Naming Things

For twelve years my agency ran the same check on every brief before we quoted. It caught the projects that would have hurt us. It never had a name, so nobody outside the building could ask for it, repeat it, or pay for it on its own. The check was real. Without a name it was invisible.

By Jordi Buskermolen7 min read
ai-systemsagency-operating
Naming Things

Two consultants say the same true thing to a prospect.

The first: "I will review your brief carefully. I have been doing this for twenty years, I know what to look for."

And the second: "I will run your brief through our nine-point scope check, the one that tells us whether a project is clear, bounded and safe to quote."

Same review. Same twenty years. Possibly the same nine things checked. But the second sentence does work the first one cannot. It can be examined: what are the nine? It can be repeated: they have this nine-point brief check. It can be requested: can you run that scope thing on this one? It can be delegated: Sarah is trained on it. And it can be bought: the review is nine hundred euros. The first sentence can only be believed.

Trust me, versus check this. The name is the difference.

The check we ran for twelve years and never named

At This Page Amsterdam we had that check. Before any quote went out, someone senior read the brief against the same handful of questions, in the same order, every time. Is the decision maker in the room. Is "done" defined. Who supplies the content. What happens to the timeline if the client is late. We had learned every one of those questions the expensive way, and the check caught the projects that would have hurt us.

It had no name. We called it "having a look at the brief". So nobody outside the building could ask for it. A client who had been through it could not describe it to a friend, except as "they were careful". We could not put it on a proposal as a line, because it was not a thing, it was a mood we were trusted to have. And when a new project manager joined, the check lived in two people's heads and was taught by sitting next to them.

I did not think of this as a loss at the time. It is only since I started naming things that I can see what it cost.

The instruments I run now have names. RecoLog is the monthly measurement of what AI assistants answer when someone asks who to hire. SCREEN is how I score a lead against the firms I actually want to work with. The visibility check that someone in South London ran on their own site a few weeks ago is a fixed set of seven tests with a name on the page. None of those methods is smarter than the brief check was. Every one of them travels further, because a name is a handle, and a handle is what lets other people pick a thing up.

What a name actually does

Five things, and they are mechanical rather than cosmetic.

A name makes expertise inspectable. Vague expertise asks for faith. A named method invites the question "what is in it?", and answering that question is the sales conversation, on your terrain. The prospect who asks what the nine categories are is already engaging with your thinking. The one who hears "twenty years of experience" has nothing to grab.

It also makes quality reproducible, and therefore claimable. "I check briefs carefully" describes a mood that varies with your sleep. "The method checks these nine things, in this order, every time" describes a machine, and machines can be promised. It is the same move that makes a team delegable and an AI tool trustworthy: judgment extracted into fixed structure. The name is the handle on the structure.

Named methods travel without you. People cannot repeat your expertise, but they can repeat a name. "You should talk to Jordi, he is really good" is warm and dies in one hop. "They have this scope check that catches project risk before you quote" survives dinner tables and Slack threads and being retold badly, because the name carries the concept intact through messengers who do not understand it.

Then there is price. Vague expertise gets priced by time, because time is the only visible unit. A named method is a thing, and things get priced by value. "The audit is nine hundred euros" is a normal sentence. "Nine hundred euros of my looking at it" is a negotiation.

And a name claims territory, with people and, increasingly, with machines. Publish a defined, findable method and you become its source. The AI systems answering "how do I check a brief for scope risk" surface defined, named, crawlable things far more readily than diffuse good advice. I measure this every month, and the pattern holds: an unnamed insight is common knowledge, a named method has an author.

You have more of them than you think

The raw material is not invented. It is extracted, and the test is simple: anything you do the same way every time is a method wearing casual clothes.

Start with your repeated checks, the things you always look for in a brief, a candidate, a codebase, before your conscious mind engages. Your repeated questions, the five you ask every new client, which is an intake method. Your decision rules, the gut call you have made a hundred times, which has criteria you have never listed. Your sequences, the order you do things in because you learned what happens in other orders. And your scars, systematized: every "never again" became a check you now run. The check is the method, the scar is its origin story, and together they are the most credible package expertise comes in.

The extraction is one slow session per method. Do the thing once more, narrating, and write down what you actually checked or decided as a numbered list. Most real methods land at five to nine items, not because of magic numbers but because that is roughly how much structure one judgment has. Twenty-three items means you have merged three methods. Two means you have found a principle, which is fine, and not yet nameable.

Then name it plainly. What it examines plus the count, when there is a count. The natural shape, when the concept has one. The process, when the doing is the identity. Not acronym gymnastics, where letters are tortured until they spell a word and the audience feels the torture. Not grandiosity, where the name promises what the method will be blamed for. And never a name before the extraction, which is the emptiest move in the genre: a beautiful label wrapped around three bullet points.

The line between a method and a costume

Everyone has seen the bad version. The internet is full of named frameworks that are three obvious points in a trench coat, and their existence is the strongest objection to this whole piece. So, four tests, and I run them on my own before anything gets a name.

Do you actually use it, yourself, on real work, when nobody is watching? A method you only mention in marketing is a costume.

Can it say no? A real check sometimes returns "not ready" or "do not proceed". A framework that blesses whatever enters it is a deposit slip.

Could a competent colleague run it from the written version and reach roughly your conclusion? If only you can operate it, you have named your intuition. That is fine, but do not sell it as transferable.

Does every item earn its place, or did the count come first? Padding a real six into a rounder eight is how good methods acquire the smell of bad ones.

Meet the four and the name is load-bearing. Miss them and the name is a liability, because named things get examined, which was the whole point, and an examined gimmick costs more credibility than vague expertise ever did. Which is also why not everything needs a name. Two or three methods you would defend in a room of skeptics, used constantly, build a practice's identity. Fourteen build a parody of one.

One more thing a real method turns out to be, and it is the reason I care about this more than I did at the agency. A fixed set of checks, in a fixed order, producing a fixed output, is a specification. It is exactly what turns expertise into software. Every custom AI tool I have built that worked had a named method at its core. Every one I have seen fail had a vibe.

The brief check at This Page would have made a good tool. It would have made a good line on a proposal, a good onboarding document, a good thing for a client to tell a friend about. It was real for twelve years and did its job every week. It just never had a name, so the rest of the world never found out it existed.

Want more of this?

I write regularly on LinkedIn about what I'm building and learning: agency growth, AI development, product judgment, and the messy reality behind making things work.

Follow on LinkedIn