Nobody can tell you what the EU Digital Product Passport will require of a knit garment, because the textile-specific requirements do not exist yet. The Ecodesign for Sustainable Products Regulation is in force and textiles are named as a priority category, but the European Commission currently plans adoption of the textile delegated act in Q4 2027. Economic operators will then have a transition period of at least eighteen months before the new product-specific requirements apply.
So when someone tells a factory in Gazipur or Narayanganj that there is a deadline next year and they should buy something today, the honest response is: which deadline, and written by whom?
But there is a second question underneath that one, and it is the one that decides whether the next two years are wasted. Not when does the law arrive. When does the buyer arrive.
Those are not the same date. The second one is much earlier.
The law has never been first
Ask anyone who has run a floor here since the nineties. No Bangladeshi statute made ISO 9001 universally mandatory for export factories. BSCI, SEDEX, WRAP and Higg were not blanket legal obligations on every Bangladeshi exporter. Yet many of them became commercially mandatory for suppliers because buyers wrote them into sourcing or purchase conditions, and factories that could not meet those conditions risked losing orders.
Look at what actually separated factories in those cycles. It was almost never who understood the standard best. It was who had the underlying records already in order when the questionnaire arrived. A factory with clean, dated, attributable production records could satisfy a new audit framework in weeks, using its own staff. A factory without them hired a consultant at three times the cost, produced a binder assembled backwards from memory, and passed on paper while staying exposed to the first auditor who pulled a thread.
The standard was the visible thing. The record underneath it determined the outcome.
That pattern is about to repeat — and this time the record is the entire point.
Why the buyer cannot wait for the delegated act
There is a structural reason the questionnaire arrives before the rule.
Under the regulation, the obligation to produce a passport falls on the economic operator placing the product on the EU market. For a Bangladeshi factory supplying an EU brand or importer, that responsibility may sit downstream rather than with the factory itself, depending on the market arrangement.
That operator has a problem. It has to publish, at product level, data it does not itself generate — fibre content, mill of origin, dyeing and finishing inputs, who did what and when. All of it originates upstream, in suppliers it does not own and cannot audit at will. Assembling that pipeline across dozens or hundreds of suppliers is a multi-year programme, not a procurement exercise.
Which means a buyer that expects future textile DPP obligations has reason to start testing and collecting supplier data well before the legal application date. It cannot wait for the final schema to discover whether its suppliers can record information consistently at all.
So the sequence is not: rule published, then brands comply, then factories get asked. It is: buyers can start asking earlier, find out which suppliers can answer, and identify data gaps and stronger supplier relationships before the final rule applies.
That sorting can begin years before the legal deadline.
What the question actually sounds like
It may not arrive as “are you DPP-ready.” Brands do not necessarily ask that, because they do not know the answer for themselves yet.
It arrives as something narrower and much harder to bluff.
For this style, on this purchase order, what was the fibre composition — and which batch does that number belong to?
Which mill supplied that fabric? Which spinner supplied that yarn?
When was that recorded, and by whom?
And if I asked you the same question about a shipment from fourteen months ago, would I get the same answer?
The last one separates factories. A merchandiser can reconstruct an answer to the first three from files and phone calls. Nobody can reconstruct consistency. Either the record was captured at the time, by a named person, and has not been rewritten since — or it was not.
What is worth building before the rule exists
The distinction that matters is between the format layer and the record layer.
The format layer is the passport itself: the schema, the identifier standard, the data carrier, the registry submission. Parts of the horizontal DPP infrastructure already exist, including the Registry and published technical standards, but the textile-specific data requirements are not yet final. Anything built today as if the final apparel specification were already settled risks being rebuilt.
The record layer is different. Whatever the delegated act says, it will demand the same three properties of the underlying data, because no evidence system has ever worked without them.
Attribution. Every claim has an author. Not “the factory says,” but a specific person or system, identifiable afterwards. A number with no author is an assertion, and assertions do not survive being challenged.
Timing. The record was created when the event happened, not when the question was asked. A production entry dated to the shift is evidence. The same number typed in six months later, during a buyer escalation, is a reconstruction — and it reads like one.
Scope. The record states what it covers and what it does not. A factory that says “cut-and-sew and fabric documented, spinning not attested” is more useful to a brand than one presenting an unbroken chain it cannot support. The first can be relied on selectively. The second cannot be relied on at all, because the brand has no way to know which part is soft.
Those three properties are buildable today, against no final textile schema, and they carry forward into every version of the rule. That is the work that is not wasted.
What this does not solve
Being straight about the limits, because the market currently is not.
There is no final ESPR-compliant textile passport available today, because the textile-specific delegated act has not yet been adopted. The EU DPP Registry is live and organisations can enrol and use the testing environment, but the product-specific textile requirements are still being developed. A vendor can build a working digital product record today; it cannot honestly present that record as proof of compliance with a final textile rule that does not yet exist.
Registry enrolment or economic-operator verification is not the same as textile DPP compliance. It establishes identity and access within the Registry framework; it does not settle the future textile requirements.
Records built now do not make a factory compliant, because the final textile requirements are not yet in force. They make a factory answerable, which is the only thing a buyer can actually evaluate in 2026.
And a record a factory generates about itself remains its own account of its own operation. Attribution and timing make it checkable, and hard to quietly revise. They do not make it independently confirmed, and no honest system should tell a factory otherwise. What it earns is narrower and still valuable: a buyer can see exactly which claims are backed by a document, which are the factory’s own statement, and which are simply not covered — and can act on that instead of guessing.
The sentence worth being able to say
When the questionnaire comes — perhaps from a merchandiser, by email, with a short turnaround and little warning — the factories that win the next order will be the ones that can say:
Yes. Here is the record. It was captured at the time, by a named person, and here is precisely what it covers and what it does not.
Everything else is a scramble. The factories that ran the last four certification cycles here know exactly what that scramble costs, and know it is always more expensive than having kept the record in the first place.
The textile rule is not final yet. The buyer does not have to wait for it.