Most fashion brands have never seen a Digital Product Passport up close. The whole conversation happens at the level of regulations and deadlines, almost never at the level of what a customer sees after scanning a label. This article shows what actually sits on that page, and lets you build a working one in about a minute, for free.
What is actually on the page
The JRC study for apparel published in May 2026 lists 49 data points across four categories. That number scares people. In practice, more than half of it is information you already print on a care label or keep in a product sheet:
- Identification - model, batch number, and a product identifier (a GTIN if you have one; small brands often do not, and that is fine).
- Material composition - fibres with percentages, plus recycled content.
- Care - washing, drying, ironing.
- Origin and manufacturer - country of production and the legal entity placing the product on the EU market.
- Substances of concern - a declaration, not a chemistry paper.
- Durability, repair and warranty - what the product is built to survive and what you will fix.
The heavier end (carbon footprint, tier-by-tier supply chain detail) depends on what the textile delegated act finally demands. Nobody knows that yet, which is exactly why building your passport around the certain parts is the sane move today.
Batch, not item
This is the single most useful thing a small brand can learn about textile passports. The recommended minimum granularity is the batch, not the individual garment. One passport covers a production run, not every hoodie that leaves the warehouse.
The difference is enormous. Item-level passports mean unique codes, serialisation and a system to track which code went on which piece. Batch level means one QR code per production run, printed on the whole run, pointing at one page. A brand doing 120 pieces of a style has one passport to fill in, not 120.
The QR code is only the carrier
Standard EN 18220 defines the data carrier, and it points at ISO/IEC 18004. In plain language: an ordinary QR code. There is no special DPP symbology to buy, no proprietary tag, no licence.
What matters is the address behind it. Where a GTIN exists, the canonical form follows GS1 Digital Link, so the same code can work at the till and as the entrance to the passport. Where it does not, a short stable URL does the job.
The part almost nobody thinks about until it hurts: the URL has to outlive the print run. Labels are ordered in bulk, months before they are sewn in, and a garment can sit in stock for years. If the address dies with a subscription, the code on a physical product becomes a dead link. Ask any vendor how long their URLs live before you print anything.
Build one and look at it
Reading about this only gets you so far. We published three example passports - a coat, a chair and a battery pack - with full data and a QR code that resolves to a live page. Scan one and you see what a customer sees.
Being honest about what they are: the companies and numbers are made up. They are there so you can see a real passport rather than a screenshot, and so you can put a code in front of a colleague and argue about what belongs on the page. When you want your own, a free account builds it, and the first two passports are published at no cost.
A label, not just a code
A QR image is not a production file. Whoever prints your labels needs vector artwork at the right size, with bleed and crop marks, not a PNG dragged out of a browser.
So the panel gives you the label, not only the code: a 50x80 mm hangtag with 3 mm bleed, crop marks, your brand and product name, the composition lines, and the passport address as text in case the code gets damaged. It is an SVG, so your label supplier can scale it to whatever your tag actually is. That file is the one thing in this whole category that tends to be sold as an expensive add-on, and it is the part we think should be standard.
What the law actually requires today
Nothing, for your clothes. It is worth being blunt about this, because the noise around the EU registry has convinced a lot of people otherwise.
- The EU DPP Registry went live on 20 July 2026, and Implementing Regulation (EU) 2026/1778 sets its operating rules from 6 August 2026. A working registry is not an obligation for your products.
- The first mandatory passport is the battery passport on 18 February 2027.
- For textiles, the delegated act is expected in 2027, with real obligations realistically landing in 2028-2029.
- Separately, since 19 July 2026, large companies may not destroy unsold apparel and footwear. That one is already in force.
Which leaves you with a two-year window and a choice: treat the passport as paperwork you will panic about in 2028, or start using it now as a page that tells your product story, with compliance data already sitting in the right shape.
FAQ
Do I need a GTIN to have a product passport?
No. A GTIN gives you the GS1 Digital Link address, which is useful if you already sell through retail. Brands without one can use a short stable URL instead and add a GTIN later.
Does one passport per batch really satisfy the requirements?
Batch is the recommended minimum granularity for apparel in the JRC study, with item level left voluntary. The delegated act is the document that will settle it, but batch is the level to design for.
Is a QR code alone enough to count as a DPP?
No. The code is the data carrier. The passport is the structured data behind it, plus the guarantee that the address resolves for as long as the product exists.
Are the example passports real?
The pages are real and the QR codes work, but the companies, products and numbers are made up. They show the format, they do not host a real product. Permanent, editable passports come with a free DPPera account.
Can I print the example QR on real labels?
No. It points at a fictional product. Print codes only from your own published passport, whose address stays the same for as long as the product exists.
Łukasz Głuch
Builds DPPera, a digital product passport tool for small and medium-sized companies. Reads the delegated acts so you do not have to. Writes about what he checked in the registry, not about what vendors promise.
LinkedIn
