It was the morning of December 23rd. I was sitting with a CMO at a global engineering company when an internal email landed in his inbox: a new product, a preliminary name attached. We both read it and looked up at the same moment. The name was wrong.

An anatomical brain cut out of paper with torn orange and teal edges on a grey halftone background.
Fifteen years of naming logic. Until that morning it existed in exactly one place.

Not slightly wrong. Wrong in the way that normally costs three rounds of agency feedback and one irritated CEO. And the company shut down for the holidays the next day.

So we improvised. I spent 40 minutes pulling the CMO’s naming logic out of his head and into a prompt: the structure behind every product name already in the portfolio, the rules that make a name belong there, the words this brand would never use. Fifteen years of judgement, written down in the time it takes to drink a coffee.

Then we asked the AI. Two hours of back and forth produced three strong candidates. By late afternoon, the CEO had picked one.

It worked. And that’s where the problem started, not ended.

Solved in the short term, but not in the long term

Everything about that session was unrepeatable.

The CMO’s naming logic existed in the real world for two hours, in one chat window, because I happened to be in the room with 40 minutes to spare to capture it.

40

Forty minutes was all it took to get fifteen years of naming logic out of one head and into a prompt. It only happened because someone was in the room to ask.

If that email had landed at a different time, handled by a product manager and an intern, the outcome would have been drastically different. Not because they’re less capable. Because nobody would have handed them the fifteen years.

That’s when the real work became obvious. We’d proven the CMO’s judgement could travel. We’d also proven it only traveled because I was there to carry it.

So we decided to package judgement like that. This is the core of what we built at BBN AI Studio: we turn a senior person’s judgement into a ‘skill’ with a defined input, a measurable output, a quality bar a senior would defend, and a named author who reviews it every quarter. His naming logic went in. So did our own naming methodology, our workflow design, and everything we know about getting an AI model to hold a brand’s structural rules under pressure. That combination became one skill, sitting inside the AI his team already opens every morning.

The next time a mediocre preliminary name lands in an inbox, nobody has to improvise, and nobody has to call me for some AI-generated drafts.

That’s the AI Studio product. It’s also the part I find hardest. I spent twenty years becoming the person CMOs call directly. Watching a version of that run a few weeks later without me in the room was strange.

Handing it over

Expert judgement only counts once you hand it over. When it stays in one head, it’s a personal advantage. Once it’s written down and running in someone else’s environment, it becomes an asset the company owns.

Expert brains stop being something you rent by the hour.

We do it with our own expertise first. Twenty years of senior B2B marketing judgement, written down, signed, and then calibrated to a client’s brand, market, and terminology. Our competency is what makes a skill work. Their context is what makes it theirs.

Then we get out of the way. It keeps working whether we’re in the room or not.

That’s the trade. Expert brains stop being something you rent by the hour.

What this doesn’t solve

I won’t tell you this approach fixes all of your team’s AI problems, because I don’t know your team’s AI problems yet.

What I can tell you is what usually stalls B2B marketing teams is not a shortage of tools. Most CMOs I meet already have more AI access than they use. What they’re missing is a way to get their own judgement out of their heads and into the place where the work happens.

Ask where your thinking currently lives.

If the honest answer is ‘in my head, and in the edits I make on Friday,’ your brand scales at the speed of one person’s review cycle. No tool can move that ceiling.

Your brand scales at the speed of one person’s review cycle.

Start here

My suggestion is to pick one decision your team makes repeatedly without you in the room. Write down, in three sentences, how you’d make it: what you look for, what you value, what you’d never do.

That’s your first judgement asset. Put it somewhere your team can reach it on a day you’re not there.

Mine came out of a bad product name and a holiday deadline. Yours can come out of a quiet Tuesday.

Hand it over while you still have the time to do it properly.