Energy Labelling and EPREL

Discussions of digital product passports frequently proceed as though nothing comparable exists. Something does, it has operated for years, and it is worth studying for what it got right and where it stops.

What EPREL already does

Regulation (EU) 2017/1369 requires suppliers to register models in the European Product Registry for Energy Labelling before placing them on the market. The label carries a QR code resolving to the public record.

The fourth row is the most instructive. EPREL already implements the public and permissioned split that passport designs describe as novel, with a compliance section visible only to market surveillance authorities.

What it proves

Several questions that occupy passport planning were genuinely open before EPREL and are now answered by demonstration rather than by argument.

• Manufacturers will register at scale when registration is a condition of market access.
• A public database can carry commercial data without collapsing the competitive position of the registrants.
• Consumers do scan when the code is on the label at the point of decision rather than hidden on packaging.
• Authorities use the data — it is a working enforcement tool, not a compliance artefact nobody opens.

The third finding is more qualified than it sounds and worth reading carefully. Scanning happens where the code sits at the moment of comparison, in a shop, against alternatives. It does not follow that consumers will scan a code on a product they already own.

The rescaling lesson

The energy label was rescaled because efficiency improvements had pushed almost every product into the top classes, making the label useless for discrimination. The A+++ proliferation was the visible symptom.

This is directly relevant to reparability scoring and any circularity rating. A score where most products cluster at the top conveys nothing, and the pressure not to downgrade products already on the market is substantial once the scheme is running.

Where EPREL stops

Its limits are limits of scope rather than of design, and they map precisely onto what ESPR adds.

The last node is the structural difference. EPREL registers models, which is sufficient for energy performance because every unit of a model performs identically. Material provenance, service history and recycled content are not always identical across units, which is why passports need finer granularity.

What manufacturers can reuse

For anybody already registering in EPREL, some of the passport work is done and it is worth identifying which parts.

Model identification, supplier identification, the technical documentation package and the discipline of registering before market placement all transfer directly. The organisational muscle for treating a database registration as a market access condition is also genuinely valuable and hard to build from nothing.

What does not transfer is the data itself, since energy performance is a narrow slice of what a passport carries. The gap is materials, substances, reparability and end of life, none of which EPREL asked for.

The coexistence question

Products covered by both energy labelling and an ESPR delegated act will have obligations under each, and duplicate registration is a reasonable concern.

The sensible internal approach is to hold one product record and generate both submissions from it, rather than maintaining separate datasets that drift. Energy performance figures declared in two places that disagree is a compliance problem entirely of the manufacturer’s own making.

This argues for treating EPREL as an output of the product record rather than as a system in its own right. Teams that built EPREL submission as a standalone process, frequently in a spreadsheet maintained by one person, will find that process is the thing blocking a single record rather than helping toward one.

The reverse framing is also worth stating plainly. An organisation that already treats EPREL registration as a disciplined, owned, repeatable process has most of the operating model a passport programme needs, and only the data scope has to grow.