Garment Durability as a DPP Attribute

Durability has been claimed by clothing brands for as long as clothing has been sold, almost always without a number attached. ESPR changes that by making it a declared parameter with a defined method behind it.

The shift is smaller than it sounds technically and larger than it sounds organisationally. The tests already exist. What is new is that results become a public assertion rather than an internal quality gate.

What durability actually decomposes into

There is no single durability figure for a garment. The parameter decomposes into several properties, each failing differently and each mattering to a different kind of product.

The last row is worth dwelling on. A significant proportion of garments are discarded because a zip failed on an otherwise sound item, which means fastening durability has an outsized influence on actual product lifetime relative to the attention it usually receives.

Appearance failure ends more garments than structural failure

Consumer research consistently finds that clothing is discarded because it looks worn rather than because it has stopped functioning. That has a direct consequence for which properties matter.

A knitwear item that pills after five wears is finished as far as its owner is concerned, however sound its seams. A dark cotton item that fades unevenly is finished for the same reason. Both would pass a purely structural durability assessment.

Where the data comes from

Most brands selling at any scale already hold much of this data, because retailers and their own quality functions have required it for years. The problem is rarely that the tests were not run.

• Results sit with the supplier or the test house, not in a brand system, and are retrieved by email when someone asks.
• They are recorded as pass or fail against an internal specification rather than as the measured value the passport needs.
• They are held per production run rather than per product, so linking a result to a specific item requires a lot join nobody maintains.
• They expire and are re-tested without version history, so the value in force at a given date cannot be reconstructed.

The second point causes the most rework. A specification saying colourfastness must reach at least grade 4 produces a record saying "pass", and a passport attribute asking for the grade cannot be populated from it. The underlying test report has the number; the system that stored the conclusion discarded it.

Declaring a figure is a verifiable claim

A durability declaration differs from most marketing claims in being reproducible. An authority can buy the product, run the named test method and compare the result.

That makes the declared value a compliance exposure in a way a phrase like "built to last" never was. Regulation (EU) 2019/1020 gives market surveillance authorities the powers to do exactly this, and Directive (EU) 2024/825 removes the room for unsubstantiated durability messaging alongside it.

The practical implication is that declared figures should reflect what production actually achieves rather than what a golden sample achieved. A value taken from a development sample and never re-verified against bulk production is a claim waiting to be falsified.

Durability and the rest of the passport

Durability does not sit alone. It interacts with reparability and with the fibre composition data the same passport carries, and the three together describe how long a garment can remain in use.

A highly durable garment that cannot be repaired when it eventually fails, made from a blend that cannot be recycled, performs worse across a full lifecycle than a moderately durable one that is repairable and recyclable. The passport makes that trade-off visible for the first time.

What to do first

Start by finding out what testing already happens and where the results live, because the answer is usually that more testing happens than anyone at the brand can quickly enumerate.

Then change the specification so suppliers return measured values rather than pass or fail verdicts, and store them against the product rather than the production run. Those two changes convert existing testing into passport-ready data without adding a single new test.