An 18-Month ESPR Readiness Plan

When a delegated act is adopted for a product group, roughly eighteen months follow before it applies. That period is frequently described as a transition or a grace period, which encourages exactly the wrong sequencing.

It is not a grace period. It is the window in which supplier data has to be obtained, and supplier response time is the one element of the plan that no amount of internal urgency compresses.

Why the plan runs backwards

The intuitive sequence is to select a platform, configure it, then fill it with data. It produces a system configured against attributes nobody has yet obtained, followed by a reconfiguration once they arrive in a different shape.

Sequencing from the constraint instead means starting with the thing that takes longest and cannot be accelerated. Everything else fits around it.

Months one and two: know what you need

The first phase is short, cheap and unlocks everything after it. It produces one artefact: a list of required attributes, each marked held, partial or missing, with a named owner.

1. Confirm which product groups you place on the market, named as the working plan names them rather than as your merchandising hierarchy does.
2. Extract the required attributes from the delegated act’s data set for each group.
3. Mark each attribute held, partial or missing against what you actually hold today, not what you believe you could obtain.
4. Identify which supplier tier holds each missing attribute, because that determines how long it will take.
5. Flag every claim requiring third-party verification, since those start immediately.

The partial column deserves particular attention. It is where most programmes deceive themselves, because an attribute held for sixty per cent of a range looks like progress and behaves like a gap.

Months two to nine: collection

Collection begins as soon as the gap list exists and runs for most of the remaining window. Structuring it as campaigns rather than a per-product chase is what determines whether it completes.

One campaign covers one attribute set, addressed to the suppliers who hold it, with the regulation named, an example answer per field and an explicit route to report that a value is not measured. A supplier serving eleven of your products receives one request, not eleven.

Run verification in parallel rather than after collection. Laboratory and assessor lead times are independent of your data gathering, and treating verification as a downstream step is how programmes discover in month fourteen that an assurance engagement needs four months.

Months six to twelve: definitions and design

While collection runs, two internal pieces of work proceed that have nothing to do with suppliers and everything to do with whether the resulting data is coherent.

The first is definitional. Two teams using recycled content to mean different things is not a software problem, and it will not be revealed by a platform. It surfaces when two returned figures disagree and nobody can say which is correct.

The second is the access policy: which audience reads which attribute. This has to be settled before publication because it cannot be retrofitted to a passport that is already resolvable, and commercial teams reliably object after launch rather than before.

Months ten to fifteen: build

Platform selection happens here rather than at the start, and by this point it is a considerably better-informed decision. You know the attribute set, the volumes, the issuance level and which claims carry credentials.

Carrier work runs alongside. A printed code is an artwork change on the next production run; anything applied to the product itself is a line change with its own lead time, which is why the carrier decision belongs in the first phase even though the work happens here.

What to do when the window is shorter

Companies frequently discover the obligation late, leaving nine months rather than eighteen. The plan does not change shape; it changes scope.

The move that recovers most ground is narrowing to the products that actually need to be compliant first, rather than attempting the full range. A single product group published on time, with the remainder following, is a materially better position than a whole catalogue that misses the date together.

The second is accepting documented gaps rather than estimates. An attribute recorded as outstanding with a named owner and a date is a programme in progress; the same attribute filled with a plausible figure is a claim that will fail verification later, at a point when correcting it is considerably more expensive and more visible.

Months fifteen to eighteen: publish and rehearse

The final phase is deliberately not a launch. It is a pilot on one product group, followed by rehearsal of the things that will go wrong later.

Rehearse the correction process specifically. A supplier figure will turn out to be wrong after publication, and a programme that has never corrected a live passport will discover its approval chain, its versioning and its audit trail all at once, under time pressure, in front of a customer.