ESPR Readiness: A Complete Assessment

Most ESPR readiness work starts in the wrong place. It starts with a technology question — which platform, which data model, which carrier — when the binding constraint is almost always that nobody inside the company can say who holds the recycled-content figure for a component bought through two intermediaries.

This document is the assessment we would run before any of those technology questions. It has four parts, in a deliberate order, and each one can be completed before the delegated act covering your product group exists.

What the ESPR actually obliges

The Ecodesign for Sustainable Products Regulation (EU) 2024/1781 entered into force in July 2024. It is a framework regulation, which is the single most important thing to understand about it: it creates powers rather than product requirements.

Article 4 empowers the Commission to adopt delegated acts setting ecodesign requirements for specific product groups. Those requirements can be performance requirements, information requirements, or both. The Digital Product Passport sits in the information category.

The practical consequence is that "when does ESPR apply to us?" has no single answer. It applies to your product group when a delegated act covering that group applies, and the regulation provides that such acts allow a transitional period before their requirements bite.

Determination one: will an act reach your products?

The ESPR working plan names the product groups the Commission intends to address first. Iron and steel, aluminium, textiles and apparel, furniture, tyres and mattresses appear in the early prioritisation, alongside a set of horizontal measures.

Read the working plan against your own product catalogue rather than against your industry. Companies routinely discover that a component or an accessory falls into a prioritised group even where the headline product does not — a furniture manufacturer selling with a textile cover, an electronics business selling a steel enclosure.

The second row is where most companies actually sit, and it is the row where the most money is wasted. An adjacent group does not justify a platform selection. It does justify knowing where your data lives, because that work has a long lead time and is useful whatever happens.

Determination two: which data do you already hold?

Annex I of the regulation lists the ecodesign parameters a delegated act may draw on. That list is fixed even where the acts are not, which is what makes assessment possible before adoption.

The parameters include durability, reliability, reusability, upgradability, reparability, the presence of substances of concern, energy and resource efficiency, recycled content, the possibility of remanufacturing, and the carbon and environmental footprints of the product.

Take that list to your existing systems and mark each parameter with where the data lives today. In our experience the result divides into three groups of roughly equal size, and the middle group is the one that determines your timeline.

Be honest about the second layer. A recycled-content figure that exists in a PDF attached to an email from 2023 is not data you hold in any sense that survives an audit, and treating it as held is the most common way a readiness assessment produces an optimistic answer.

Determination three: who owns the data you do not have?

This is the question that decides the programme, and it is the one most often skipped because it is uncomfortable. For every parameter in the third layer, name the legal entity that holds the answer.

Three patterns recur, and they have very different lead times.

• A direct supplier holds the data and will provide it under the existing contract. Lead time is weeks, and the work is a data request.
• A direct supplier holds the data and will not provide it without a contractual change or a commercial concession. Lead time is a contract cycle, often a year.
• The data sits below your direct supplier, with a party you have no relationship with. Lead time is unbounded, and the work is deep-tier engagement rather than a request.

The third pattern is where recycled content, substance declarations and carbon data usually sit. A tier-one supplier who assembles a component frequently cannot tell you the recycled content of the polymer in it, because they bought a compound rather than a resin.

Determination four: what does closing the gap cost?

Cost falls into four buckets, and only one of them is software. Estimating the other three is what turns an assessment into something a finance function can act on.

The fourth bucket is systematically underestimated. A passport is not a document produced once; it is a record that must remain accurate for as long as the product is on the market, through every specification change and supplier switch.

Size that bucket by asking how many product changes your business made last year, and who would have had to update a passport for each one. If the answer is nobody, the operating model is the gap rather than the technology.

How to sequence work against an act that does not exist

The objection to assessing early is reasonable: why build against requirements that are not fixed? The answer is that the parameters are fixed even where the thresholds are not, and the long-lead work depends on the parameters rather than the thresholds.

Running these in the published order is what makes an unadopted act tolerable. Running them in reverse — platform first, suppliers last — is how programmes arrive at the transitional deadline with a working system and no data to put in it.

What a finished assessment looks like

A readiness assessment is finished when it can answer a specific question from the board: if a delegated act covering our largest product group were adopted next quarter, what would we not be able to do?

1. A product-group scope statement naming which of your catalogue is in, adjacent, or out.
2. A parameter-by-parameter data map with each parameter marked held, unstructured, or absent.
3. A named owner for every absent parameter, with a lead time attached to each.
4. A cost estimate across the four buckets, with the operations bucket sized from your actual change rate.
5. A sequencing plan whose first two phases start now and do not depend on the act.

Note what is not on that list: a platform choice, a data model, and a carrier decision. Those follow from the assessment rather than preceding it, and an assessment that produces them first has usually skipped the third determination.

Who should actually run the assessment

An ESPR readiness assessment run entirely by a compliance function produces an accurate list of obligations and an unusable plan. The reason is structural: three of the four determinations require authority the compliance function does not hold.

Scope determination is a regulatory reading, and compliance owns it. Data mapping requires access to PLM and ERP, which engineering and IT control. Supplier ownership requires the ability to open a contract, which procurement controls. Cost estimation requires the change-rate figures that operations holds.

The second row is the one that most often goes wrong quietly. An engineer who knows a test report exists will mark a parameter as held; a person who has to extract two thousand of them will not. Ask the question in the form "could we produce this for every SKU next month?" and the answer changes.

Sponsorship matters more than headcount here. The assessment produces a procurement programme, and a procurement programme that originates in a compliance team without executive sponsorship does not survive contact with a supplier negotiation that has commercial terms attached.

What market surveillance will ask for

Readiness is ultimately measured against what an authority can demand, so it is worth working backwards from that rather than forwards from a data model. The regulation puts the burden of demonstrating conformity on the economic operator placing the product on the market.

In practice that means three artefacts, and a readiness assessment should end knowing which of them your company could produce today.

1. Technical documentation demonstrating that the product meets the requirements in the applicable delegated act.
2. An EU declaration of conformity, signed, in which a named legal person takes responsibility for that demonstration.
3. The information required to be made available, in the form the act specifies — which for most prioritised groups is expected to include a passport.

The first artefact is where readiness usually breaks. Technical documentation is not a summary of your data; it is the evidence chain behind each claim, including the test method, the date and the party who performed it. A recycled-content figure with no traceable basis is not documentation, whatever system it is stored in.

This is also the reason the operations bucket in the cost model is larger than it looks. Evidence ages. A test performed against a formulation that has since changed does not support a claim about the current product, and knowing which claims have gone stale requires linking specification changes to the evidence that depended on them.

What changes if you import rather than manufacture

An importer placing a product on the EU market carries obligations in its own right, and a readiness assessment built for a manufacturer transfers badly. The difference is that an importer cannot generate the underlying data; it can only obtain it.

That inverts the second and third determinations. For a manufacturer, the data map is mostly internal with a supplier tail. For an importer, almost every parameter sits with a party outside the EU, frequently outside any contractual regime that anticipated these requirements.

The practical consequence is that an importer readiness assessment should start at determination three and work backwards. If the manufacturer will not supply parameter data, no amount of internal system work produces a compliant record, and the commercial question — whether to change supplier — arrives much earlier than it does for a manufacturer.

What to do in the week an act is adopted

The value of an assessment is measured on the day it stops being hypothetical. When a delegated act covering your product group is adopted, the transitional period starts running and the questions change from strategic to operational.

A completed assessment turns that week into a comparison rather than a discovery exercise. You already know your parameters, your data map and your supplier lead times; what the act adds is thresholds, the conformity assessment route, and the exact information format.

1. Diff the act against your parameter map: which parameters it actually sets, and which of those you marked absent.
2. Recheck lead times against the transitional period, because the gap between them is the only number that matters now.
3. Confirm the conformity assessment route, since that decides whether a third party has to be involved and how early.
4. Freeze the information format, which is the point at which carrier and data model decisions become answerable.
5. Escalate any absent parameter whose supplier lead time exceeds the transitional period — that is a commercial decision, not a compliance one.

The last step is the one worth rehearsing before it is needed. If a parameter requires a contract change that takes eighteen months and the transitional period is shorter, no amount of programme management closes the gap; the choices are to change supplier, to change the product, or to withdraw it from the market.

Discovering that in the week of adoption is survivable. Discovering it three months before the requirement applies is not, and the difference between those two positions is whether determination three was done early.

The argument against doing this now

There is a real counter-argument, and a readiness assessment that does not state it is selling something. It runs: the delegated acts are not adopted, the thresholds are unknown, and work done now may be work done twice.

That argument is sound for phases three and four. It is not sound for phases one and two, because a supplier data map and a set of contract changes retain their value whatever the thresholds turn out to be, and because both have lead times measured in quarters rather than weeks.

The honest position is therefore partial: begin the work whose value does not depend on the act, and defer the work that does. That is a narrower commitment than most vendors ask for, and it is the one the regulation actually justifies today.

One caveat on the assessment itself. Its output ages, because supplier relationships change and product specifications change with them. A data-ownership map eighteen months old will misstate who holds what, and a plan built on it will misstate the lead times that follow.

Refresh it annually, and refresh it immediately on two triggers: a new working plan that names an adjacent product group, and any supplier change affecting a parameter you marked absent. Both change the answer to determination three, which is the determination the whole timeline rests on.