Why your brand colour looks wrong in print

Screen colour and printed colour are two different systems, and most brand documents only specify one of them. What goes wrong, and the four numbers that stop it.

Almost every brand I rebuild arrives with the same unopened problem: the colours were chosen on a screen and have never been specified for anything else. They look right in the logo file, right on the website, and then a box arrives from the printer and the green is a different green.

This is not a printer error and it is not a mistake in the design. It is two different colour systems being asked to describe the same idea, and only one of them was ever written down.

Screens add light, print subtracts it

A screen starts black and adds red, green and blue light until it reaches the colour you asked for. Push all three to full and you get white. That is why screen colour can be so intense — it is emitted, not reflected.

Print starts with a white sheet and subtracts light using cyan, magenta, yellow and black ink. Every layer of ink removes something. Push all four to full and you get a muddy near-black and a sheet too wet to run. The brightest thing a printed page can ever be is the paper itself.

So the two systems do not have the same reach. RGB can describe colours that CMYK physically cannot make. The ones that fall outside are predictable: saturated greens, electric cyans, pure oranges and anything neon. Those are exactly the colours that look best on a website, which is why the problem is so common.

What actually goes wrong

Three failures account for most of it.

Nobody specified the CMYK build

The brand document lists a hex value and stops there. The printer, needing four numbers, converts the hex automatically — and an automatic conversion optimises for mathematical closeness, not for what the colour is supposed to communicate. A bright brand green converted by default often lands duller and slightly blue. Nobody chose that. It was just the last decision in the chain, made by software.

Small type printed in rich black

Rich black — a black built from all four inks — looks deep and expensive in a large area. In 8pt type it is a liability. Four plates have to register perfectly; if any one of them shifts by a fraction of a millimetre, the type gains a coloured fringe and looks blurred. Small type should be 100% K and nothing else. Large solid areas can use a rich build. The rule is set by size, not by taste.

The stock was never part of the decision

The same build prints differently on coated board, uncoated board, kraft and film. Uncoated stock absorbs ink, so the dot spreads and mid-tones print heavier and darker than the proof suggests. A palette that was approved on a glossy proof and then run on uncoated card will come back looking like somebody turned the contrast up.

The fix is a document, not a trick

There is no clever conversion that makes this go away. The fix is that every brand colour is specified in every system it will ever be used in, and that specification travels with the logo files.

For each colour, that means four things:

  • The hex and RGB value, for screen.
  • A CMYK build that has been chosen, not auto-converted — adjusted by eye against a printed reference, so it reads as the same colour rather than measuring as the closest one.
  • A Pantone reference, for the jobs where colour accuracy matters enough to justify a spot ink.
  • A note on where the build changes — if the uncoated version needs different numbers, those numbers are in the document too.

When to pay for a spot colour

A Pantone spot ink is a pre-mixed ink rather than a four-colour simulation. It is consistent across runs, it can reach colours CMYK cannot, and it costs more because it is an extra plate.

It is worth it when the colour is the brand — when somebody would recognise the company from the colour alone at a distance. It is worth it on packaging that sits next to competitors. It is usually not worth it on a one-off flyer, and it is irrelevant for digital press runs, which simulate spot colours in CMYK anyway.

A practical middle path: specify the Pantone so the option exists, and specify a CMYK build that gets acceptably close so the job can run four-colour when the budget says so. The decision then belongs to the person buying the print, which is where it should sit.

Proof before you run

If colour matters on a job, ask for a proof on the actual stock, and sign it off in daylight rather than under an office fluorescent. A screen proof tells you the layout is correct. It tells you very little about the colour, because you are looking at emitted light approximating reflected ink.

None of this is difficult. It is just a step that gets skipped because the brand looked finished the moment the logo did. Specifying colour properly costs an afternoon once, and it stops the same conversation happening on every print job for the life of the brand.

If you have a brand whose colours have never been specified for print, the identity work I do includes a full production sheet as standard, and the packaging pages cover what that means on a physical pack.