Fifty QR Code Colour Pairs, and Which of Them Survive Print

What the Number in This Table Is, and What It Is Not

A contrast ratio is the relative luminance of the lighter of two colours divided by that of the darker, offset so the scale runs from 1:1 for two identical colours to 21:1 for black on white. It is the same figure WCAG uses for text legibility, and it is computed rather than observed: linearise each sRGB channel, weight them 0.2126 for red, 0.7152 for green and 0.0722 for blue, then divide.

That definition is worth stating precisely, because it is the reason every row below is checkable. Nobody has to trust this table. Two hex values and the formula produce the same number on any machine, which is a different kind of claim from "we tested this and it worked".

Which is also the limit of it. No QR code was printed and scanned to produce this table. The verdict column applies two design thresholds that this site already publishes - around 7:1 for print and around 4:1 for screen - to a computed ratio. It is a filter for deciding which pairs are worth testing, not a substitute for testing them.

The reason those thresholds sit so far above what a decoder needs is covered in the contrast scan test: the readable open-source decoder gives up somewhere near 1.2:1, and everything between there and 7:1 is margin for the things print does to a code after the file is correct.

The Ten Pairs That Clear Print

These reach 7:1 or better and have enough margin left for ink spread, paper reflectance and an awkward light source. Every one of them is a very dark colour on white.

The practical reading of this group is that the printable palette is narrow and deep. Navy, oxblood, wine, deep green, deep purple - these are the colours that work, and a brand whose identity lives at that end of its own palette can print a coloured QR code without thinking about it.

The Eleven Pairs That Belong on a Screen

These land between 4:1 and 7:1. On a display, where contrast is perfect, light is even and edges are hard, they are fine. In print they have spent their margin before the ink is dry.

This is the band where most brand colours actually sit, and it is the band that produces the specific failure of a code that scanned in the studio and did not scan on the shelf. The studio test was on a monitor.

The Thirteen Pairs That Should Not Carry a Code

Below 4:1, a QR code is relying on the decoder's tolerance rather than on being legible. These pairs will sometimes scan, on a good phone, in good light, at a generous size - which is worse than failing outright, because it passes the one test anybody runs.

The pattern across this group is the finding worth taking away. It is not that some hues fail and others pass - it is that lightness is the variable and hue is not. Green appears in all three bands: Deep green passes at 7.68:1, Green 700 is screen-only at 5.02:1, Green 500 fails at 2.28:1. The colour is the same idea three times; only the luminance changed.

Dark Codes on Tinted Stock

Printing on anything other than white lowers the ratio, because the light modules of the code are the paper. Kraft, cream, recycled and coloured stock all start lower than the file suggests, and the file cannot tell you by how much.

The last two rows are the instructive ones. Forest green on white passed at 7.13:1; the same green on a pale green tint drops to 6.81:1 and no longer clears the print threshold. A tint that is barely visible has moved a passing pair into the marginal band.

Light Codes on Dark Grounds Are a Different Question

Inverting a QR code - light modules on a dark background - produces high ratios easily, and the ratio is not the problem with it. The problem is that the contrast ratio formula is symmetric and decoders are not.

A QR code is specified as dark modules on a light background, and while many decoders will try the inverted polarity, not all of them do and none of them promise to. That is a scanner-population question rather than a colour question, and no ratio answers it. If a design calls for an inverted code, the verdict column says test it because testing is the only way to find out.

What the Ratio Does Not Tell You

Four things sit between a passing ratio and a code that scans off paper, and none of them is visible in the file.

• Ink spread. Every press grows dark areas slightly. On a QR code that means the light modules get narrower, which lowers effective contrast at exactly the scale a decoder samples at.
• Substrate. The ratio assumes the background is the colour in the file. On uncoated, recycled or coloured stock it is not, and the paper is doing half the job.
• Finish. A gloss laminate can put a specular highlight across part of the symbol regardless of how good the underlying pair is. A 21:1 code under a reflection is a 1:1 code in that patch.
• Camera exposure. A phone meters the whole scene. A small code on a large bright surface gets exposed for the surface, and the code arrives at the decoder flatter than it left the printer.

This is why the verdict column is worded as a threshold rather than a promise. Clearing 7:1 means the pair is not the reason the code fails; it does not mean the code will not fail.

How to Check a Pair That Is Not in This Table

Most brand palettes contain a colour that is not one of the fifty. The check takes seconds and is the same one used to build the table.

1. Get the Ratio for the Two Hex Values. Put the code colour and the background into the QR contrast checker. It returns the same figure this table uses, for any pair.
2. Compare It Against the Job, Not Against a Single Number. Around 7:1 for anything printed, around 4:1 for a code that will only ever be shown on a screen. Under 4:1, change the colour rather than the size.
3. If It Fails, Darken Rather Than Re-Hue. The failures in this table are a lightness problem. A brand orange that fails at 2.80:1 usually has a darker sibling in the same palette that clears 7:1 and still reads as the brand.
4. Measure the Exported File, Not the Swatch. The QR code validator reads the finished export and reports the contrast it actually contains, along with the quiet zone and module size. A gradient or a soft background changes the answer the swatches gave.
5. Run the Ladder Before a Large Print Run. The contrast ladder in the contrast scan test produces a threshold for your own phones and your own stock, which is the only number that describes your job rather than the general case.

For the design side of the same decision - which parts of a code can take colour at all, and where an accent belongs - coloured QR codes covers the placement rather than the arithmetic.