25 Brands Checked. Three Put the Origin in the Product Data.

Every product page speaks two languages

One is for people: images, description, specification list. The other is for machines — a data sheet behind the page from which search engines, price comparison services and shopping assistants draw their details. It holds name, item number, price, availability, often even the accepted payment methods.

A field for the country of manufacture is provided there too. It has been part of the standard for years and costs one line.

We looked at 25 brands. Three had filled it.

The three that do

Loro Piana records “Italy” in the data sheet of a €1,300 cashmere jumper. Givenchy does the same, also Italy. And Kenzo fills two fields.

All three belong to LVMH. That sounds like a group decision — but it is not. Within the same house, Louis Vuitton, Dior, Céline, Fendi, Berluti, Pucci, Patou and Marc Jacobs leave the same field empty.

So it is not the group that decides but the individual sales platform. Brands that deliver origin machine-readably do so because their shop system provides for it — not because a policy requires it.

It is not quite that simple, though. Tommy Hilfiger and Calvin Klein run on the same platform, right down to the naming of their collapsible panels. At Calvin Klein the country of manufacture is stated; at Tommy Hilfiger the same line sits empty.

More precisely, then: the platform decides whether the statement is possible. Whether it is maintained is decided by the brand.

Kenzo shows why a field name is not enough

At Kenzo two fields sit side by side, and they say different things.

countryOfAssembly: Portugal — where the sewing happens
countryOfOrigin: France — that is the brand

Both are accurate. But read only the second — and it is, after all, called “country of origin” — and you take a garment sewn in Portugal for a French one.

This is not hair-splitting. It is why an origin statement is worth little without the question “origin of what?”. Most other houses lack the distinction not because they reject it, but because they carry no field at all.

And the other 22

Everywhere else the field stays empty — across every price band, country and ownership structure. From the €20 t-shirt to the €4,300 bag.

It stands out most where the visible text says a great deal. Natalino names Naples 58 times in its descriptions, Florence eight times, Okayama four — and stays mute to any automated reading. Asket discloses seventeen production steps with supplier names, and likewise.

How precisely a brand discloses, and whether that disclosure can be queried, are therefore two separate decisions. Neither tells you anything about the other.

The Asket case that is not one

At one further brand the search appeared to strike. Asket does carry origin details machine-readably — even the supplier’s name.

But they sit in Asket’s own data layer, not in the format search engines and comparison services read. There the field is as empty as everywhere else. A comparison service finds nothing here either.

An address is not an origin

At Windsor, a full postal address in Switzerland appears under “Manufacturer”. That looks like an origin statement but is not one.

The European product safety regulation requires a responsible economic operator to be named with an address — who is liable, not where the sewing happened. The two rarely coincide. Anyone taking that address for the place of manufacture is reading a liability notice as proof of origin.

What it means in practice

A service reading brands automatically sees the same at Natalino as at SIR of Sweden, which states nothing at all: nothing.

The disclosure exists; it simply cannot be queried. For the brand it means its openness shows up in no comparison. For buyers it means any filter by origin comes up empty — not because the data is missing, but because it sits in the wrong place.

What we corrected here

This page has been wrong twice.

The first version rested on thirteen brands and called the finding universal. As a statement about the industry that went too far; Loro Piana was the counter-example.

The second version spoke of one exception among 23. That was wrong too: Givenchy and Kenzo had been checked in an earlier round and carry the same fields — we had built the table from the current round without reading our own records against it.

Both admissions sit here, because a page about missing disclosure has to show its own gaps. The result is more precise for it: not “nobody does”, not “one does”, but “three do — and all three from the same group, in which eight others do not”.