Do you need GS1 for a digital product passport? No - and the reason is not the price

By Łukasz Głuch · 17 August 2026 · DPPera

Ask five DPP vendors what you need to get started and at least three will begin with "first, get your GTINs from GS1". It is the single most repeated piece of misinformation in this market. No EU rule names GS1 as the way to identify a product in a digital product passport. What follows is what the rules actually say, what the Commission's own analysts chose when they had to pick a real identifier, and what GS1 costs in Poland - because the honest answer about cost is not the one you expect.

What the regulation actually asks for

The Ecodesign Regulation (EU) 2024/1781 requires a passport to carry a unique product identifier. It does not say who issues it. The same is true of the unique operator identifier and the unique facility identifier that sit alongside it.

The harmonised standard covering this area, cited in the Official Journal on 15 July 2026, is EN 18219:2026 - Digital product passport - Unique identifiers. Note the title. Not "GS1 identifiers". The standard is vendor-neutral by construction, because a European standard cannot hand a market to one issuing body.

Which identifier scheme applies to your product is decided in the delegated act for your product group - and those acts are still being written. We covered the six standards and what "harmonised" buys you separately.

The proof: the regulator did not pick GS1 either

This is not a theoretical argument. When the Commission's own Joint Research Centre had to propose concrete passport content for the first product group to get a delegated act, it did not reach for a GTIN.

The JRC study behind the iron and steel act proposes the heat number - the steel industry's existing batch identifier - as the mandatory identifier, with item-level identification optional. GTIN appears in the proposed dataset as a public field, but as one attribute among many, not as the backbone. We went through that study field by field.

The pattern is consistent: the preparatory work reuses whatever identifier an industry already has, at batch level, because retrofitting a new serialisation scheme is the expensive part. For steel that is the heat number. For clothing it is your production run. In neither case does the analysis start by sending you to buy a number.

What GS1 actually costs in Poland - and why that was never the point

Most articles arguing against GS1 argue about price. Let us be accurate instead, because the real figures undercut the usual scare story.

Per the GS1 Polska price list, for a company with net revenue up to 2 million zloty:

• the registration fee for the first pool of up to 10 numbers is waived (new participants only, one such pool per participant);
• the annual fee is 136 zl gross.

So for a small Polish brand, GS1 costs about the same per year as two months of a mid-range SaaS subscription. Cost is not the reason to avoid it.

The reason to think twice is ten. Ten numbers is a small pool if you identify at item or variant level - and the next pool up, 100 numbers, carries a 307,50 zl gross registration fee, payable once. Fees also scale with company revenue, not with how many numbers you use: cross 2 million zloty and the annual fee moves to 267 zl gross, and upward from there. None of that is ruinous. It is simply a commitment you should make on purpose, for reasons that have nothing to do with DPP compliance.

So when do you actually want GS1?

There are good reasons, and they are all commercial rather than regulatory:

• You sell through retail or marketplaces. Chains and large marketplaces require a GTIN to list a product. If you are already there, you already have one - and you should absolutely reuse it in your passport rather than inventing a second identifier.
• You want one code to do two jobs. GS1 Digital Link lets a single 2D code work at the checkout and as the entrance to a passport. That is a genuine operational win if your product crosses a scanner.
• Your supply chain already speaks GS1. If your partners exchange data on GS1 standards, matching them removes translation work.

What none of those is: a precondition for having a passport at all.

What has NOT changed

A passport still needs an identifier that is unique, stable and resolvable - one that keeps working years after the product left your warehouse. Rejecting GS1 does not mean improvising: published schemes exist. W3C DID, for decentralised identifiers, is free to read. So, incidentally, are GS1's own specifications - GS1 Digital Link and EPCIS cost nothing to download, which is worth knowing before you assume "not using GS1" saves you reading. Two other reference points, IEC 61406 for identification links and ISO/IEC 15459 for unique identification, are sold by IEC and ISO the way EN standards are sold by national bodies. Self-issued does not mean self-invented - but it does not automatically mean free either.

And the identifier is not the hard part of a passport. The hard part is the data behind it, and who is allowed to see which field.

What to do with this

• If you already have GTINs, use them. This article is not an argument for throwing away an identifier you own and your buyers recognise.
• If you do not, do not buy them because of DPP. Buy them when a retailer, a marketplace or a real customer asks. That is a sales decision.
• Identify at batch level, not item level. One passport per production run. Two independent preparatory studies, in two unrelated industries, reached that same answer.
• Refuse any tool that hard-blocks you without a GTIN. A vendor whose onboarding cannot start without one has built their product around an assumption the regulation does not make - and has quietly outsourced part of your onboarding to a third party.

FAQ