All articles
Research27. srpna 2026

A tenth of France's reliability score measures nothing

Pavel Kopczyk

We rebuilt the French durability index from the complete register — 2,764 models instead of the 274 we had in July — and separated every sub-criterion. The warranty and quality-process criterion carries a nominal 10% of the reliability score and explains −0.1% of its variance. Delete it entirely and the ranking barely moves.

Following up on a smaller sample

In July we looked at six manufacturers in France's Indice de Durabilité and found that one criterion — commercial warranty and quality process — came back near the floor for every one of them. That was 274 records.

We now hold the complete consolidated register: 2,769 models, of which 2,764 are analysable, covering every washing machine and television legally required to publish a score. With the full set we can ask a sharper question than "is this criterion low?" We can ask what it does to the score.

The answer is close to nothing.

First, a correction to how we read the scale

The July piece reported this criterion as landing between 0.25 and 0.5 "out of 10". That was our misreading of the published files, and it is worth fixing plainly.

The register publishes criterion values in two different units, and does not say which is which. Most of the time a criterion is published as its weighted contribution to its block — so it tops out at its own weight, not at ten. The warranty criterion carries a weight of 0.10, so its contribution can never exceed 1.0 for a television or 0.5 for a washing machine. A published 0.25 is not "0.25 out of 10". On the actual 0–10 scale it is 2.5.

One manufacturer — 366 washing-machine models — publishes the same criteria on the raw 0–10 scale instead. Comparing the two without correction makes that manufacturer look roughly twice as good as its competitors on every criterion, purely as a unit artefact.

The conclusion from July survives the correction. The arithmetic behind it changes, and the corrected reading is what everything below uses.

How the score is built

The index runs 0–10 and splits into two equally weighted blocks — repairability and reliability. The reliability block is itself three criteria:

Criterion What it covers Weight
B1 Resistance to stress and wear 50%
B2 Maintenance and servicing 40%
B3 Commercial durability guarantee + quality process 10%

Those weights are not published as such. We recovered them from the data by two independent routes that agree to three decimal places, and they are identical for both product categories.

What the tenth actually does

If a criterion carries a tenth of the weight, you would expect it to explain something in the neighbourhood of a tenth of the differences between products. Here is what it explains:

B1 wear B2 maintenance B3 guarantee
Washing machines 39.3% 60.6% −0.1%
Televisions 53.4% 45.5% 1.3%

Delete B3 entirely, redistribute its weight to B1 and B2, and compare the resulting ranking to the published one:

Washing machines Televisions
Rank correlation with the published score 0.9988 0.9950
Product comparisons that change direction 0.03% 2.68%
Mean score, published 7.605 7.144
Mean score, B3 removed 8.154 7.636

So B3 does one thing and one thing only: it takes about half a point off everyone's score. Three product comparisons in ten thousand come out differently because of it.

Why: a floor and a ceiling

B3 is built from two sub-criteria. The guarantee duration counts for three quarters of it; the manufacturer's quality process for the remaining quarter.

  • Guarantee duration is zero for 87.5% of washing machines and 97.5% of televisions. Almost nobody offers a commercial durability guarantee beyond the statutory minimum.
  • Quality process is full marks for 92.2% of washing machines and 99.8% of televisions. It is a self-declaration that the manufacturer runs a continuous-improvement process, and essentially everyone declares it.

Combine a floor weighted 0.75 with a ceiling weighted 0.25 and the arithmetic only has one likely answer: 2.5 out of 10. 86.6% of washing machines sit on exactly that value. Across the whole register the quality-process sub-criterion takes just two distinct values.

A criterion where almost every product scores identically cannot separate products, whatever weight the decree assigns it.

The exceptions are real, and there are more than we thought

In July we reported a single record scoring full marks. With the complete register that turns out to be wrong in both directions.

104 models — 3.8% of the register — score 5 or better on this criterion, so the low scores genuinely are a commercial choice rather than a technical impossibility. Eleven Metz Display televisions score a full 10.0. Among washing-machine brands, Miele is the only one whose average sits meaningfully above the floor.

The Easypitch record we cited in July as scoring 10.0 should not have been cited. Its declared reliability block reconciles to neither of the two published scales — the value exceeds the criterion's own ceiling by a factor of ten — so it is a malformed entry rather than an exceptional product. We now flag such records instead of reading them. There are five in the register, plus three placeholder rows with randomised manufacturer names and exemple1.fr URLs.

What this means if you use the index

The practical advice from July stands and gets sharper: read the sub-scores, not the headline number.

Specifically, on the reliability half of the index, only two of the three criteria carry information. Resistance to wear and ease of maintenance are where products actually differ. The warranty criterion tells you something true and useful — that almost no manufacturer will stand behind its product past the legal minimum — but it tells you that about the entire market at once, not about the product in front of you.

That is a finding about policy, not about a washing machine. It belongs in an argument for extending guarantee obligations. It does not belong in a per-model score presented to a shopper as a comparison tool, where it functions as a near-constant deduction that looks like a measurement.

Ten percent of a legally mandated score is not a small thing to spend on a criterion that separates 3.8% of the market from the rest.

Method

Source: the consolidated indice de durabilité files for lave-linge and téléviseur published on data.gouv.fr under Licence Ouverte, retrieved 27 August 2026 — 2,769 models, of which 2,764 are analysed. Criterion weights recovered empirically by two independent methods (maximum observed contribution divided by ten; least squares of the block score on the raw criteria), agreeing to three decimal places. Variance shares are covariance-based and additive across criteria by construction. Scores in the register are banded rather than continuous and are treated as ordinal throughout.

Sources

Decree of 5 April 2024 on durability index criteria for televisions and washing machines (Légifrance); schema.data.gouv.fr (etalab/schema-indice-durabilite); economie.gouv.fr; ADEME. Consolidated manufacturer datasets on data.gouv.fr, retrieved 2026-08-27. Supersedes the sample-based figures in The hidden gap in France's Durability Index.