The Most Valuable Thing AI Can Do for Your Operation Isn’t Replace People. It’s Remember Them.

A friend sent me Pam Didner’s recent piece in Inc., “Forget Prompting. The Most Important AI Skill of the Next 5 Years Is Something Else.” Her argument stopped me because it lands exactly where I’ve spent my career living: the gap between what your best people know and what your organization can actually reproduce.

Her thesis is simple. The next real AI skill isn’t better prompting or vibe coding. It’s writing a skill file — a one-page document that tells an AI tool how to do a specific task your way, every time. Not a style guide. Not a prompt library. One task, your standards, your format, written down so the work runs the same whether or not the person who mastered it is in the room.

I’d put it more bluntly, from an operator’s chair: a skill file is institutional knowledge that finally stops walking out the door.

The problem I’ve watched play out for years

Every operation has a handful of people who are quietly holding it together. They know how the stakeholder update should read. They know which line item in the proposal always gets questioned. They know the difference between a customer summary you’d send and one you’d never send. That knowledge is real, it’s valuable, and for most companies it lives in exactly one place: someone’s head.

Then that person takes a vacation, gets promoted, or leaves — and the standard drops. Not dramatically. Just a step behind. And usually nobody notices until a client does.

I’ve spent a lot of time building the systems that prevent that quiet slippage: the runbooks, the checklists, the “here’s how we actually do this” documents that turn one person’s judgment into a repeatable process. What Didner is describing is the same instinct, now aimed at AI. You capture how your best people think and operate, you write it down, and the process keeps running.

Why this is an operations skill, not a technical one

The part I want to underline, because it’s the part people miss: you don’t need to be a developer to do this.

Didner’s build process is refreshingly un-technical. Pull three examples of work you’d be proud to send and one you’d never send. Hand them to Claude, ChatGPT, or Copilot. Ask it to write a one-page instruction file from the pattern. Run it on real work, fix what’s off, run it again. Most people have a working draft in an afternoon.

Notice what that actually requires. Not code. Judgment. Knowing what “good” looks like. Knowing when a task holds its shape week to week versus when it changes every time. Knowing which rule you forgot to write down when the output missed. That’s operational thinking — the same muscle that builds a pricing model or a hiring scorecard. The AI is just the place you finally get to store it.

This is why I keep telling teams that the people who will get the most out of AI aren’t the most technical ones. They’re the ones who understand the work deeply enough to explain it.

The line that matters most

Here’s where Didner and I land in exactly the same place, and it’s the whole reason I wanted to write about her piece.

A skill file can’t replace your best people. It can hold what they know, run their process when they’re out, and onboard the next hire faster. But it doesn’t learn — you do, and then you have to teach the file what you learned. Her framing is sharp: a person’s skill grows with use; a skill file’s effectiveness decays without it.

That’s the truth I’ve been repeating in every operations conversation I’ve had this year. AI done right doesn’t thin out your team. It amplifies it. It takes the repetitive, standardizable 60% off your best people’s plates so they can spend their judgment where judgment actually matters. The file makes your AI more useful. Your people make the file better. Neither one replaces the other.

And this is where efficiency stops being a cost story and becomes a growth story. When you give your best people back their time, you don’t just get faster execution — you get room to think. Efficiency creates space for strategy. Strategic people are innovative. Innovative people grow companies. That’s the chain most leaders miss when they treat AI purely as a way to do the same work with fewer hands: the real return isn’t the hours you save, it’s what your people do with the hours you hand back.

Companies that treat AI as a headcount-reduction lever are solving the wrong problem, and they’ll feel it when the quality quietly drifts and there’s no one left who notices. The ones that treat it as a way to preserve and scale their best thinking are the ones who’ll pull ahead.

What I’d tell an operations leader to do Monday

Pick one task worth repeating. Give the skill file one owner — the person who does the task best and will feel it first when the output slips. Treat it like any other operational asset: it lives on the shelf, it gets maintained, and it doesn’t leave when someone resigns. Start with one file that works, then build the library by function.

The judgment that made the file worth writing still lives in a person. That’s not a limitation of the technology. It’s the entire point — and it’s exactly why the smartest operations leaders should be writing these down now.

Inspired by Pam Didner’s “Forget Prompting. The Most Important AI Skill of the Next 5 Years Is Something Else,” Inc., July 2026.