Judgment as a non-commoditizable asset

Two professionals hand you the same flawless report, critical to your business continuity. One took three weeks; the other took three hours with well-directed AI. If we’ve spent a century billing for effort and effort no longer measures value, what should?

Professional services have been invoicing by the hour for a century, but that proxy is broken. In the legal sector, agentic AI has already cut complaint response times from 16 hours to just 4 minutes — a gain of over 200x ¹. When execution accelerates by that order of magnitude, persisting with hourly billing penalizes efficiency and destroys margins. These aren’t isolated cases: according to Deloitte, 67% of consulting buyers now prefer fixed-price contracts, up from 41% three years ago ². The client is no longer paying for effort. They’re paying for certainty of outcome.

Because value doesn’t disappear — it migrates. From “doing” — execution that is now automatable — to “deciding” — the human judgment that determines what should be produced, in what context, and with what accountability. Not the report or the campaign itself, but the chain of decisions that shaped it. Why this structure. What was discarded and why. That is judgment.

The Harvard and BCG experiment with 758 consultants confirms it: AI-assisted professionals were 25% faster and produced higher-quality results. But in tasks requiring fine contextual synthesis, those who delegated to the AI were 19% less likely to get it right ³. The tool multiplies expert judgment; without it, it multiplies mediocrity.

And here is the paradox. The biggest obstacle to billing for judgment is that thinking is invisible. A 50-page report seems to justify a high fee; a one-sentence recommendation that saves a company is typically met with skepticism about its cost. Those leading this transition resolve it by making the decision process visible: alternatives evaluated, risks mitigated, the logic behind each step. Authorship migrates from the output to the reasoning that produced it.

Steinmetz understood this a century ago. Called in by Ford to fix a generator no one could repair, he listened to the machine for two days, made a chalk mark, and ordered 16 coils removed from the windings at that exact point. When Ford received the $10,000 invoice and asked for a breakdown, Steinmetz replied: “Making a chalk mark: $1. Knowing where to make it: $9,999” ⁴.

AI is today’s chalk marker. The professional who understands this migration doesn’t compete on producing faster — they compete on knowing where to mark. And that knowledge — sediment from thousands of real decisions, mistakes owned, and consequences learned over the course of a career — cannot be replicated with prompts. What AI cannot reproduce is not what you know how to do. It’s what you’ve learned not to do.