What data must a Digital Product Passport contain?
Material composition and sourcing, durability and repairability, carbon footprint, substances of concern, and end-of-life handling. The precise fields are fixed per product group in its delegated act, which is why mapping data to a standards-based model beats building to a form that has not been published yet.
Data
Does a passport contain personal data?
It should not, and designing so it cannot is the safer position. Passport content is product data. Where a repair or resale record could identify an individual, the identifying part belongs in your own systems under normal GDPR controls, not in a record designed to be publicly resolvable for fifteen years.
Data
What if we do not have the data yet?
That is the normal starting position, and it is a supplier problem before it is a software problem. Most missing attributes sit with tier-2 and tier-3 suppliers who have never been asked. Start by listing the named gaps against your product group’s data set, then run one collection campaign per attribute set rather than per product.
Data
How accurate does the data have to be?
Accurate enough to defend to a market surveillance authority, and attributable to whoever asserted it. That is why signed credentials matter more than precision theatre: a figure with a named issuer and a method behind it survives scrutiny, and a confident number with no provenance does not.
Data
How often must a passport be updated?
Whenever the underlying facts change — a new supplier, a reformulation, a corrected footprint, a withdrawn certificate. Passports are versioned records rather than one-time publications, which is why the ability to correct them matters and why immutable storage is the wrong architecture for the content.
Data
Why can we not just publish our existing PDFs?
Because a passport attribute asks for a typed value against a product identifier, and a document containing that number is not addressable by anything. The information exists in the organisation without being reachable, which is the gap most programmes underestimate at the outset.
Data
How do we prove a published claim if challenged?
Through provenance recorded with the value: who asserted it, when, by what method and on what evidence. A figure without its method is not interpretable, and a self-declared number and an audited one are both legitimate while being very different evidence.
Data
What is the most dangerous kind of bad passport data?
Stale values, because they look complete. A gap is visible and everyone treats it as a gap, while a value that was correct three years ago and is no longer true gets acted on. Recording when each value was established is the defence.
Data
Will schema validation catch our data errors?
Only format errors, which is the smallest category. Unit confusion, granularity mismatch, silent defaults and stale values all produce structurally valid data that is simply wrong. Distribution analysis across similar products catches what validation cannot, by flagging values orders of magnitude away from their peers.
Data
What should we measure instead of data completeness?
Proportion of attributes with recorded provenance, proportion resting on primary rather than estimated data, and median value age. None of the three can be improved by filling a field with a guess, which is exactly what a completeness percentage rewards.
Data