# How to Build a Textile Digital Product Passport From Just a Handful of Inputs

Source: https://activatedigital.ai/knowledge/guides/the-data-you-already-have
Last verified: 5 October 2026
Summary: ActivateDigital builds the passport. You confirm the few facts only you can know.

## In 60 seconds

You do not need to build a textile Digital Product Passport from zero. Give ActivateDigital the product data and evidence you already have. We do the hard work of building the passport: connecting product identity, finding usable information, resolving evidence and calculating what the available inputs support across our 22-field textile DPP readiness model. You confirm the few facts that only you or your supplier can know.

- In current testing, up to **~71%** of product-data decisions were settled before merchant input. That is the upper observed record, not a typical-store forecast.
- The initial queue averaged 5.1 product-specific confirmations per product, with shared company identity handled separately.
- The test covered 10 textile products and 34 variants from one Shopify test shop, in a report dated 3 October 2026.
- The 22-field model is not a final statutory textile field list. The Commission describes textile DPP requirements as being defined through a future delegated act.
- The test did not measure completed merchant journeys, elapsed time or legal compliance.
No 22-field form. No need to turn your catalogue into a completed passport before you begin. Start with your existing records and move towards one governed passport, with the evidence behind it kept intact.

## From existing data to a governed passport

| Result | What it measures |
|---|---|
| Up to ~71% | Product-data decisions settled before merchant input in current testing |
| ~5 | Initial product-specific confirmations per product on average in the test flow |
| 22 | Textile DPP readiness fields covered by ActivateDigital's model |
| 1 | Governed product passport to publish once required checks are met |

The percentage includes resolved values and method-specific not-applicable outcomes; it varies with the starting record. The question count excludes shared company identity and is measured before answers, so follow-ups can arise. See how we counted the test.

**Your starting point is what you already have. Your job is to confirm what only you can know.**

[Build a textile passport from existing product data](https://activatedigital.ai/textile-digital-product-passport-software).

## See what happens to one product record

**One product. Useful facts, a clear evidence trail and specific decisions. Not 22 empty boxes.**

This fictional navy hoodie shows a starting record becoming useful without pretending every field is settled. It's an **illustrative example**, separate from the measured catalogue test above, not a captured engine run or a compliance assessment.

### Start with the records already in the business

- **Product:** zip hoodie, style HX-01, navy, size M, batch B-2609.
- **Catalogue:** SKU HX-01-NAV-M, title and variant options. No GTIN supplied.
- **Technical pack:** the same style, variant and batch; unlined construction; net garment mass of 480 g. Evidence ILL-TECH-008, dated 8 September 2026.
- **Supplier declaration:** body fabric stated as 80% cotton / 20% polyester. Evidence ILL-DECL-014, dated 10 September 2026.
- **Test report:** body fabric measured as 65% cotton / 35% polyester for a sample from the same variant and batch. Evidence ILL-LAB-027, dated 18 September 2026.
- **Care statement:** wash at 30°C, with no drying instruction supplied. Evidence ILL-CARE-009, dated 9 September 2026.
The composition comparison assumes the product, component, method, mass basis and tolerance checks are done. A different batch, method or component would need a different conclusion.

### See what the information supports

| What this example contains | What the record should say |
|---|---|
| The catalogue contains SKU HX-01-NAV-M and navy/M options | Keep the starting identity. The SKU does not become a GTIN because it was imported |
| The catalogue and technical pack refer to the same style, variant and batch | Keep the matched product association and the source supporting it |
| The technical pack's 480 g net mass converts to 0.480 kg | Keep the input and the method: divide grams by 1,000. This is a unit conversion, not a new measurement or an impact calculation |
| The care statement supports “wash at 30°C”, but not a complete care set | Retain the supported instruction without inventing the missing detail |
| Two comparable evidence records give different body-fabric compositions | Keep both values and both evidence references. The composition needs review |
| Nothing supplied establishes recycled content | Keep it unknown. Do not substitute zero or infer recycled origin from the fibre blend |
| The technical pack specifies an unlined garment | A separate lining-composition entry is not applicable to this construction. That says nothing about other fields or legal obligations |

The matched association here is supported by the example records. It does not mean an independent certification of the garment. The lining “not applicable” is also different from the environmental-method applicability outcomes counted in the earlier test.

### Focus attention where it changes the result

The next work is specific: investigate the body-fabric disagreement, complete care information where the intended output requires it and obtain recycled-content evidence before making a recycled-content claim. There's no reason to retype the product identity or ask a supplier to recreate every fact already supported.

**The result is not “100% complete”. It is a record you can act on without losing what you know or hiding what you do not.** This example is not presented as ready to publish: the unresolved composition still needs the appropriate decision.

See [how the two-source conflict is handled](https://activatedigital.ai/knowledge/evidence/conflicting-values#conflict-example), including the evidence retained and the review required. For your own textile products, [build from your existing data](https://activatedigital.ai/textile-digital-product-passport-software).

## How do you build a textile Digital Product Passport?

Connect the product records and evidence you already hold. ActivateDigital resolves the passport, brings you the open questions and prepares the governed record for publication once the required checks are met.

The work starts with your product, not a regulation-shaped questionnaire. A store record can supply identity and variant information. A composition declaration can unlock material-dependent work. A supplier document can provide a fact that was never on the public product page.

ActivateDigital builds from what you already have, then asks for the facts that still need you or your supplier.

### 1. Connect what you already have

Start with Shopify, an existing catalogue export, product data from a PIM or ERP, or supporting files. A WooCommerce export, CSV, tech pack, supplier file, certificate, barcode or product page can be starting material through the available connection or supported input route. You do not need to retype everything into 22 boxes.

### 2. Resolve the product

ActivateDigital works out what your information supports and calculates results where the inputs and method allow. The aim is a product passport with a basis for its values, not simply a document filled with plausible text.

### 3. Keep the evidence and any disagreements

Each fact needs to belong to the right product, variant or component. A declaration remains tied to what it covers. Where credible sources disagree, that disagreement stays visible rather than being silently overwritten.

### 4. Answer only the genuine gaps

The engine asks for the specific confirmation, declaration or evidence item still needed. That might be a responsible business detail, a recycled-content declaration or a supplier-held fact. It is a focused question because the resolution work has already happened around it.

### 5. Publish the governed passport

Address the required publishing checks and publish a record that keeps its evidence, declared facts, calculated results and explicit unknowns distinguishable. When the product or its evidence changes, there is a controlled record to update rather than another document to rebuild from scratch.

**Connect → Build → Activate.**

## What is actually left for you to answer?

The remaining work is concentrated in facts an external resolver should not guess. ActivateDigital does the research and resolution; you or your supplier confirm the private facts behind the product.

### What does the merchant still need to provide?

Provide only the relevant confirmations and supporting evidence that the existing record has not already established. In the current test flow, the initial questions averaged **5.1 product-specific confirmations per product**, with shared company identity handled separately. That is a handful of focused questions, not another passport-sized research project.

### Product facts

These can include country of manufacture or origin, a recycled-content declaration, the repair route, substance declarations or care instructions where they are unavailable. The exact questions depend on what your records already answer.

For example, confirming origin means supplying the manufacturing fact and its basis, not researching every passport field. The [country-of-origin guide](https://activatedigital.ai/knowledge/fields/country-of-origin) explains what supports that answer.

### Business facts, often answered once

Responsible operator details, an EORI where relevant and target markets or registrations can apply across several products. Reuse them where the business, role and scope are the same. You should not have to repeat one company fact for every size in a range.

### Supplier evidence, only when needed

A factory or facility relationship, relevant certificate, chemical declaration or recycling evidence may be held upstream. Some of it may already be in your files. For chemical information, the question is [whether the evidence reaches the finished garment](https://activatedigital.ai/knowledge/evidence/chemical-data).

When something is missing, [request the missing evidence from a supplier](https://activatedigital.ai/knowledge/evidence/supplier-evidence). Ask for the particular declaration or record, not another complete questionnaire.

**These are confirmations of private facts and targeted evidence requests. They are not 22 research tasks.** Some answers may need a supplier document or lead to a follow-up; the question count is the initial queue, not a fixed limit.

## What data do you need for a textile Digital Product Passport?

Start with product identity and the product information and evidence you already hold. Composition, origin, weight, care and supplier documents are useful wherever they are available. You do not need to finish collecting every field before ActivateDigital can begin.

Look across the business, not just the public product page.

| What you already have | What it can contribute to the passport |
|---|---|
| Store or catalogue records | Product identity, variants and populated material, origin, classification or weight information |
| Spreadsheets, PIM or ERP records | Supplier references, business details and product relationships |
| Tech packs, specifications or care labels | Material, component, construction and care information |
| Certificates, declarations or test reports | Evidence supporting a particular product or material claim |

A tech pack can do a different job from a sales description. See [what a tech pack can and cannot establish](https://activatedigital.ai/knowledge/digital-product-passport/textiles/tech-pack-dpp-readiness). Composition must also retain its fibre and component scope; the specialist guide explains [how fibre composition is interpreted](https://activatedigital.ai/knowledge/fields/fibre-composition).

ActivateDigital brings the information into its **22-field textile DPP readiness model**. Use [the 22-attribute textile readiness guide](https://activatedigital.ai/knowledge/fields) for the detailed definitions.

The model is not a final statutory textile field list. The European Commission describes textile-specific DPP requirements as being defined through a future delegated act. Read [what textile DPP requirements are established today](https://activatedigital.ai/knowledge/digital-product-passport/textiles) and [what a delegated act decides](https://activatedigital.ai/knowledge/regulation/what-a-delegated-act-decides) for that separate legal question.

Before you start, you can [measure what you already hold against what a passport needs](https://activatedigital.ai/dpp-readiness-checker). It runs in your browser, on your own data, and nothing is uploaded.

## Can you create a textile DPP from Shopify?

Yes. ActivateDigital can use your Shopify catalogue as the starting point for a textile Digital Product Passport. Connect the product data you already use to sell, then let the resolver do the passport work before you answer the remaining questions.

Your product and variant records can contribute identifiers, relationships and populated customs information. Composition or care may already sit in a description or metafield. Supporting evidence can add what the store record does not hold.

The engine checks what those values mean. A SKU is your internal reference; a barcode value needs checking before it is used as a GTIN. The information available depends on the populated fields and the connection or export, not simply on whether Shopify has a field with that name.

For the field-by-field platform detail, see [which Shopify fields a passport can use](https://activatedigital.ai/knowledge/guides/shopify). To begin the commercial journey, [start from your Shopify store](https://activatedigital.ai/shopify-digital-product-passport).

## Why is this more than AI autofill?

Autofill starts with a box and tries to put something in it. ActivateDigital starts with the product and establishes what can support the answer.

That means resolving identity, matching evidence to the right product, applying suitable calculations and preserving the basis for each result. It also means recognising when a source cannot answer the question, or when two credible sources disagree.

The difference is where the work happens. You are not being asked to find every value, interpret every document and rebuild the same company information across a catalogue. The engine handles the resolution work and brings you the decisions that need your knowledge.

A certificate is useful for the claim it supports. Read [what certificates, declarations and test reports establish](https://activatedigital.ai/knowledge/evidence/documents). A modelled figure also keeps its method and inputs: [what a modelled environmental figure actually establishes](https://activatedigital.ai/knowledge/evidence/environmental-figures) is different from a laboratory measurement.

## How much of a textile DPP can be resolved automatically?

Existing data can carry substantial passport work before merchant input. In the current controlled textile testing, the upper observed result was **about 71% of scoped decisions settled from the starting record**, including resolved values and method-specific not-applicable outcomes.

That is evidence of what a richer starting record can enable, not a universal completion percentage. A record with usable identity and material information gives the engine more to work with than a product title alone. The passport also has product, variant and business facts, so the measured decisions are not simply 22 equally sized boxes.

The more useful customer question is: **what still needs me?** Across the test flow, the initial product-specific queue averaged about five questions. The engine's task is to resolve what the evidence supports; your task is to confirm the facts it should not invent.

A percentage is not a measure of merchant effort. One company answer can apply across products, while a supplier-held declaration may need a specific request. So [what drives a textile passport programme's cost](https://activatedigital.ai/knowledge/guides/cost) is a separate question from the number of fields in a model.

## What happens when product data is missing?

ActivateDigital keeps a genuinely missing fact visible and asks for the next useful action. It does not silently substitute zero, “none” or a guessed answer.

A fact missing from a public search may already be in your records, and one missing from the business may need a supplier or specialist. That access boundary is in [which facts a small business can verify itself](https://activatedigital.ai/knowledge/evidence/small-business-product-fact-observability). Then [decide who can close a genuinely missing fact](https://activatedigital.ai/knowledge/guides/missing-product-information-small-business).

That distinction prevents unnecessary work and false certainty. For example, [what a missing SCIP reference does and does not mean](https://activatedigital.ai/knowledge/fields/scip-reference) is different from a conclusion about whether a notification duty applies.

## What happens when two sources disagree?

ActivateDigital keeps the disagreement visible rather than choosing a convenient answer, so it can be addressed with evidence.

Two different composition declarations should not become an invented third composition. A test result should not be overwritten just because a later supplier file says something else. See [what to do when a test and a declaration disagree](https://activatedigital.ai/knowledge/evidence/conflicting-values).

This is part of keeping the passport trustworthy, not a reason to send you back to the beginning. The evidence guide explains [how evidence states and unknowns are recorded](https://activatedigital.ai/knowledge/evidence).

## Do all textile DPP fields have to be known before publication?

Not necessarily. The required checks must be complete, while permitted optional gaps or supported “I don't know” answers can stay clearly stated in the record.

This lets the passport distinguish established facts from information still being sought. It does not give permission to omit information required by an applicable law. Read [when an unknown can be stated and when it cannot](https://activatedigital.ai/knowledge/evidence/stating-what-you-do-not-know) for that distinction.

The goal is a controlled, publishable record under the applicable checks, not a page that looks complete because unknowns have been disguised.

## What does the finished passport look like?

The output brings product identity, established values and supporting evidence into one readable record. It also shows where a result is calculated, where it rests on a declaration and where information remains unknown.

You can [read an annotated textile passport](https://activatedigital.ai/knowledge/passport) to see how the record fits together. Its invented product details make the structure easy to follow; it is an illustration, separate from the measured test.

The value continues after publication. When a material, supplier declaration or calculation input changes, the affected fact can be reviewed against its evidence. You have a record to govern, not just a document to file away.

## Four checks make existing product data reusable

Existing information becomes more useful when its meaning stays attached to it. Four checks help the same record serve a passport and other business destinations.

| Check | What needs to stay clear |
|---|---|
| Object | Whether the information describes the finished product, a component or a facility |
| Level | Whether it applies to the model, a variant, a batch or an individual item |
| Source | Who stated, measured or calculated it, and where the supporting record lives |
| Coverage | Which product, process, period and claim the evidence covers |

Keep units and denominators with the value. A fibre-mass percentage is not automatically a whole-product percentage. A shipment weight is not necessarily the net weight of the garment.

For the hierarchy, see [how textile facts belong to product, variant, batch or item](https://activatedigital.ai/knowledge/digital-product-passport/textiles/product-variant-batch-item-data). For a changing classification, see [when the commodity code changes](https://activatedigital.ai/knowledge/regulation/when-the-commodity-code-changes). These distinctions help you preserve useful work rather than repeat it.

Choosing the software to do this? [Test it with a representative product, not a perfect demo](https://activatedigital.ai/knowledge/guides/choose-digital-product-passport-software#representative-pilot). Check the data it establishes, the evidence it keeps and the work still left with your team.

## Before you reuse a value from another EU system, ask six more questions

The four questions above decide whether a value is usable at all. Reuse across a regulatory system adds another layer. A value can be true, well evidenced and current and still be the wrong value to copy into a DPP because the target system is asking a different legal question.

Add six checks before you move it:

- **Same definition?** `Origin`, `producer`, `weight` and even `product` can mean different things in different rules.
- **Same unit or method?** A number produced under one regulated method is not interchangeable with a same-named number calculated another way.
- **Same legal role?** The company acting as supplier in one system may be manufacturer, importer, producer or another actor in the next.
- **Same jurisdiction and version?** Registrations, nomenclatures and sector rules can be Member-State, year or product-group specific.
- **Same access class?** Authority-only technical data does not become public just because the value is technically available to a passport system.
- **Same lifecycle?** Ask what makes each record stale and which system must be updated when the product, evidence or rule changes.
The result is a ten-question gate rather than a new score. **Same fact and same label are not the same thing.** The cross-system matrix for DPP, SCIP, EPREL, EUDR, EPR, Safety Gate, PPWR, customs and technical documentation is on [Product Data Across EU Systems](https://activatedigital.ai/knowledge/product-data/eu-product-data-systems).

## What the Registry holds is separate

A published passport and the EU Registry are different records, and neither replaces the governed product data behind them. For what the Registry holds, see [what the EU DPP Registry records about you](https://activatedigital.ai/knowledge/evidence/what-the-registry-records).

## What Product Data Should Businesses Prepare Now for Digital Product Passports?

The textile example shows one use of existing product data. The wider preparation principles below apply across categories, while the relevant product rules determine what each passport must contain.

## A passport is a governed projection, not a second product master

A Digital Product Passport should project the product information and evidence the business already controls, plus new information required by the applicable product rule. It should not slowly diverge from Shopify, ERP, PIM, PLM or the technical file.

Reuse is real; zero new work is not. Establish the product and variant identity, how your internal IDs map to each other, authoritative facts, evidence, change owner and missing data before treating the passport screen as the system of record.

Publishing is not the end. Continue to [what a product code can do after the sale](https://activatedigital.ai/knowledge/guides/product-code-after-sale-services).

Businesses do not need to choose between **doing nothing until every delegated act is final** and **hard-coding an imagined universal DPP checklist today**. A safer approach is to separate preparation into three buckets:

## PREPARE

Low-regret data capabilities that are repeatedly useful across adopted DPP regimes, current product law and future implementation.

## WATCH

Information areas with credible official development signals but unresolved product-specific requirements.

## What product data should you prepare across categories?

Assumptions that current evidence does not establish as universal DPP requirements.

This is a **readiness model**, not a universal statutory field list.

All categories: [Digital Product Passport Requirements by Product Category](https://activatedigital.ai/knowledge/digital-product-passport).

### The readiness map

| Information or capability | Action now | Why |
|---|---|---|
| Product identity and hierarchy | **PREPARE** | UPI and granularity recur, while exact model/batch/item rules vary. |
| Manufacturer / responsible operator master data | **PREPARE** | Actor identity recurs across adopted DPPs and existing product law. |
| Product classification / commodity coding | **PREPARE where applicable** | Appears in adopted passport and customs/Registry contexts. |
| Existing legally required composition / ingredients | **PREPARE** | Already regulated data are low-regret to structure, provided they are not mislabelled as universal DPP fields. |
| Conformity evidence and document references | **PREPARE** | Adopted regimes use conformity/legal references and businesses need stable evidence lineage. |
| Evidence provenance | **PREPARE** | A value without source, scope and date becomes expensive to defend or update. |
| Versioning and change history | **PREPARE** | DPP information must remain controlled as products and evidence change. |
| Machine-readable data and export capability | **PREPARE** | Interoperability is a horizontal system requirement. |
| Access classification | **PREPARE** | DPPs are not universally public. |
| Textile production-site information | **WATCH** | Strong preparatory signal, not final textile law. |
| Textile recycled/environmental/durability information | **WATCH** | Official development evidence, but final selection/methods remain open. |
| Iron/steel, tyre and aluminium product-specific fields | **WATCH** | Delegated acts are pending. |
| DPP service-provider rules | **WATCH** | Horizontal delegated action remains pending. |
| Universal GTIN | **DO NOT HARD-CODE** | UPI is the legal concept; GTIN is not universal. |
| Universal GS1 Digital Link | **DO NOT HARD-CODE** | Useful implementation option, not universal EU mandate. |
| Universal QR | **DO NOT HARD-CODE** | Data carrier is the horizontal concept; QR is sector-specific in some regimes. |
| Universal item-level serialisation | **DO NOT HARD-CODE** | Adopted regimes use model, product type and hybrid architectures. |
| Universal factory/carbon/recycled/supply-chain fields | **DO NOT HARD-CODE** | Product data are category-specific. |

### PREPARE: product identity and hierarchy

Know how your systems represent:

- product family
- model/style
- variant
- batch/lot
- individual serial/item.
Do not assume the final DPP will use all of those levels. The value is in preserving their relationships.

ESPR lets product-specific rules choose the relevant granularity, while adopted regimes already differ: Battery is hybrid, detergents and toys are model-level and construction uses product type.

So a flat SKU table is a weak long-term foundation.

See [Digital Product Passport Granularity: Model, Batch or Item?](https://activatedigital.ai/knowledge/guides/granularity).

### PREPARE: manufacturer and responsible-operator data

Keep legal actor data controlled rather than buried in free text.

Useful fields include:

- legal entity name
- legal role
- registered/contact address
- country
- operator identifier where applicable
- valid-from / valid-to dates
- evidence or authoritative source.
Do not collapse manufacturer, importer, authorised representative, responsible person and “brand” into one generic actor field. The applicable product law can assign different responsibilities to each.

### PREPARE: classifications you already need

Where your products already use commodity, customs or regulated category codes, keep them controlled and versioned.

Detergent and Toy DPP regimes include commodity-code concepts where applicable, and Registry/customs workflows make classification strategically relevant.

The advice is not “invent a code for every DPP”. It is “do not leave existing regulatory classification trapped in spreadsheets with no ownership or history”.

### PREPARE: product information already required by current law

This is one of the safest areas to improve now.

Examples include:

- textile fibre composition under existing textile law
- ingredient and allergen information under food law
- cosmetics PIF/notification and label information
- tyre label/product-database information
- product-safety identity information.
These are **Required under other law** where their regimes apply. They are not automatically DPP fields.

Structuring them pays off if a later DPP law reuses an existing regulated fact: the organisation already has a governed source.

The legal label still matters. Do not rename adjacent regulation “DPP compliance” before the DPP instrument says so.

### PREPARE: conformity evidence and stable document references

A DPP can point to or carry conformity-related information. The adopted detergent, toy and construction frameworks show this.

Make declarations, test reports, certificates and technical documents addressable by stable metadata:

- document ID
- document type
- version
- date
- issuer
- product/model/batch scope
- legal basis
- expiry/review trigger if relevant
- source location
- confidentiality/access classification.
The product fact and the document that proves it should not be the same database field.

### PREPARE: evidence provenance

For every important product claim, be able to answer:

- where did this value come from?
- who supplied or measured it?
- what exactly does the evidence prove?
- which product, batch, site or time period does it cover?
- when was it last checked?
- what happens if a second source disagrees?
This is a product-evidence question, not the same thing as the regulatory methodology used to decide what DPP law says.

Retain [Passport Evidence: How We Know, and What a Blank Means](https://activatedigital.ai/knowledge/evidence) as the canonical product-evidence destination, with [When a Test Contradicts a Supplier Declaration](https://activatedigital.ai/knowledge/evidence/conflicting-values) for source conflict.

### PREPARE: machine-readable data

Critical product truth should not exist only in:

- PDFs
- email attachments
- supplier free-text notes
- image scans
- disconnected spreadsheets.
The horizontal DPP architecture now has concrete interoperability standards covering data exchange, identifiers, carriers, storage, APIs and system interoperability.

A business does not need the final textile field list to know that machine-readable, versioned product data will be easier to activate than PDF-only records.

### PREPARE: access classification and ownership

Not all DPP information is public.

Add metadata that can support:

- public/consumer
- authority
- customs
- notified body
- legitimate-interest or controlled partner access
- internal/restricted evidence.
Do not guess the final product-specific permission matrix. Build the data so permissions can change without duplicating the source of truth.

See [Who Can See What in a Digital Product Passport?](https://activatedigital.ai/knowledge/passport/who-sees-what).

## Which category-specific evidence should you watch?

The Watch bucket exists for areas where there is enough official evidence to justify preparation, but not enough to freeze the legal schema.

### Textile factories and production sites

Official textile preparatory work considers production-site and processing-location information.

Good preparation:

- map factories and facilities internally
- give facilities stable identifiers
- retain product/batch-to-site relationships
- capture supporting evidence.
Bad preparation:

- publish “factory is mandatory in the textile DPP” as current law
- expose every supplier site publicly before access rules are known.

### Textile recycled content, environmental information and durability

These are material development territories in the JRC study.

Businesses with credible data can improve methodology and evidence now. They should not assume the final delegated act will select every proposed field or preserve every proposed methodology unchanged.

Treat the category as:

**In official development — evidence useful for readiness, legal requirements not final.**

### Tyre circularity, abrasion and future DPP data

[Tyres](https://activatedigital.ai/knowledge/category/tyres) are on the 2027 ESPR track.

Existing tyre-label information is already structured under Regulation (EU) 2020/740.

Keep that information clean. Monitor how future DPP rules treat abrasion, circularity, material/recycled content and existing product-database fields. Do not prelabel those future choices as settled.

### Iron and steel

The Commission currently plans the iron and steel delegated act for Q4 2026 and discusses possible data territories.

For businesses in the category, this is close enough to justify active source mapping now.

It is not close enough to justify a self-invented final statutory field list.

### Service providers and horizontal implementation

ESPR provides for delegated rules concerning DPP service providers, and the Commission currently places relevant action in 2027.

Architectural implication:

- keep your data exportable
- avoid provider lock-in
- document ownership and continuity
- make storage and resolution components replaceable
- watch final provider qualification, backup and persistence rules.

## Which requirements should remain flexible?

These are the assumptions most likely to create expensive rework if converted into immutable platform rules.

### “Every DPP uses GTIN”

Not established.

UPI is the horizontal legal concept. GTIN can be useful and may satisfy requirements in some implementations, but it is not the universal definition.

### “Every DPP uses GS1 Digital Link”

Not established as a universal EU legal rule.

Support it if useful. Do not make the entire compliance truth model depend on it being legally compulsory for every category.

### “Every DPP uses QR”

Battery law specifically uses QR. The horizontal concept is a [data carrier](https://activatedigital.ai/knowledge/passport/choosing-a-carrier-that-still-works).

### “Every DPP is item-level”

False as a universal claim. Detergents and toys are adopted model-level regimes and construction uses product type.

### “Every DPP needs factory data”

False as a universal claim. Battery plant location is adopted, textile site data are development evidence and Toy factory/storage addresses sit in technical documentation rather than the mandatory DPP dataset.

### “Every DPP needs carbon or recycled content”

False as a universal claim.

Battery law contains category-specific sustainability information. Textile and other categories have development signals. Horizontal DPP law does not make these universal product fields.

### “Every DPP exposes the full supply chain”

Not established.

Traceability, responsible sourcing, factory data and upstream supplier information are different concepts with category-specific rules and access implications.

### Should I wait until the final regulation?

No, if “wait” means leaving product identity, regulated data and evidence governance in poor condition. Yes, if the alternative is inventing final legal requirements that haven’t been adopted.

Do both at once: improve the substrate now and keep the regulatory layer configurable. [What to do now, and what to refuse to do now](https://activatedigital.ai/knowledge/guides/what-to-do-now) sorts the rest by which decisions can’t be undone.

### Should I implement GTIN now?

If your business already uses or benefits from GTIN, maintain it well.

Do not implement it solely because someone has told you every future EU DPP legally requires GTIN. Current primary evidence does not support that universal claim.

See [Digital Product Passport Identifiers](https://activatedigital.ai/knowledge/fields/identifiers).

### Should I make everything item-level?

No.

Serialisation can be valuable for authenticity, service, warranty, high-value assets and certain regulated products. Those business cases are separate from the claim that DPP law requires item-level passports everywhere.

Design the hierarchy so item identity can be added or used where needed.

### Should I collect factory information?

It can be a low-regret internal capability if your supply chain already knows production sites, especially for textile, battery and industrial contexts.

The status still matters:

- Battery plant location: adopted passport information
- textile production sites: official development
- toys: manufacturing/storage addresses in technical documentation, not mandatory DPP Annex VI fields
- universal DPP factory requirement: not established.

### Should I calculate carbon now?

Only if there is a real legal, customer, product or internal need and a defensible methodology.

Do not manufacture a carbon number merely to fill an imagined universal DPP field.

Where category law eventually requires carbon, method, scope, verification and underlying evidence matter as much as the number itself.

### How should I handle unknown data?

Do not convert missing evidence into a guessed value.

A product-data system needs explicit states for:

- known and evidenced
- known but evidence incomplete
- conflicting sources
- not collected
- not applicable
- not established as a regulatory requirement.
The last of those is a **regulatory evidence state**, not a product-value state.

Retain [What You May Publish When You Do Not Know](https://activatedigital.ai/knowledge/evidence/stating-what-you-do-not-know) for the product-publication problem.

### What architecture should remain flexible?

At minimum:

- identifier scheme
- carrier type
- model/batch/item level
- access rules
- facility/origin fields
- category-specific sustainability data
- service-provider integration
- Registry interfaces
- data schema extensions
- evidence and method versions.
The stable core should be governance and relationships, not today's speculative field list.

### Direct answers

What product data should companies prepare now?

Governed product identity, actor identity, classifications where applicable, information already required under current law, conformity evidence, provenance, version history and machine-readable export capability.

Which fields are low-regret?

The lowest-regret investments are capabilities and governed master data, not a universal set of speculative DPP fields.

Should I wait for final delegated acts?

Wait before freezing unconfirmed legal fields. Do not wait to improve poor product-data governance.

What evidence should I retain?

Source document, issuer, date, scope, version, product/batch/site applicability, legal or claim context and change history.

## How we counted the test

ActivateDigital's controlled catalogue test covered **10 textile products and 34 variants from one Shopify test shop**. The supplied report is dated 3 October 2026 and was reviewed in the evidence pack on 5 October 2026. The run was offline, before merchant answers.

A decision sits at the level where the fact belongs. It is not necessarily a complete field or a separate merchant question.

The approximately 71% result is the upper observed record, the Seamless Sculpt Legging: **37 of 52 scoped decisions settled**, including shared company rows. Those 37 comprise **29 resolved values and eight method-specific not-applicable outcomes**. The latter relate to material not mapped in the selected environmental method; they do not mean zero impact or exemption from a legal duty. A resolved value may also carry a partial qualification. This is a bounded test result, not a typical-store forecast.

The initial required-question queue averaged **6.1 per product**, with a median of six and a range of four to eight. Excluding the shared company-identity question, the average was **5.1 product-specific questions**. These were measured before answers. Follow-ups can arise and optional gaps can remain open. The test did not measure completed merchant journeys, elapsed time or legal compliance.

No environmental model fetch completed in this offline run. The conditional online scenario is not presented as an observed result. No resolved row came from document ingestion or an external register in this catalogue test, so it does not measure those routes' coverage. The fixture contained no conflicted rows and does not measure a conflict-handling success rate.

## What would change this page?

A new matched engine run can update the measured result. Completed merchant journeys can support broader effort claims. New textile-specific product rules can change the legal requirements. These are separate reasons to update the evidence, product guidance or legal-status page.

## Sources

**First-party test evidence:** ActivateDigital controlled textile catalogue report, as-of 3 October 2026, reviewed 5 October 2026. The sample and counting method are set out above. This is internal test evidence, not an independent audit or market benchmark.

**Legal status:** European Commission, [Textile Apparel and the Digital Product Passport](https://single-market-economy.ec.europa.eu/single-market/digital-product-passport/textile-apparel_en), reviewed 5 October 2026.

**Shopify field definitions:** Shopify's [InventoryItem reference](https://shopify.dev/docs/api/admin-graphql/2026-10/objects/InventoryItem), [ProductVariant reference](https://shopify.dev/docs/api/admin-graphql/2026-10/objects/ProductVariant) and [product CSV guidance](https://help.shopify.com/en/manual/products/import-export/using-csv), reviewed 5 October 2026. Field availability is separate from what a store has populated or a connection retrieves.

### Sources retained for the cross-category preparation sections

- [Regulation (EU) 2024/1781 (ESPR), Articles 9–13 and Annex III](https://eur-lex.europa.eu/eli/reg/2024/1781/oj) EU law
- [Regulation (EU) 2023/1542 concerning batteries and waste batteries, Articles 77–78 and Annex XIII](https://eur-lex.europa.eu/eli/reg/2023/1542/oj) EU law
- [Regulation (EU) 2026/405 on detergents and surfactants, Articles 21–25 and Annex VI](https://eur-lex.europa.eu/eli/reg/2026/405/oj) EU law
- [Regulation (EU) 2025/2509 on toy safety, Articles 19–23 and Annex VI](https://eur-lex.europa.eu/eli/reg/2025/2509/oj) EU law
- [Regulation (EU) 2024/3110 on construction products, Articles 75–80](https://eur-lex.europa.eu/eli/reg/2024/3110/oj) EU law
- [Commission Implementing Decision (EU) 2026/1736 on harmonised DPP standards](https://eur-lex.europa.eu/eli/dec_impl/2026/1736/oj) Delegated or implementing act
- [JRC/Product Bureau, Study in support of the Digital Product Passport for textiles](https://susproc.jrc.ec.europa.eu/product-bureau/sites/default/files/2026-05/Textiles_DPP_20260513.pdf) Official study
- [ESPR and Energy Labelling Working Plan 2025–2030, COM(2025) 187 final](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52025DC0187) Official guidance
- [Regulation (EU) 2020/740 on tyre labelling](https://eur-lex.europa.eu/eli/reg/2020/740/oj) EU law
- [European Commission, DPP for Iron & Steel](https://single-market-economy.ec.europa.eu/single-market/digital-product-passport/iron-steel_en) European Commission
- [European Commission, Digital Product Passport implementation timeline](https://single-market-economy.ec.europa.eu/single-market/digital-product-passport_en) European Commission
This is a readiness resource, not a universal statutory DPP checklist or personalised legal advice.

The navy-hoodie record is fictional and illustrates evidence states only. It is not part of the controlled catalogue test. Its unit conversion is arithmetic; no environmental result or new performance claim is introduced.

## For a small business

### If you do not have a PIM, look across the systems you already use

For a smaller business, product information is often distributed across ordinary operating tools rather than one product-data platform. Check the ecommerce catalogue, product spreadsheets, shared drives, supplier specifications, certificates, test reports, customs/origin records and accounting or ERP records where present.

Do not treat “not visible on the website” as “not held by the business”. Use the access boundary instead:

| State | Meaning |
|---|---|
| `PUBLICLY_OBSERVABLE` | directly visible in a public source |
| `PUBLICLY_INFERABLE_WITH_CAVEAT` | inferable only with an explicit caveat |
| `INTERNAL_ACCESS_REQUIRED` | company records are needed |
| `SUPPLIER_ACCESS_REQUIRED` | supplier/upstream access is needed |
| `NOT_MEASURABLE_RELIABLY` | the available method cannot establish it reliably |

**NOT FOUND PUBLICLY ≠ BUSINESS DOES NOT HAVE IT**

For each important value, record both the working source and the evidence reference. A Shopify field can be the publication destination without being the independent evidence behind the value.

### Suggested onward links

- [Which Product Facts Can a Small Business Verify Itself?](https://activatedigital.ai/knowledge/evidence/small-business-product-fact-observability)
- [Which System Should Own Each Product Fact?](https://activatedigital.ai/knowledge/product-data/which-system-owns-each-fact)
- [Getting Evidence Out of Suppliers](https://activatedigital.ai/knowledge/evidence/supplier-evidence)
- [How to Organise Product Data Without a PIM](https://activatedigital.ai/knowledge/product-data/small-business-without-pim)

## Don't start with a blank passport. Start with what you already have.

Connect the product data and evidence you already hold. ActivateDigital does the hard work of building the textile passport, asks only for the facts that remain and turns the result into a governed product record ready for publication once the required checks are met.

For preparation that serves more than one destination, explore [which product data is worth fixing once](https://activatedigital.ai/knowledge/guides/product-data-worth-fixing-once). For specialist field, evidence and regulatory questions, use [Textile DPP Knowledge](https://activatedigital.ai/knowledge/digital-product-passport/textiles).

**We resolve the passport. You confirm the few facts only you can know.**

## Related

- [DPP Data Requirements: Mandatory, Conditional and Voluntary Information](https://activatedigital.ai/knowledge/passport/mandatory-conditional-voluntary-data) — The obligation-strength model

## Keep exploring

The question this one usually raises next.

- [Evidence behind thisEvidence & suppliersWhat you are allowed to put when you do not knowWhat named instruments already permit when a value cannot be established, and where an unbounded unknown stops a sale.→](https://activatedigital.ai/knowledge/evidence/stating-what-you-do-not-know)
Also worth reading

- [CompareImplementationWhat level is this value true atThe second of the four questions this page asks, worked through on its own. A value that is true of a style and recorded against a garment is the commonest way the level question goes wrong, and the gap between the levels runs to three orders of magnitude on an ordinary range.→](https://activatedigital.ai/knowledge/guides/granularity)
- [CompareImplementationWhat happens when the value movesThis page is about a value at rest, in the system it was born in. The moment it moves between systems a different set of things goes wrong, and they go wrong silently for the same reason.→](https://activatedigital.ai/knowledge/guides/what-does-not-map)
- [Evidence behind thisSmall businessWhich Product Facts Can a Small Business Verify Itself?Not sure whether a blank means missing, internal or supplier-held?→](https://activatedigital.ai/knowledge/evidence/small-business-product-fact-observability)