SME
A 90-Day DPP Pilot
How to run an end-to-end Digital Product Passport pilot on one product family — a week-by-week plan, team, outputs and success criteria.
The most common mistake in DPP projects is starting with the entire catalogue. The outcome is usually the same: supplier questionnaires go unanswered, data cleaning never finishes, the project slips.
The alternative is simple: one product family, 90 days, end to end.
The week-by-week plan
| Week | Task | Output |
|---|---|---|
| 1–2 | Scope and product selection | Chosen product family, target market |
| 3–4 | Data inventory and field mapping | Field mapping table |
| 5–6 | Supplier data requests | Requests sent |
| 7–8 | Identifier and carrier decisions | GTIN/URI scheme, QR design |
| 9–10 | Data entry and validation | Draft passport |
| 11 | Label and scan testing | Test report |
| 12–13 | Publication, registry, review | Published passport plus lessons |
Choosing the right product
Look for:
- Medium complexity: the simplest product teaches nothing, the hardest stalls.
- Data access: a supplier relationship that works.
- Commercial relevance: something you would want to show someone.
- Representativeness: a typical example from your catalogue.
The team
Keep it small: one owner (from quality or product development), one technical contact and a link into procurement. Thirty to fifty per cent of the owner's time is enough. See DPP project roles and RACI.
Success criteria
The pilot succeeds if it answers:
- Which fields do we hold, which sit with suppliers, and which do not exist at all?
- What is our supplier response rate, and why?
- How many person-days does producing one passport take?
- Does our label scan in real conditions?
- What is our cost and time estimate for 500 products?
See DPP cost breakdown and a 12-month roadmap.
What not to do in a pilot
- Try to collect data for the whole catalogue.
- Wait for perfect carbon data.
- Start writing your own platform. See build vs buy.
- Finish without publishing — publication is the most valuable part of the learning.
Frequently asked questions
Is 90 days realistic?
With one product family and a limited supplier set, yes. If supplier responses lag it can stretch to four or five months.
Do I really have to publish the passport?
Yes. An unpublished pilot never exposes the real obstacles: labels, registry, multilingual content.
How should I use the results?
For the scaling budget and timeline — and they also produce the project definition a funding application needs.
