From SKU to Passport: Organising Product Data

Most of the data you need for a Digital Product Passport (DPP) you already have — it is just scattered. It sits in the ERP system, in spec sheets, in declarations of conformity and in supplier files. This guide shows how to go from a list of SKUs to a finished, consistent passport.

A passport starts with data, not technology

Rolling out a DPP is largely a data-tidying project. Before you pick a carrier or a platform, you need to know what product information you hold, where it lives and whether it is up to date.

The good news is that a passport rarely requires creating data from scratch. The challenge is bringing existing information together into one complete and trustworthy record.

Start with a SKU inventory

An SKU is your starting point, but an SKU is not yet a passport. First assemble the list of products and work out which ones actually fall under a DPP obligation, and on what time horizon.

Sort the list by a few criteria:

• product group and the future delegated act that applies to it,
• the markets you sell into,
• volume and business priority,
• availability of source data for the product.

That way you begin with the most urgent products rather than random ones.

Decide the level of detail

Before you gather data, decide at what level you will describe the product: model, batch or individual item. This choice follows from the sector and regulatory goal, and it drives how many identifiers and records you will maintain.

The model level is the simplest to maintain; the item level is the most detailed but also the most demanding. Batteries, for example, require a passport at the level of the individual item. More in model, batch, item.

Map the data sources

The next step is a map: which piece of information comes from which system. Typical sources are the ERP, PIM, technical sheets, declarations of conformity and documentation from material suppliers.

For each required field, establish:

• where the source data sits,
• who is responsible for keeping it current,
• how often it changes,
• whether it can be pulled automatically or must be entered by hand.

This exercise reveals gaps before they become a problem at publication.

Organise by data layers

Instead of treating data as one big bag, arrange it into the layers that will appear in the passport anyway:

• product identity — identifiers, model, variant,
• responsible operator — who places the product on the market,
• composition and materials,
• environmental and life-cycle information,
• traceability and evidence — links to compliance documentation.

This is a helpful framing, not a rigid legal list — the specific fields will be set by the delegated act for the group. We describe the detail in what data a DPP requires.

Ensure quality and consistency

Completeness is one thing, but consistency matters too: the same units, date formats and material names across the whole catalogue. Discrepancies between systems are the most common cause of errors in passports.

Establish a single "source of truth" for each field and avoid keeping the same information in several places at once. Also plan who will update the data after publication and how, because a passport lives alongside the product.

Prepare for publishing and testing

Once the data is in order, you can translate it into a passport, generate the carrier and test the whole flow from scan to display. Well-organised data makes this stage a formality rather than firefighting.

We describe the next step — end-to-end testing and rollout — in testing and deploying a DPP.

Involve the right people

Tidying product data is rarely a one-person job. The data comes from different departments — from purchasing and production, through quality and compliance, to marketing and IT. Without their cooperation, gaps and conflicting versions are easy to create.

Decide at the outset who owns which fields and who approves the final record. A clear division of roles shortens the project and reduces the risk of data getting stuck between departments.

Automation and common pitfalls

With a larger catalogue, retyping data by hand is not only slow but also error-prone. Where possible, pull data directly from source systems instead of copying it into yet another spreadsheet.

Watch for the usual pitfalls: the same materials described under different names, inconsistent units, outdated declarations or data that diverges between the ERP and the spec sheet. Catching these during the tidying stage is far cheaper than fixing published passports.

Plan data upkeep for the future

Tidying data is not a one-off clean-up but the start of an ongoing process. Products change, suppliers update materials, and sector requirements will be refined in delegated acts.

Decide how often you review the data, who is responsible for updates and how you mark versions. A simple review rhythm keeps passports current instead of letting them age in the background. It also lets you add new products faster, reusing the order you established once.

Key takeaways

• Rolling out a DPP is above all a project of tidying existing data, not creating it from scratch.
• Start with a SKU inventory and work out which products fall under the obligation, and when.
• Decide the level of detail (model, batch, item) before you gather data.
• Map the data sources and arrange information into the passport's layers.
• Consistency of units, formats and names is as important as completeness.

See how CyfroPass helps turn product data into a finished passport. Visit cyfropass.pl and build your first passport from your own data.