The DPP data collection playbook: why readiness starts months before the label

Every Digital Product Passport programme we have seen fail has failed the same way: the team started from the label and worked forwards, instead of starting from the data and working backwards.

The QR code is the visible artefact, so it gets the attention. But printing a QR code takes an afternoon. Assembling the verified data behind it takes two to four quarters — and most of that time is spent waiting on other people. This playbook lays out the sequence that works, based on what battery teams learned the hard way ahead of their 2027 deadline.

6–12 mo

Typical supplier data lead time

40–70

Data points per passport, category-dependent

4–6 wk

Artwork freeze before production print

Phase 0 — Establish scope before anything else (2 weeks)

One workshop, three questions, written answers:

1. Which delegated act applies to us, and when? Batteries have a fixed 2027 date; textiles, tyres, steel and furniture ride the ESPR Working Plan 2025–2030. If you span categories, you are running multiple timelines, not one.
2. What granularity of identity? Unit-level serial (batteries, premium goods), batch-level (most textiles), or model-level (some furniture)? This decision drives cost by orders of magnitude and is painful to change after labels are printed.
3. Who is the economic operator? The legal entity placing the product on the EU market owns the obligation. For import-heavy structures this is often not the brand entity everyone assumed.

Phase 1 — The data audit (4–6 weeks)

Take the data requirements for your category and build a spreadsheet with one row per data point and four columns: source system, owner, format, confidence.

The pattern that emerges is always the same:

• Green (~50%): lives in your PIM/ERP already — dimensions, materials, weights, product identity.
• Amber (~30%): exists somewhere internal but unstructured — test reports as PDFs, substance declarations in email threads, carbon models in a consultant's spreadsheet.
• Red (~20%): does not exist inside your company at all. Supplier-held: recycled-content evidence, process chemistry, tier-2+ due-diligence documentation.

The red cells are your critical path. Everything else in the programme is scheduled around how long they take.

Phase 2 — The supplier campaign (1–2 quarters, runs in parallel)

Start the red-cell campaign the same week the audit ends. Rules that separate successful campaigns from ignored questionnaires:

• Ask for the minimum viable field set. Every extra field halves response rates. You can enrich later; you cannot recover a supplier who has decided your programme is homework.
• Accept evidence in the forms suppliers already have. A GRS certificate, an existing audit report, a test certificate reference — reuse before you demand new formats.
• Structured attestations over raw data. "Batch X contains ≥18% recycled content per certificate Y" travels; "give us your process database" does not. (More on this in our textile supplier piece.)
• Escalate through commercial channels, not sustainability ones. Data requests attached to purchase orders get answered. Requests from an email alias with "ESG" in it get archived.

Phase 3 — Identity and carrier decisions (2–3 weeks, once)

While supplier data trickles in, fix the identity layer:

• Identifier scheme: GS1 Digital Link on your GTINs. This is the one decision you truly make once — the codes outlive every software contract. (The full argument.)
• Resolver domain: your own subdomain (id.yourbrand.com) or your provider's. Own domain costs one DNS record and buys permanent portability.
• Carrier placement: survivability requirements differ — a battery QR must outlive 15 years of vibration and heat; a care-label QR must survive 50 washes. Involve packaging engineering now, because their artwork deadlines are earlier than anyone else's.

Phase 4 — Pilot on one product line (4–6 weeks)

Do not attempt the catalogue. Pick one line with cooperative suppliers and real complexity, then run the full path end to end:

1. Map audited data into the category data model.
2. Mint real passports (unit or batch level, per your Phase 0 decision).
3. Print physical labels on production materials — not office-printer mockups.
4. Scan-test every persona view: consumer phone, authority access, recycler tooling.
5. Rehearse a data update — a corrected substance declaration, a state-of-health event — and watch it version through to the live passport.

Step 5 is the one everyone skips and everyone regrets. Passports are living records; if your pipeline can only create and never amend, you have built a brochure, not infrastructure.

Phase 5 — Registry, artwork freeze, launch

The end-game items with unforgiving dates:

• EU DPP Registry: the central registry (live since July 2026) needs your passport identifiers synced before products ship. Platform-side this is automated — but the registration duty sits with the economic operator, so confirm it contractually. (What the registry actually does.)
• Artwork freeze: production printing typically locks artwork 4–6 weeks before units leave the line. Your passport URLs must be final by then. Work the calendar backwards from your first compliant shipment and put the freeze date in front of the whole programme team.
• Operational handover: who answers when a passport field is wrong in production? Compliance data has a lifecycle; assign it an owner with a rota, not a project team that disbands at launch.

Phase 6 — Governance: keeping the data alive

The programme does not end at launch; it changes shape. Passport data decays the moment nobody owns it — suppliers change inputs, test certificates expire, regulations tighten. The teams whose passports are still accurate a year after launch all put the same three structures in place:

• A data owner per field group. Composition data belongs to product engineering; substance declarations to quality; due-diligence evidence to procurement; carbon models to sustainability. Named roles, not a shared mailbox. When a field is wrong in production, the routing is already decided.
• Change triggers wired into existing processes. A supplier change order, a reformulation, a new test report — each already flows through some system of record. Add "update the passport" as a mandatory step in those workflows rather than inventing a parallel review cycle that nobody attends.
• A quarterly accuracy sample. Pull twenty random live passports, verify every field against source documents, log the defect rate. It takes an afternoon and it is the difference between discovering drift yourself and having a market surveillance authority discover it for you.

Budget roughly 10–15% of the original programme effort per year for this stewardship. Teams that skip the line item don't skip the work — they just do it as an emergency, at emergency prices.

The one-page version

When	What	Failure mode it prevents
Week 0–2	Scope: act, granularity, operator	Building for the wrong deadline
Week 2–8	Data audit, RAG-classified	Discovering red cells in month 6
Month 2–6	Supplier campaign	The universal critical path
Month 2–3	Identity + carrier decisions	Re-labelling regret
Month 4–5	One-line pilot, incl. an update	Brochure-ware passports
Month 6+	Registry sync, artwork freeze	Missing the only immovable dates

Start earlier than feels necessary. Every team that has shipped a compliant passport says the same thing afterwards: the technology was the easy part — the calendar was the opponent.

Want this playbook applied to your catalogue? Book a consultation and leave with a dated readiness plan for your category.