How to read a fabric shade report

A shade report is the most-circulated and least-read document in garment manufacturing. It arrives as a PDF full of decimals, somebody looks at the pass/fail column, and the order moves on. Then a bulk delivery gets rejected for shading and nobody can explain why the paperwork said it was fine.

Here is what the numbers mean and, more usefully, where they stop being enough.

The colour space, in one paragraph

Colour is measured in a three-dimensional space called CIELAB, because it was designed to be roughly perceptually uniform — a given distance means roughly the same amount of visible difference wherever you are in it.

• L* is lightness. 0 is black, 100 is white.
• a* runs green (negative) to red (positive).
• b* runs blue (negative) to yellow (positive).

A spectrophotometer reads your standard and your batch, and the report gives you the difference on each axis.

Reading the deltas

ΔL — is the batch lighter or darker than the standard? Positive means lighter.

Δa — redder (positive) or greener (negative).

Δb — yellower (positive) or bluer (negative).

ΔE — the total distance between the two colours: one number combining all three. This is the number everyone quotes.

ΔC and ΔH — chroma and hue difference, from the same data expressed in cylindrical coordinates. More useful than Δa/Δb for diagnosis, because they separate how saturated from what hue.

That last distinction matters practically. A batch that is off in chroma usually means a dosing problem — too much or too little dye. A batch that is off in hue means the ratio between dyes in the recipe is wrong. Same ΔE, completely different conversation with the dyehouse.

Which ΔE are you looking at?

This trips people up constantly. There are several formulas, and they do not give the same answer:

• ΔE*ab (CIE 1976) — the original. Simple, and it overstates differences in saturated colours.
• CMC (l:c) — developed by the textile industry precisely because ΔE*ab did not match what dyers saw. Usually quoted as CMC(2:1) for acceptability.
• ΔE 2000 (CIEDE2000) — the current international standard, better aligned with perception across the whole space.

A ΔE*ab of 1.2 and a CMC of 1.2 are not the same claim. If your tolerance is written as “ΔE < 1.0” without saying which formula and which illuminant, it is not really a specification. Fix that in your first tech pack and you save an argument in month six.

Illuminant and metamerism

Every measurement is taken under a specified illuminant — D65 (daylight), TL84 (European retail fluorescent), CWF (US cool white fluorescent), A (incandescent), and so on.

Metamerism is when two samples match under one illuminant and visibly differ under another. It happens when the batch was dyed with a different combination of dyes from the standard — the colours arrive at the same place by different routes, and change lighting exposes the difference.

This is the single most common cause of “but the report said it passed”. The report was measured under D65. The garment was rejected in a shop lit with TL84.

Two protections, both worth insisting on:

1. Specify your illuminants — usually the daylight standard plus whatever your main retail environment uses. A good report shows ΔE under each.
2. Ask for the same dye combination, not just the same shade. A dyehouse that keeps to one recipe across a programme gives you far less metameric risk than one that reformulates each lot to hit the number.

Where the numbers stop being enough

A shade report describes a flat piece of fabric measured under laboratory conditions. A garment is none of those things.

Surface changes what you see. On velour, the pile has direction, and direction changes how light reflects. Two panels within tolerance on paper can look obviously different in a finished jacket because one panel was cut with the nap running the other way. This is a cutting rule, not a dyeing problem — but it will be reported to you as a shading complaint.

Different fabrics in one garment. A terry body with a rib collar means two constructions, two dye behaviours, two shade reports. Both can pass individually while the garment looks wrong, because the eye compares them directly at the neckline and forgives nothing.

Continuity across the order. This is the one that actually causes rejections. Each roll may sit inside tolerance against the standard, while roll 4 and roll 40 sit on opposite edges of it — so they are nearly twice the tolerance apart from each other. Any garment made from both is visibly two-tone.

Which is why shade continuity is a separate requirement from shade approval, and belongs in your specification as its own line. On our floor it is why fabric is shade-banded before cutting and why panels for one garment are bundled from the same band. It costs a little in cutting efficiency and saves an enormous amount in returns.

A practical checklist

Before approving a bulk shade:

• Confirm the formula (CMC 2:1 or ΔE 2000), the tolerance, and the illuminants — in writing, in the tech pack.
• Ask for ΔL, ΔC and ΔH separately, not just ΔE. You want to know the direction of the error.
• Check for metamerism across at least two illuminants.
• Ask for the spread across the lot, not just the average. Highest and lowest ΔE, and how many rolls.
• For pile fabrics, check with the nap in both directions.
• For multi-fabric garments, view the components together under the retail illuminant, not separately.

None of this requires a colour science degree. It requires asking for the right five columns and reading them before, rather than after, the fabric is cut.

---

We dye on soft-flow machines with automated add-tank dosing so recipes repeat lot after lot, and we shade-band before cutting. If you want to see how we handle shade continuity on a long programme, come and look at the floor.