Against the category specialist

Working in one clothing category can give a studio valuable depth. It can also make familiar answers arrive before a new brand's problem has been understood. The risk we would test is whether the studio can explain what makes this collection different, rather than whether its category list is long or short.

01What a specialist knows that nobody can reason out from first principles

The case for the specialist deserves to be put properly first, because it is a real case and it is not weak. Some categories carry safety and labelling requirements that bite on the design itself, on cords, fastenings, small components, fabric behaviour and what has to be printed on a label. Those requirements vary by market and have to be checked for the market being sold into, and a studio that has worked one of those categories for years already knows which of those questions come before the drawing rather than after it. That sequencing is worth a great deal. A requirement discovered after a fit sample has been approved costs the sample and the pattern behind it.

Fit depth is the second real advantage. A specialist has a fit model of one body type, a base block corrected over many seasons, and a memory of every correction that was tried and rejected. Its first sample is often closer than a wider studio’s second. The same is true of its suppliers: a narrow field produces a network that is deep rather than broad, where the quality standard is known without being renegotiated and a difficult construction can be placed with somebody who has already done it. A founder buying a specialist is buying compressed time, and that is a legitimate thing to buy.

There are three further things a specialist holds that a wider studio has to rebuild each time, and they are worth naming because they are the part that is genuinely hard to acquire. The first is a failure library: not the list of solutions that work but the list of ones that were tried and abandoned, and why. Knowing that a particular closure fails in a particular place after a particular kind of wear is knowledge that only comes from having shipped it and had it come back. The second is tolerance intuition, meaning a sense of which measurements wander in production and which stay put, so the tight tolerances get spent where they change the garment and are not scattered evenly across a specification that nobody can then hold. The third is the category’s standard of ordinary: the details a customer of that category does not consciously notice but registers the absence of, which is why work outside a studio’s home category can be technically correct and still read as slightly foreign.

None of that is decoration on the argument. It is the argument the specialist should be allowed to make at full strength, because the counter-case only means anything if it survives it.

02How a reference set stops growing without anybody deciding it should

What is bought alongside it is a settled set of solutions. Every studio designs out of a reference set, meaning the constructions, finishes, proportions and supplier capabilities it can actually reach for. A specialist reference set is drawn entirely from one category, so it converges. The problems repeat, the answers that worked are available, and reuse is what makes the studio efficient. The brand that arrives fourth gets a version of a solution that was designed for somebody else, and it will look like it.

The mechanism is economic rather than lazy, which is why naming it as laziness misses it. Every solved problem lowers the cost of solving it the same way again. The pattern exists, the supplier has made it, the fit has been proved, the specification is written and the risk is known. A studio choosing between a proved answer and an unproved one is choosing between a known cost and an unknown one, and it will keep choosing the known cost for perfectly good reasons until the reference set has hardened into a house style that was never designed. The studio does not notice, because from inside, a solution that works and a solution that is right look identical.

The observable symptom is a proposal that arrives too early. A studio that has answered a question many times will offer the answer before the founder has finished describing the problem, and the first time that happens it feels like competence, because it partly is. The second time it happens, on a different brand, is the moment worth noticing. The other symptom is a portfolio in which the garments could be swapped between the brands without anything looking out of place. That is not a failure of taste. It is the reference set showing through the work of several different clients at once.

The argument for working across categories is not that variety is stimulating. It is that a construction developed under the pressure of one category solves a problem in another, and the transfer is specific rather than atmospheric.

03A solution travels because the second category never had to solve it hard

Sportswear has to design the seam around the movement, because a garment that restricts a shoulder or a stride fails immediately and visibly. That produces habits: a gusset that adds reach without adding width, a sleeve set forward of where the flat drawing suggests, a seam moved off the point where the body bends. Carry those into a coat and the coat lets somebody drive in it. Tailoring also has a long history of solving movement. Comparing those approaches can reveal a useful construction choice rather than assuming either tradition owns the answer.

That last sentence is the whole mechanism, and it generalises. A solution travels not because two categories resemble each other but because one of them was forced to solve a problem the other one only ever had to accommodate. Where the pressure was real, the answer was tested to destruction and refined. Where the pressure was mild, the conventional answer was good enough and was therefore never examined. Transfer moves knowledge from the place it was interrogated to the place it was assumed, which is why it so often looks obvious once it has been done and occurs to nobody before.

Knitwear has to decide what happens at every edge, because a knitted fabric does not simply accept a hem. It answers with rib, with linking, with a fully fashioned edge shaped during knitting instead of cut afterwards. A studio that has spent time there stops treating a jersey neckline as a trim decision made at the end. It arrives with a view about how the edge should behave after twenty washes, a useful durability question to test rather than assume.

Denim brings measurement discipline, because a fabric that shrinks and is then washed forces every dimension to be recorded after wash and not before. Applied anywhere else, that habit catches the specification error that produces a garment correct on the table and wrong on the customer. Childrenswear brings a logic of fastening and dressing sequence, worked out because the wearer cannot be relied on to help. It turns out to be the same logic that makes an adult garment easy to put on with one hand.

Workwear brings the most portable habit of the group, which is putting the reinforcement where the load enters rather than where the wear eventually shows. A pocket can fail at the corners of the opening, where the whole force of a hand going in arrives at a point. Workwear answers at the corner, and the rest of the pocket is left alone. Take that somewhere with no workwear tradition, a bag for instance, and the same reading applies without translation: the panel is rarely the problem, the anchorage of the strap is, and reinforcement spread evenly over a bag is weight added everywhere and strength added without necessarily addressing the likely failure point. A studio without that habit reinforces the thing that looks vulnerable, which is usually the thing that is not carrying anything.

Breadth of this kind is not the same thing as generality. A studio that has passed shallowly through eight categories has eight sets of assumptions and no reference. The transfer only happens where each category was worked deeply enough to leave a real technical memory, and where somebody is looking across them deliberately rather than waiting for the connection to occur.

04A first brand’s hardest problems sit above the category, not inside it

For a first brand the balance favours breadth, because the founder’s hardest problems are rarely category specific. What the brand is for, how much to launch with, what the clothes can cost to make against the price they will sell at, which supplier is the right one, how to hold fit consistent across a group of styles, and what to say about any of it: none of those questions change when the category does. The genuinely category specific parts are narrower than they look. A block, a size chart, a set of technical requirements and a cloth platform can be built per project, and the requirements have to be checked per market in any case.

Those problems are worth separating, because they fail in different ways. Positioning failures show up as a range that cannot be priced: the clothes were designed before anybody decided what they were for, so the construction was chosen first and the shelf it has to sit on was found afterwards. That order cannot be undone by a better sample. Range architecture failures show up as a set of styles in which every piece is trying to be the memorable one, which leaves nothing carrying the volume and nothing carrying the margin, and a founder discovering that their whole range is made of exceptions. Specification failures show up as arguments with a supplier that both sides can win, because the document described an intention rather than a measurable state, and intentions cannot be checked against a garment on a table.

A studio whose experience is entirely one category tends to enter all three of those at the construction end, because that is the end its expertise faces. It will answer a positioning question with a make, and a range question with a style count borrowed from how that category is usually bought. Neither answer is wrong in itself and both are arrived at from below.

The second order effect lands about six months later, at the first extension. A brand set up as its category has to change what it says about itself in order to add anything: the name, the language and the customer’s idea of it were all built around the category, so a second one reads as a departure and has to be explained. A brand set up around a proposition can add a category as a range decision, and nothing has to be re-explained to anybody. Which of those a founder ends up with is decided in the first weeks, in work that has nothing to do with garments, by whoever is asking the questions then.

05Where the specialist is the correct answer, stated plainly

There is a case where the specialist is the better choice, and it should be stated plainly. A brand that is entirely one technically demanding category, that intends to stay there, and that competes on a standard of construction the category already understands, is better served by a studio that has made a thousand of the thing. Knowledge like that cannot be reasoned out from first principles in a single project. That case is real, and it is not the common one, because a founder may still be deciding how far the brand should extend.

The case gets stronger the more the calendar itself belongs to the category. Where testing, component lead times and approvals dominate the schedule, most of the work is knowing the sequence, and a studio that has run the sequence before saves a season rather than a fitting. It gets stronger again when the construction is the product rather than the vehicle for it, so that the thing being sold is a technical achievement a customer can identify. And it applies to a different kind of founder: somebody deepening inside a category they have already proved, rather than deciding what to be. Depth is the right purchase when the question has already been settled.

There is a test that separates the two situations without much argument. Ask what was tried in this category and rejected, and why. A specialist answers immediately and in detail, because the failure library is the asset. A studio that answers with categories it has worked in rather than problems it has been beaten by is not offering depth, whatever else it is offering.

06Whether breadth is just a polite word for shallow

The strongest objection is that breadth is what a studio claims when it cannot claim depth, and it deserves a direct answer rather than a deflection. The objection is frequently correct. A studio listing many categories is easy to assemble and hard to verify, and most of the time the list is a record of what walked through the door rather than evidence of anything. The claim that a construction transfers is also unfalsifiable in the abstract: anybody can assert that a habit from one category informs another and never be shown to be wrong.

What separates the two is that depth is held per problem, not per category. A studio with real depth can describe a specific technical fight, name the thing it got wrong, and say what it does differently as a result. A studio with breadth and no depth describes categories, which is to say it describes the market rather than the work. The second tell is convention: work produced without depth tends to arrive looking like the ordinary version of its category, because in the absence of a tested view the conventional answer is the only one available. Depth shows up as a decision that had to be argued for. Breadth without it shows up as competence with no position.

The cost of breadth should be said as plainly as the cost of specialism was. A wider studio’s first sample in a category it has not lived in is further from finished than a specialist’s would have been, and the first fit round is longer because the block is being established rather than recalled. That is a real price and a founder pays it in time at the start. What is bought with it is that the third brand does not receive the first brand’s answer, and that a problem can be recognised as something other than the category’s usual version of itself. Whether that trade is worth making depends on whether the brand intends to look like its category or like itself.

The useful test of a studio is not how many times it has solved this problem before. It is how many different problems it can recognise this one as.