One Colour, Many Materials

A brand colour is never made once. It is made in an ink, in a dye, in a pigment loading, in a coating — each time by different people, in a different factory, out of different raw material. This is how it survives that, and why the answer has to be spectral.

The designer is not choosing a recipe

Think about where a colour ends up. A folding carton. A shrink sleeve on a bottle. A woven label. A moulded cap. Then, later, a paper flyer, a poster, a screen. Six materials, six technologies, six suppliers — and one colour, which the customer expects to recognise every time.

The designer cannot know that list in advance and should not have to. Their job is the product, not its production. So the colour they choose has to be defined by what it is, not by the ink that made the first sample of it.

The colour is the fixed thing, and the ink is the variable. In production you match the ink, never the colour.

A small reversal that decides everything downstream. A library built around an ink recipe hands every later supplier the wrong question — how do I reproduce this ink? — instead of the one that matters: what do I mix so that my material, under my customer’s light, looks like this?

The five stages a colour passes through, and what each one hands on A designer chooses a ChromaSpot code; the specification travels as CxF; an ink is formulated for each substrate and technology; the ink is drawn down on the production substrate and characterised as CxF/X-4; and production adds metamerism, fluorescence and observer differences. Only the first stage is substrate independent. SUBSTRATE INDEPENDENT A VARIANT IS MADE AT EVERY STAGE BELOW 1 · Choose Designer Product, not press HANDS ON ChromaSpot code 2 · Specify Design file, brief The spectrum travels HANDS ON CxF 3 · Formulate Ink kitchen, dye house One per substrate HANDS ON An ink, and a recipe 4 · Draw down Little Joe, IGT Tints, white and black HANDS ON CxF/X-4 5 · Produce Press floor, shelf More than one light HANDS ON The product every variant answers to the same code

Only the first stage is substrate independent. Everything after it is a variant, made deliberately, for one material and one process — and every one of them answers to the same definition.

What actually has to travel

Between the designer and the factory something has to carry the colour. A name will not do it: names live in one language and one catalogue, and the factory’s software cannot mix a word.

Three numbers — L, a, b — look like they will, and the industry behaved for a long time as though they did. Everything else on this site follows from why they do not.

Lab is not a description of a colour. It is the answer to a question. The question is: if this object were lit by this light and looked at by this observer, what would be seen? Change the light and the answer changes. Change the observer and the answer changes. The object on the table did not move.

That is not a theoretical worry. In this library, swapping the standard observer from the 2° to the 10° — same light, same sample, same instrument — moves the average colour by about 2 ΔE00, which is further than the distance to its nearest neighbour in the book. Moving from daylight to a tungsten lamp moves it by more than three.

A colour recorded as three numbers is a photograph taken under one set of conditions. Hand it to a factory working under another and you have handed them a plausible, precise, wrong target.

The spectrum is the object

A spectral measurement records something else: how much light the surface gives back at each wavelength, from violet through to deep red. It does not assume a light. It does not assume an observer. It is a property of the material.

From it you can compute Lab for any light and any observer — daylight, shop fluorescent, the LED strip over a supermarket shelf — and be right each time. The reverse is impossible: three numbers cannot be turned back into a spectrum, because the information was thrown away when they were written down.

Keep the object, not the answer. Everything ChromaSpot does follows from that one sentence.

What each factory does with it

In production the colour is deliberately redefined, once for every route: a printing ink for a substrate, a dye recipe for a fibre, a pigment loading for a polymer. Each is honest work, because it is the ink together with the substrate that makes the colour.

Two things the designer should know about that step.

The limits are real. Every technology has a set of pigments it can actually use. Some colours cannot be reached in some processes, and no amount of measurement changes that. Where a compromise is needed it is a technological fact, not a failure of the design — and knowing which colours are safe across many processes is exactly what the Ranges filter is for.

The same target can be hit by different recipes, usually several, from different pigments. They agree under the light they were formulated for and part company under another — that is metamerism, the most common way a job passes in the booth and fails on the shelf. It is also the risk an ink kitchen takes when it builds a colour out of leftover ink: good for waste, worth measuring rather than hoping about.

The moment the ink meets the substrate

One step separates a good supplier from a guessing one. Before the run the new ink is drawn down on the production substrate — a Little Joe proof press, an IGT printability tester — and measured as a ladder of tints over white and over black, not one solid patch. The black backing shows how opaque the ink is; the ladder shows how its halftones will behave.

There is an ISO standard for recording exactly that: CxF/X-4. It describes this ink on this substrate — which is precisely the thing ChromaSpot deliberately does not describe. The two are not competitors. ChromaSpot is what you specify before anyone has chosen a material; CxF/X-4 is what you measure once they have. A workflow that uses both has a fixed target and an honest record of how each supplier met it.

And then, the real world

Every ΔE number quietly assumes one observer and one light. Production has neither: the booth is not the shop, brighteners fluoresce under some lights and not others, and two people at the same booth do not see the same colour — the standard observer is an average and real eyes scatter around it.

Each of those effects can be predicted, and only from spectral data. That is why a library that ships spectra is not a specialist luxury: it is the only kind that survives the journey from a screen to six factories and back onto a shelf.

Read next