One Must-Read Article

The Judgment Audit, Part 1: Survival Is Not Skill

Part one of a two-part series.

If you ask a founder how he knows his judgment is any good, you usually won’t get a direct answer. Instead, he might just gesture around the office. The company is still running. Payroll has been met this month, last month, and for many months before that. Half the original team is gone. Whatever he’s been doing seems to be working.

I’ve heard that answer for twenty-five years. I used to give it myself for a long time.

But that’s not real evidence. It’s just survivorship, which is the weakest kind of proof in business. What follows is the first half of a judgment audit — an honest look at whether you have any evidence at all about the quality of your own decisions.

Your record doesn’t prove what you think it proves

Three things get in the way.

Outcomes are unpredictable. Sometimes good decisions fail, and bad ones work out, and this happens often enough that it’s hard to judge the thinking behind them. Over your whole career, you might only make a few hundred truly important decisions. That’s not nearly enough to tell skill from luck. In any other part of your business, you’d never trust a sample that small. But here, you base your whole self-assessment on it.

Feedback comes late and is often mixed up with other factors. For example, a pricing decision might take eighteen months to show results. By then, the market has changed, competitors have shifted, and maybe you hired a salesperson who made a big difference. The real signal is buried in all that. You can’t separate it out, so you end up giving yourself credit by instinct.

The survivor tells the story. This is the subtle part. You’re not really remembering the decision—you’re piecing it together backwards from the outcome you already know. You remember the reasoning that matches the result. You’re not lying; that’s just how memory works. But it means your memory has edited the story, and every time you review your own decisions, you’re looking at the edited version.

The one I’d have defended

Early in my career, I was the director of electronic publishing at a pharmaceutical publishing company. We had a product that reported on clinical trials for drugs in development presented at medical conferences around the world. It was tough to produce, and as far as we knew, no one else was doing it at that scale.

I had to decide whether to keep the product going. I said yes, and my reason seemed solid: customers told me they wanted it. Pharmaceutical companies were clear and specific about their interest. I asked the market directly, and it answered.

The product continued to succeed—not in a huge way, but it was clear and happened early. That seemed to confirm my decision. I moved on to the next project and counted this as proof of my good instincts.

But the real outcome took a year to show. Gathering trial data from so conferences was hard on the staff and never got any cheaper. Adoption leveled off at a mediocre level. The clients who said they wanted it weren’t willing to pay anywhere near what it cost to produce. We ended up discontinuing the product.

Here’s what I want you to notice: it took me a long time to realize this, and that’s why I’m writing this piece. My reasoning was wrong from the start. When clients say they want something, that’s just an interest statement, but I treated it like a price statement. Those are two different things, and only one decides if a product survives. That mistake was there from the beginning. I could have caught it.

What stopped me from seeing the mistake was the early success. The win came before the real truth, and I accepted it. I kept that decision in my mind as a point in my favor—not the discontinuation, which I blamed on the market, but the launch, which I credited to myself.

That’s the whole problem in a nutshell. I got a good result from bad reasoning, counted it as skill, and by the time the real feedback arrived, I wasn’t paying attention anymore.

Why this stopped being tolerable

For most of the past twenty-five years, you didn’t have to measure your judgment, because that’s not what you were paid for. You were paid for execution, and execution can be measured—cycle time, close rate, gross margin, on-time delivery, churn. If you tracked the numbers, you had a real scoreboard for what set you apart.

But that’s changed, and I’ve spent most of the past year in Carter Reports explaining why. As AI makes execution a commodity, judgment becomes more valuable. The skill you’re now valued for is the only one you’ve never measured.

You have dashboards for everything that’s no longer your advantage. But you have nothing to track the thing that actually sets you apart now.

The judgment audit takes thirty seconds

Before you read on, try this.

Think of the last three big decisions you made—a hire, a market, a price, a partnership, or anything important. Most founders can do this quickly.

Now, for each one, recall what you actually predicted would happen. Not what you hoped for, and not what you’d say now. What specific outcome did you expect, when did you expect it, and how confident were you at the time?

Almost nobody can do this. It’s not because founders are careless, but because no one ever writes these things down. If you never recorded your prediction, there’s nothing to measure. Every judgment you’ve made has gone unscored, so your confidence in your own judgment is built on an empty record.

That’s the first finding of your judgment audit. Next week, I’ll talk about the scoreboard.

The Judgment Audit, Part 2: Building the Scoreboard — next week.