What Actually Breaks Made-to-Measure Consistency Across Stores

Three measurement sheets landed on my desk in the same week last spring, all from the same ten-door account, submitted by three different stores. I sorted them into the right pile without reading the store code printed in the corner of each one, and got all three right on the first try.

Handwriting had nothing to do with it. Store B's chest-to-waist drop has run about a centimetre tighter than everyone else's for two years, and Store D's shoulder point sits a centimetre wider than the shoulder point on any sheet Store A or Store C has ever sent me. Ask either manager and they will tell you they are following the same measurement form as everyone else in the group, and they are. Their hands are just not landing on the same centimetre when they reach the shoulder bone.

You do not catch that by looking at one sheet in isolation. Each sheet is internally consistent and sits inside every tolerance the form checks for. The pattern only shows once you have forty sheets from ten doors in the same folder, and by then several of them have already become finished jackets.

Why the factory gets the blame

Most of these conversations start from the same sentence: a client in one city should get the same suit as a client in another. Said like that, it sounds like a manufacturing promise, so when two suits from two stores do not match, the manufacturer is where the account looks first.

Almost nobody says the alternative out loud, because "our stores measure inconsistently" is a harder sentence to bring to a buying meeting than "the factory has a quality problem." One puts the fix on people you do not manage directly. The other means retraining your own payroll.

The part of this chain that is most controlled is the part everyone suspects first. Our cloth is European or Australian by origin. The cutting and the sewing happen in one place, our own atelier in Shanghai, under our own tailors, on every order regardless of door. What we cannot control from our side is what happens before an order reaches us: who took the measurements, on what day, following habits nobody wrote down.

Ten stores means ten separate sets of hands, trained at ten different moments by ten different people, each one confident they are doing it the standard way.

What forty-seven fields catch

Our measurement form runs to forty-seven fields, and you can see how our fit and sizing system works rather than take my word for it. Multiply forty-seven fields by ten stores and what you are running is ten sets of habits, each wearing the same paper form.

Several of those fields check each other. A sleeve length that does not sit within a plausible range of the arm measurement gets flagged before it becomes a cutting instruction, and so does a chest-to-waist drop too tight to belong to the body described on the rest of the sheet. That catches the obvious mistakes, the transposed digit, the field left blank and guessed at.

What it cannot catch is a whole store leaning the same direction on the same field, the way Store B does on the waist and Store D does on the shoulder. A systematic bias does not look wrong from inside that one store. It looks like a correct, repeatable measurement, because it is repeatable; it just repeats the wrong number every time. The cross-checks built into the form catch a measurement that contradicts itself. They were never built to catch one that is wrong the same way, consistently, for two years running.

One store measured twice

This is the part of the conversation I would not skip past, even though it is the least exciting slide in any rollout plan.

Before a group adds a third door to a shared programme, let alone a tenth, I would want proof that the first door can measure the same client the same way twice. Take your best person, have them measure a willing colleague, write it down, and repeat the measurement three weeks later without looking at the first sheet. If the numbers hold, you have a repeatable method worth training everyone else against. If they do not, you have found the problem this piece is about, on one store, before it multiplied across nine more.

Adding doors does not average this out. I would expect the opposite. A group that cannot make one store repeat itself has no realistic path to making ten stores agree with each other, because scale does not smooth over an inconsistent habit. It gives the habit more places to live.

I would say the same about the review, not only the measuring. If nobody on your team is willing to read every sheet line by line before it goes to production, the way I did with those three, this should not launch across a network yet. That job does not do itself, and the store that submitted the sheet cannot do it for you.

Pooling cloth is a brand decision

The order minimum on any single piece is one suit, whether it is a new door's first order or your oldest door's two-hundredth. Because that floor never moves, there is no volume threshold ten stores need to clear together to earn a better price, which is the assumption most people bring into this conversation and the first one worth dropping.

So if pooling ten stores' orders onto four to six cloths does nothing for price, what is it for? Making ten doors read as one brand instead of ten local ideas of what "the programme" is. Leave the full range open to every manager and you are running ten small businesses under one sign. Our range holds 347 style options across jacket, trouser, waistcoat and shirt, 159 in the jacket alone, and a manager free to choose from all of them will choose differently from the manager two cities over. A shortlist agreed once at the centre closes that gap. Price was never going to.

The shortlist itself should not lean on the word "European." I used that word myself a few paragraphs ago, and it is about as checkable as it sounds, which is not very. So put something harder next to it. Ask for the bunch reference behind each cloth on the list, not just the country, and put that reference on the paperwork every store receives, so a manager in one city and a manager in another are looking at the same numbered swatch book rather than two different cloths that both happen to answer to "navy wool."

Weight decides the rest of it. Of the 822 cloths in our range with a stated weight, 597 sit between 240 and 300 grams, the band most business suiting belongs in regardless of city or climate. Vitale Barberis Canonico's Four Season, a 280 gram cloth, works across ten storefronts because nobody has to think about the season before selling it. Go under 240 grams for something worn from a morning meeting into an evening one and the jacket creases before the second appointment, which is not a complaint nine doors want explaining back to the centre.

Price belongs in the same conversation, after construction, and the two get inverted more often than not. A Classic Fused two piece starts at 108 euro ex-works. Built full canvas instead, the same silhouette starts at 254 euro ex-works. Do not let ten stores set a shared retail price against those figures without agreeing which construction the group is selling. A customer who compares notes with a friend from another door, and finds she paid the same price for the cheaper build, will not blame the store. She will blame the brand, which by then is you.

A pooled season order once it ships

Pooling the fabric choice does not pool the delivery date, and treating "the group order" as a single line on a spreadsheet is where a lot of programmes lose control of their own timeline.

Each store still submits its own measurements on its own schedule, and each set starts its own clock the day it arrives complete, not the day the season programme was announced. Our production window is 14 to 20 working days from a confirmed order, so a store that submits three weeks into the programme starts its own 14 to 20 days three weeks late, and pooling the fabric changes none of that. A ten-door programme run this way is managing ten delivery dates in parallel, and whoever owns the programme should track it that way from the first week rather than discover it when a store asks where its suits are.

Reordering mid-season is where I would slow a group down. If Store G sells through its first batch in week six and wants forty more of the same cloth, I would not promise the second batch is cut from the same bolt as the first, and I would be careful of any supplier, us included, who tells you it is without qualifying that. A mill runs its own production in its own batches. What is realistic to promise, and worth getting in writing, is the same bunch reference, weight and composition on both orders, not an identical shade under two different stores' lighting. Put that distinction in front of the group before week six.

Responsibility in writing

Most of what comes back wrong on a multi-door programme is a measurement problem wearing a manufacturing complaint, and nobody finds out which one it was until somebody asks the right question in writing.

Before you place a first pooled order, get a plain answer from your supplier, ours included, covering three things: what counts as their error versus a measurement your store supplied, who pays for the remake, and who covers shipping while it happens. We wrote up the fuller list of questions worth asking any manufacturer before you commit. The answers matter more across ten doors than one, because a policy left vague for a single boutique becomes a different argument in every city at scale.

A promise that a supplier will make things right sounds reassuring on a sales call and means nothing when Store F is holding a jacket with a shoulder that does not sit correctly and needs to know whether that cost lands on the factory or on whoever filled out the form. Get the line in writing, store by store if you have to, before that jacket exists.

The same discipline that keeps one store's economics honest keeps the whole programme honest. The single number worth building a made-to-measure model on is gross profit per order, not a percentage, and landed cost is still calculated door by door regardless of how centrally you buy the cloth. Freight, duty and payment fees follow the store's address, not the group's. A shared shortlist makes ten doors look like one brand; it does not make their landed costs the same number.

None of this is a case against growing past one door. It is a case for getting the sequence right. Prove the method at the first store, write the shortlist and the fabric references down, put the responsibility policy in writing, and only then let a ninth and tenth door join a programme that already works. Skip the sequence and scale will not create consistency. It will show you, faster and in more cities, where it was missing.

If you are dressing one team for a single event rather than running a season programme across a network of doors, the planning problem is different, and we wrote about what breaks a corporate group order separately.

---

Tell us how many doors you are coordinating and where the measuring is happening today, and we will walk through what a shared programme should look like before you commit a season's fabric order to it. Get in touch here.