I (Andrew) train at a gym on Flinders Lane called Maddox. It's in the old Customs House and the entrance is discreet enough that you'd walk past it.

Inside: stainless steel, dark walls, original Baltic pine floors with a patina nobody could fake. Designer furniture — not equipment, furniture. Aesop in the showers. A custom scent. Playlists carefully chosen to set a mood rather than fill a silence, and lighting set the way a great hotel sets lighting. The space has a cocoon-like quality I'd never encountered in a gym and don't expect to again.

It gets written about as an interiors project, which it is. That's also the least useful way to look at it.

Three years in, the part that sits front of mind is that nobody has ever shouted a count at me.

All of it came from somewhere. Annie assembled the people who made the experience — the scent, the sound, the light, the materials — and Davey brings the same attention to the programmes and to managing his clients across years rather than blocks of ten. Those aren't two different disciplines. They're the same decision made twice.

Nothing here behaves like an asset. It reads like a vocation, and that's the part that shows.

Hecker Guthrie did the interiors and The Company You Keep did the brand. I've had nothing to do with any of it, which makes the mechanics easier to see.

Maddox has been copied. Anything that works gets copied, and the copies are assembled from the same material every time.

Whatever was photographable. Whatever got the fanfare. Whatever you could glean by standing in the room for an hour and paying attention.

So the concrete gets copied. The polished stainless steel. The lighting. The restraint in the copywriting. Reproduced thoughtfully, sometimes with more money behind it than the original had.

What opens is a gym with a different interior finish.

It's parity with a variation in trim, and the trim was the only part anybody controlled.

It's the shape of the car. Not the engine.

You can buy a custom scent. You can't buy the thought that a gym should smell like anything at all. You can put Aesop in the showers without ever having decided that the ten minutes after a session are worth designing.

The engine is a set of positions, and most of them can't be seen. Whether there's a lock-in contract. Whether the programme is built for the next quarter or the next thirty years. Whether anyone remembers what you were working on back in March.

Copy the expression and all you have is a sentence in a language you don't speak.

Everyone says brand architecture. Almost nobody means it.

If you did mean it, the sequence would be obvious. No building starts at the cladding. You work out what the thing is for, then how people move through it, then what holds it up. The outer face comes last. It's the expression of everything decided before it.

The façade still matters enormously. It's the introduction, and a good one is why anybody walks in at all.

But it's a claim about the inside, and the inside is where people spend their time. Where the light falls. Where the doors are. The decisions that determine whether a place is any good to be in, none of which can be revised later without taking the whole thing apart.

Copy a façade and you have a façade. You're now committed to a front that promises an interior you haven't built.

Clients often arrive wanting the visible answer first. A reference, a mood, a look that already exists somewhere else. Studios are perfectly capable of providing it.

Sometimes the brief is the problem. Sometimes it's the fee. More often everyone in the room knows exactly what's happening and takes the work anyway.

It's easy to talk about the difference between design and decoration when you're looking at someone else's building. Harder when a client is paying you to decorate theirs, and the alternative is asking questions nobody budgeted for, or telling them the thing they asked for isn't the thing they need.

Visual work goes hollow when it's asked to compensate for decisions nobody has made.

That's harder to sell, and we don't always manage it.

There's a version of every good chair that costs a tenth as much. Most are visibly worse.

The good ones aren't. Close enough that nobody would pick it in a photograph, and most people couldn't pick it in a room.

People who buy the real one aren't paying for how it looks. They're paying for provenance. There's a record of where the object came from and what was refused along the way, and that record is the part nobody can manufacture. It's most of what the money is for.

Take it away and you have the same silhouette with nothing behind it.

Brands work the same way. One assembled from other people's conclusions is a replica. It photographs well. It holds up until somebody asks it a question it wasn't designed to answer — a second site, or an offer that doesn't fit the existing shape — and then there's nothing underneath to consult, because the reasoning was written afterwards to justify decisions already made.

Which is also why polish doesn't rescue a hollow brand. It accelerates the exposure.

Polish raises expectation. That's what it's for. It tells people this has been considered and the price is the price for a reason. If the substance is there, the gap closes on arrival. If it isn't, the better the work, the further the fall.

An ordinary-looking business with an ordinary product is at least consistent. Nobody feels misled. A beautifully designed one with nothing underneath has spent real money making the disappointment sharper.

Subcultures worked this out a long time ago and they police it harder than any market does.

You can buy the jacket, learn the references, use the vocabulary correctly in front of people who'd notice if you didn't. The visual layer has always been available to anyone with money and an afternoon.

And the temptation is obvious. Building the real thing takes years. It means making decisions before you know whether they're right, being wrong in public, developing taste the slow way. None of that scales.

A copy lets you skip the expensive part and arrive at the photograph. Which is a rational proposition if the appearance of having arrived is all you're trying to buy.

The problem is that the people who'd know can tell within about four minutes, and almost none of them can explain how.

Not because the imitation is inaccurate — it's often extremely accurate — but because the visual layer was the residue of the thing rather than the thing.

The poser isn't rejected for bad taste. They're rejected for taking the signal without paying for it.

That instinct doesn't switch off when people become customers. Something reads as assembled rather than arrived at, and they file it accordingly.

None of this is an argument against the visual layer. It's ours. Good businesses fail with bad brand work, and mediocre ones outrun better competitors on the strength of it.

For a while.

It's an argument about sequence.

A brand is the culmination of everything someone experiences of a company. The product, the room, the people, the thing handed to you before you asked for it. The visual layer is how all of that gets introduced. It isn't what any of it is made of.

A brand that knows what it is can be designed, because there's something to translate.

A brand that doesn't can only be decorated.

It won't hold when somebody leans on it.