, watch what happened when people answered them. Some said things like "I can't, because I can't do X." Some had been trying at something for years without getting it. Some ranked one of these as harder than another thing they'd already called impossible, with no consistent logic to the order.

So I asked a plain question. What does baking a cake actually require? Ingredients, a recipe, and doing it. What does losing body fat require? Calories in, calories out, done consistently over time. Decompose almost any of these and you get the same three things: a known target, a known method, and enough sustained action over enough time.

Which raises the real question.

Strip these tasks down and most of them don't differ in kind. A cake, body fat, a language, a mile, even 10x-ing a business. The same structure sits underneath all of them: a known target, an available method, and required time-under-tension. The information isn't secret. The path is public. What differs is how much sustained action each demands, and whether the person believes the method applies to them.

That second thing is the only place these truly diverge in fundamental nature.

Two owners, identical on paper. Same task, same available method. One believes the path is for him and runs it long enough to get the result. The other doesn't quite believe it, so he never supplies the sustained action, so he gets nothing, which then "confirms" he couldn't. Same starting point, opposite trajectories, and the entire fork happened at a belief, not at the task.

Worse, it compounds. The believer's result strengthens the belief; the non-believer's non-result hardens his. They diverge further every year, from the same starting line.

So when someone says "I can't, because I can't X," they've almost always mislocated the constraint. They're pointing at the task. The constraint is upstream: in whether they believe the method is for them, which is the switch that decides whether the method ever runs at all. The cake question is that mislocation shrunk small enough that you can't argue with it. The business is where the same mislocation runs invisibly, and costs a fortune.

And the business calls sit on the same layer. What the real problem is. What a fix would take. Which way the market's moving. Who to trust and when to move. Those are read off the same stock of beliefs the survey just showed to be divergent, mostly unexamined, and, on the checkable ones, provably off. Every one of those calls has a belief underneath it doing the deciding, usually without anyone noticing it's there.

Take one and run it forward. If an owner believes real capital isn't available to him, he doesn't pursue the idea that would need it. And that belief never gets corrected, because he never creates the evidence that would correct it. It isn't describing his options. It's deciding them.