The 22 fields we track for a textile passport

Twenty two attributes, and not one of them is a statutory textile passport requirement today, because the delegated act that would set those has not been adopted. The set is still worth having. The evidence problems underneath it are real now, and most of them will not change when the act lands.

Sources as at: 28 August 2026

LinkedIn X Email

On this page

Governed, tracked, anticipated
What the columns ask
The attribute set
What evidence each needs
The five classes
How much work this is
How many passports
Reading a row
Sources

What is governed, what we track and what is still anticipation

Three different things get called textile passport requirements and only the first of them is law.

Already governed, and applying now

Some of what a passport would carry is already regulated, independently of any passport. Fibre composition has a closed legal vocabulary of fibre names and a duty to make composition available before purchase. Articles containing a substance on the REACH Candidate List above the threshold carry a notification duty. Producer registration for textiles is live in some markets already. None of that waits on a delegated act and none of it is new.

What ActivateDigital tracks

Twenty two fields, chosen from the framework, from the instruments that already apply to textiles and from what a passport has to hold to be useful to somebody who is not the brand. That is the table below. It is what we carry, not what anyone is obliged to carry.

One thing about the number in the heading, because a reader should not have to work it out. Twenty two is this business's own tracking set and not a statutory count, since nothing in force names a textile passport field list. The heading says what we carry. If an act lands with a different list, the heading changes with it.

What is still anticipation

The Ecodesign for Sustainable Products Regulation creates the passport, the unique product identifier and the data carrier. It is in force and it builds the machinery. It does not say what a pair of trousers has to carry, and the delegated act that would has not been adopted. The Commission's working plan names textiles as a first priority with an indicative adoption year, which is an indication in a work plan rather than a schedule.

So when the act arrives, some of the twenty two will be wrong. Saying so costs nothing and it changes how the tables below should be read.

Where footwear sits

Footwear falls on both sides of this line, which is why a business selling trainers usually cannot get a straight answer. It is outside the textile scope of the working plan, which indicates a separate study for footwear rather than a product act, so footwear is not on the textile route to a passport act at all. It is inside the scope of the delegated act prohibiting the destruction of unsold goods, which is already in application and whose scope names apparel, clothing accessories and footwear.

The practical consequence is the opposite of what most trainers sellers expect. They are already reached by a duty under this framework, and they are not inside the scope of the one they are usually asked about.

Why the set is useful before then

Because the hard part was never the list. Composition, weight and origin are difficult to evidence today, for reasons that have nothing to do with which act names them. A brand that fixes those has done work that survives the delegated act whatever it says. A brand that waits for the list will start from nothing on the day it lands.

For what is settled, what is adopted and what is merely signalled, including the indicative year for textiles and the study indicated for footwear, the status page is the only page here that asserts a date.

attributes we track for a textile passport

Our own tracking set rather than a statutory count. No delegated act sets a textile passport field list today, so the number in the heading says what we carry and not what anybody is obliged to carry.

SourceEcodesign for Sustainable Products Regulation (EU) 2024/1781, consolidated textCommission working plan COM(2025) 187Checked 28 August 2026

Five questions, asked of every attribute

Two of the five questions are about how hard an attribute is. Three are about what kind of evidence it takes, and those three are the ones that decide what a business should actually do.

Difficulty and size sensitivity are separate, deliberately

Difficulty is about the field. Water impact is hard because the input nobody holds is the geography where the fibre was grown, and a company with a billion euros of revenue is in exactly the same position as a company with two hundred thousand. Size sensitivity is the part of the difficulty that is caused by being small, and there are four mechanisms behind it: supplier leverage, a fixed cost spread over fewer units, not having a product information system and not employing a specialist.

Collapsing those two into one severity score turns a problem the whole industry has into one that looks like it belongs to small companies. Three of the five hardest fields here carry no size sensitivity at all.

The three questions that decide evidence strategy

Neither difficulty nor size sensitivity tells anybody what evidence an attribute needs. Three questions do. Can a laboratory acting on the finished product contradict the declared value. Can the fact change while the product itself does not. How far back does the evidence have to come from. Those are the second table, and they sort the list into groups that no difficulty score puts together.

The list is split across two tables rather than run out to nine columns, because a table nobody can read on a phone is a worse artefact than two that fit. The rows are the same twenty two, in the same order, with the same numbering.

What the size sensitivity column is and is not

It is our inference from implementation work and from published research, not a measurement of small companies. Nothing here says whether any particular business can comply. It classifies fields.

Two cells are held back. Each of them would read as a commercial claim about a specific regulatory instrument, so they wait for review rather than publishing on the strength of one worked example.

The attribute set, and how hard each one is

The resolution column is our own measurement on a 34 variant test catalogue. It is a test catalogue and not a customer, and it says what the engine reached without asking the merchant anything.

The attribute set, with intrinsic difficulty and size sensitivity held apart. Resolution figures are from a 34 variant test catalogue.
Field	What it asks for	Class	Difficulty	Size sensitivity	Resolved
1. Product identifier	A unique identifier, in practice a GS1 GTIN, carried into a Digital Link address and permanent once it is printed.	Catalogue hygiene	Low	Medium	24
2. Model, style and variant	The product as distinct from the variant, which decides how many passports exist.	Catalogue hygiene	Low	Low	34
3. Economic operator identity	Who is placing the product on the market, corroborated where a public register allows it.	Registration and admin	Low	Medium	34
4. Operator contact and address	A contact route for the responsible operator.	Registration and admin	None	Low	34
5. Commodity code	A Combined Nomenclature classification, which is republished every year.	Catalogue hygiene	Medium	Medium	20
6. Fibre composition	Percentages of fibre mass against the legal vocabulary of fibre names, never forced to 100.	Supplier dependent	Medium	Medium	31
7. Recycled content	A mass fraction of recycled input over the fibre mass of the whole product.	Supplier dependent	High	High	4
8. Country of origin	A declared non-preferential origin, recorded rather than adjudicated.	Supplier dependent	Medium	Medium	20
9. Product weight	Net product mass, being the mass of the goods without any packaging.	Catalogue hygiene	Low	Low	31
10. Conformity declarations	Declarations of conformity where an instrument requires one, which for ordinary apparel is usually nothing.	Registration and admin	Medium	Low	0
11. Carbon footprint	A modelled figure from a governed calculator, recorded and republished unaltered.	Method and model	Medium	Medium	18
12. Water impact	A scarcity weighted figure under ISO 14046. Not litres.	Method and model	Blocked	None	18
13. Durability coefficient	A dimensionless multiplier from a French regulatory formula over business inputs. It contains no test result.	Method and model	Medium	Held	34
14. Microfibre shedding	A semi quantitative supplement under the French official methodology. Not milligrams per wash.	Method and model	Low	None	24
15. Production facilities	A facility identifier the merchant names. Never one we searched for.	Supplier dependent	Blocked	High	0
16. Substances of concern	A disclosure about substances of concern above 0.1 per cent by weight.	Supplier dependent	Blocked	High	0
17. Recyclability route	A route or a score, on a method that has no authority behind it yet.	No governed basis	Blocked	None	17
18. SCIP reference	A reference to a SCIP notification, where the duty arises.	No governed basis	Blocked	Low	0
19. Repair route	Where and how the product can be repaired.	No governed basis	Unknown	Held	0
20. Care instructions	Care across five dimensions, read from the merchant's own words. No symbol is emitted.	Supplier dependent	Medium	Low	20
21. Safety certificates	A certificate number a scheme's own register recognises. Having none is lawful.	Supplier dependent	Medium	High	0
22. EPR registration	A producer registration per market where textile extended producer responsibility applies.	Registration and admin	High	High	0

Two labels need a word. Blocked in the difficulty column means a structural block: the obstacle is somebody else's infrastructure rather than anything the brand can work harder at. Held in the size sensitivity column means the value is not published while it waits for review, and there are two of those.

Seventeen of the twenty two rows have no page of their own, and the table names them rather than linking nowhere, because a link to a page nobody has written is a promise. The five that do have a page are listed after the next table, once and nowhere else on this page.

What evidence each attribute actually needs

Three attributes this list treats alike behave in three incompatible ways, and a difficulty score puts all three in roughly the same place. Origin is conferred at making up, so records covering the spinning and the weaving describe steps the rule says do not decide it. Composition is fixed at the mill and stays legible in the finished garment, which is checkable against the result of analysis by a method the regulation prescribes. Recycled content originates several steps upstream of anybody a brand has a contract with, no laboratory test at scale separates recycled from virgin origin in a finished product, and an unbroken record is the only mechanism that exists.

That last proposition reaches us through a summary of the preparatory study rather than from the study's own file, and the page says so rather than restating it as ours. The row is not softened for it. There is no shorter route to that attribute and nobody has one to sell.

The same twenty two attributes, with a status for each, whether a laboratory acting on the finished product can contradict the declared value, whether the fact can change while the product does not and how far back the evidence has to come from.
Field	Status	Can a laboratory acting on the finished product contradict it	Can it change while the product does not	How far back the evidence has to come from
1. Product identifier	Anticipated	No. A string carries no physical property for a test to reach.	No, if it is doing its job. Reassigning one breaks every passport already printed.	Your own catalogue.
2. Model, style and variant	Our model	No. It is a decision about a range rather than a property of cloth.	Yes. A merchandising change alters it while the goods stay identical.	Your own catalogue.
3. Economic operator identity	Study proposal	No. Nothing in a garment records who placed it on the market.	Yes. A corporate change alters it and the goods do not move.	Your own records, corroborated at a public register where one exists.
4. Operator contact and address	Our model	No. There is nothing on the product to test.	Yes. An office move changes it and nothing else does.	Your own records.
5. Commodity code	Binding today	Partly. A test settles the facts a classification turns on and it does not settle the classification.	Yes, on a known cycle. The nomenclature is republished every year and a code can move while the garment does not.	The product itself, read against the current nomenclature.
6. Fibre composition	Binding today	Yes. The regulation provides for a stated composition to be checked against the result of analysis by a prescribed method.	No. It is fixed at the mill when the fibres are blended and it stays fixed.	The finished garment. No upstream relationship is needed to establish it.
7. Recycled content	Our model	No. No test at scale separates recycled from virgin origin in a finished product, which we hold second hand.	Yes. A certificate lapses or a scope narrows and the garment is untouched.	Every step from recovery to the garment, unbroken. One break ends the claim and there is no shorter route.
8. Country of origin	Binding today	No. Nothing in the garment records where it was assembled.	Yes. Move the making up and the same specification produces a different answer.	The making-up site and the records held there. Deeper records describe steps the rule says do not confer origin.
9. Product weight	Our model	Yes. A set of scales settles it.	No, once the product is finished.	Your own stockroom.
10. Conformity declarations	Our model	Partly. A test can contradict the property declared. It cannot contradict whether a declaration was drawn up.	Yes. A declaration can be withdrawn while the goods sit in a warehouse.	Whoever drew the declaration up.
11. Carbon footprint	Our model	No. It is a modelled figure and there is nothing in the garment to test it against.	Yes. Rerun the same calculator on a revised dataset and the number moves.	The calculator's own inputs, wherever those sit.
12. Water impact	Our model	No. A scarcity weighted figure is a model rather than a measurement of the object.	Yes. The weighting depends on where the fibre was grown and the model behind it is revised.	The growing region, which is the input nobody holds.
13. Durability coefficient	Our model	No. The formula takes business inputs and contains no test result, so there is nothing on the product to contradict.	Yes. Change a warranty term and the coefficient moves.	Your own commercial terms.
14. Microfibre shedding	Our model	Partly. Shedding itself can be tested. The semi quantitative supplement published under the national methodology is not that test.	Yes. The methodology behind it is revised.	The fabric specification.
15. Production facilities	Study proposal	No. A garment does not record the building it was sewn in.	Yes. Production moves between orders while the specification does not.	The making-up site, named by the merchant.
16. Substances of concern	Our model	Yes, for the substance. A test on the finished product can find something a disclosure does not mention.	Yes. The list a disclosure is measured against is added to and the garment is unchanged.	The formulator, which is usually several steps upstream of the mill.
17. Recyclability route	Our model	No. A route is a claim about infrastructure and infrastructure is not in the garment.	Yes. It depends on what collectors in a market will take.	The recycler, and no method with authority behind it joins one to a garment yet.
18. SCIP reference	Our model	No. The database holds no product identifier and it suppresses the identity of the submitter.	Yes. A notification can be made or superseded after the goods are made.	Whoever made the notification. It can be recorded and it cannot be discovered.
19. Repair route	Our model	No. A repair address is not a property of cloth.	Yes. A repairer closes and the route is wrong with nothing done to the product.	The merchant. There is nothing upstream to ask.
20. Care instructions	Our model	Partly. A test establishes what the fabric tolerates. It does not settle what the brand chose to advise.	Yes. A brand can revise its own advice at any point.	The merchant's own words. No symbol is emitted.
21. Safety certificates	Our model	Partly. A test can contradict the property certified. It cannot contradict whether a certificate exists.	Yes. Certificates expire and scopes are withdrawn while the goods do not move.	The certifying body's own register.
22. EPR registration	Our model	No. Registration is an act with an authority rather than a property of the goods.	Yes. A market opens a scheme and the same garment now needs one.	The register itself, per market.

What the four statuses mean

The status is about the attribute as a published field, not about whether anybody has a duty somewhere else. Several rows marked as ours sit on top of duties that are perfectly real, and the row's own words say so.

Binding today Three rows. An instrument in force already requires the fact to be given for textiles, independently of any passport: composition, the commodity code and customs origin.
Study proposal Two rows. Named among the identification concepts of the preparatory study informing the textile decision. That study is pre-decisional by design, we hold it at second hand and it is never cited here as law.
Anticipated One row. The framework itself creates it and no act has set it for textiles. That row is the product identifier, because the framework creates the passport, the identifier and the carrier and then stops.
Our model Sixteen rows. In the set because we track it. Nothing we have read makes it a textile field.

Sixteen out of twenty two carrying nothing but our own judgement is an uncomfortable ratio to publish and it is the accurate one. It is also the reason the second table matters more than the first: if most of the list is a working set rather than a legal one, the useful question about any row is what evidence it would take to defend it, which is a question the delegated act does not change.

Two columns do most of the work. The laboratory column tells you whether a dispute can be settled without anybody upstream cooperating. The reach column tells you where to stop asking, and the reasoning behind it, attribute by attribute, is set out on how far back each attribute's evidence has to reach.

The rows that have a page of their own

These are the fields with enough substance underneath them to carry a page each. One page per field, at its own address, listed here once and linked from nowhere else on this page, so that the tables above stay a list of attributes rather than a list of promises.

What the five classes actually mean

The class says what kind of problem a field is, which is more useful than how hard it is, because it says who can do something about it.

Catalogue hygiene

The value is already in the business, or it is one weighing session away. The failure mode is a field holding the wrong kind of thing, and the most common instance is a merchant's internal stock code written into the barcode field. This is the cheapest useful work and it is the work that nothing else moves without, because several computed fields need weight and composition before they can produce anything at all.

Supplier dependent

The fact exists and it belongs to a mill, a factory, an agent or a certification scheme. This is the largest class, seven of the twenty two, and it holds the three heaviest evidence burdens. Software can ask better, record the silence and generate a statutory request. It cannot create the fact, and this is the class where a vendor is most tempted to imply otherwise.

Method and model

Somebody else's method computes it. The calculators are public and free, so producing a number is cheaper than it looks. The difficulty is knowing what the number is not, and a modelled figure read as a measured one is the real risk here.

Registration and admin

A duty discharged with an authority rather than a data problem. Recurring rather than one-off. Software can tell you which registers to look at and record the answer. It cannot register you. This is the most under-discussed real cost in the set, because it is a duty that already exists rather than one that is coming.

No governed basis

There is nothing to comply with yet, or nothing to look up. Effort spent here buys nothing, and the correct behaviour is to wait and record why. Three fields sit here and each one is waiting on somebody other than the brand.

How much work this actually is

Twenty two attributes across 34 variants is 748 cells. On that test catalogue 363 of them resolve without asking the merchant anything, and 24 distinct questions cover the rest.

That ratio, rather than a percentage, is the honest shape of the work. One answer often covers many rows: which markets do you sell into is asked once and lands across a catalogue. Four questions are cold asks that nothing could have answered from data, being the factory, the fabric supplier, the repair address and the markets.

Seven of the twenty two attributes resolve for nobody at all on that catalogue, and each of the seven is waiting on evidence that does not publicly exist rather than on a better engine. Conformity declarations, production facilities, substances of concern, the SCIP reference, the repair route, safety certificates and EPR registration.

A note on the denominator, because it matters. Thirty four variants is not thirty four products. A composition is a property of a product rather than of a size, so five sizes of one shirt are one composition question and five passports.

How many passports a range actually needs

The most consequential unanswered question in this set is not which fields are required. It is what a passport is attached to.

If it attaches to a model, one style in a full size and colour run is one set of product facts and one passport. If it attaches to an item, the same style is hundreds. Nobody outside the delegated act can settle that, and it moves the cost of a passport programme by more than any field on either table.

The positions on record differ. EURATEX asked in March 2026 for information at model level unless operators voluntarily provide more granular data. The Small Business Standards study argues for flexible granularity across model, variation, batch and item, from a survey of 18 respondents. The delegated act has decided neither, and until it does, any figure for how many passports a range needs is a working assumption rather than an answer.

How to read a row that resolves for nobody

A zero in the resolution column is not a failure to look. Each of those fields has a recorded reason, and the reasons are different in kind.

For the SCIP reference the reason is structural. The database is positive only, holds no product identifier and legally suppresses the identity of the submitter, so a reference can be recorded and cannot be discovered. For production facilities the reason is access, not structure: the records exist and confirming one anonymously and free of charge is not possible. For substances of concern the reason is that the evidence does not exist until a supplier writes it down, and there is one statutory route to making it exist.

Those three sentences are three different absences, and the evidence page sets out the vocabulary that keeps them apart. It is worth reading before either table above, if either table is going to be used to plan anything.

You might want to read next

Since you have read this, these may answer the questions that usually come next.

Sources

Ecodesign for Sustainable Products Regulation (EU) 2024/1781, consolidated text

Arts. 2(27), 7, 9 to 13, Annex I, Annex IIICELEX 02024R1781-20240628In forceRelevant provisions reviewed

Creates the passport, the unique product identifier and the data carrier. It sets no textile field requirement. It is the whole basis of the one attribute on this page whose status is anticipated rather than ours, and the articles on passport content were not read verbatim by this build, so nothing here is attributed to a numbered provision.

View official source
Commission working plan COM(2025) 187

IndicativeRelevant provisions reviewed

Names textiles as a first priority and gives an indicative adoption year. It is also the source of the footwear boundary on this page, because footwear sits outside its textile scope with a study indicated rather than a product act. The year and the study indication are both on the status page.

View official source
Regulation (EU) No 1007/2011 on textile fibre names and related labelling, consolidated text

Art. 5(1), Art. 16(1), Annex ICELEX 02011R1007-20180215In forceRelevant provisions reviewed

The instrument behind the composition row, and one of the few that already applies to textiles today. It is also what puts composition in the yes column for a laboratory acting on the finished product, because it provides for a stated composition to be checked against the result of analysis by a prescribed method. That analysis provision was confirmed and not read in full by us, so this page states that a prescribed method exists, quotes none of it and states no tolerance figure.

View official source
Combined Nomenclature, annual amending regulation to Annex I of (EEC) 2658/87

In forceRelevant provisions reviewed

Republished every year, which is why the commodity code row ages on a known cycle and why that row sits in the yes column for a fact that can change while the product does not.

View official source
EURATEX position on Digital Product Passport granularity, March 2026

Industry positionRelevant provisions reviewed

Quoted and attributed as an industry position. Evidence of market behaviour rather than authority on the law.

View official source
Small Business Standards study on Digital Product Passport implementation

Independent studyRelevant provisions reviewed

Quoted with its sample size of 18 respondents in the sentence, because a third party finding is never restated as ours.

View official source
Joint Research Centre, Study on DPP content for textile apparel products under ESPR, 13 May 2026

ProposedReached through a secondary reproduction, primary text not read

The preparatory study informing the textile decision. Two things are taken from it here and both are held at second hand: that no laboratory test at scale separates recycled from virgin origin in a finished product, and that its identification concepts include a unique product identifier, a unique operator identifier and a unique facility identifier. Parts of the study were read at its own address in an earlier pass and this row is deliberately recorded at the weaker of the two depths. It is pre-decisional by design and it is never cited here as law. Two propositions once attributed to it were withdrawn against the source registry rather than kept, and neither appears on this page.

View official source
Commission guidance on non-preferential rules of origin for textiles and garments

In forceRelevant provisions reviewed

Read for one rule, which decides the origin row in the second table: for garments, origin is conferred by complete making up, meaning everything following cutting or knitting to shape, and spinning, weaving and dyeing do not confer it. It is a customs origin statement and it is never used here to answer an origin claim made to a shopper. This build did not establish the guidance's own address, so the source is named without a link.
Chain of custody scheme rules, scope and transaction certificate specifications

Published scheme rulesRelevant provisions reviewed

Carried for the recycled content row. The published rules were read, and they are what allows this page to say that the evidence behind that attribute is scheme documentation rather than law, and that a scope can narrow or a certificate lapse while the garment is untouched. The wider scheme documentation behind those rules was not read, so no individual scheme is named here.

Help someone else make sense of product passports.

LinkedIn X Email