Free QR code tools and business calculators

Why a QR code that looks fine will not scan

A QR code either decodes or it does not, and the ones that fail usually look perfect. We build the generators on this site, so we test our own output by decoding the image files rather than pointing a phone at the screen. Three causes account for almost every failure we have had to fix, and none of them are visible to the eye.

The three corner squares are not decoration

Those large squares are finder patterns, and a scanner uses them to work out where the code is, how big it is and how it is rotated. It finds them by scanning a line across the image and looking for a run of dark and light in the ratio 1:1:3:1:1. That ratio holds at any angle, which is exactly why the format uses it.

Style those corners and the ratio breaks. Round them into dots and the dark runs measure short. Let them take color from a logo and one part of the run comes out lighter than the rest. The scanner then never locates the symbol, so it never gets as far as reading your data. The code is not damaged, it is invisible. Worse, partial styling fails intermittently rather than cleanly, which is how a code passes one test on your own phone and then fails in print. If you are styling a code, leave the three corners solid and dark and style everything else.

Scanners read brightness, not color

A decoder converts the image to brightness before it does anything else, then decides which cells are dark and which are light. Hue is discarded at that first step. This is why a rich orange and a mid grey can look completely different to you and land at nearly the same brightness, at which point the scanner treats them as the same cell.

It also explains the most counterintuitive result we measured. A white code on a near black background scores about 18.9:1 on a standard contrast check, far beyond anything a readability guideline asks for, and it still fails on a large share of phones. Contrast was never the problem. Most decoders assume dark modules on a light background and never try the inverse, so an inverted code is not a low contrast code, it is a code half your audience cannot read. The background checker in our transparent QR generator rejects inverted pairs outright for that reason. The full write up of what we tested covers the method and its limits.

Error correction is a repair budget, not a logo allowance

QR codes carry redundant data so they still read when part of the symbol is damaged. ISO/IEC 18004 defines four levels, recovering roughly 7% at L, 15% at M, 25% at Q and 30% at H. The common mistake is reading "level H allows 30% damage" as "I can cover 30% with my logo".

The budget is shared. A scuff on the print, a fold, glare, a slightly out of focus camera and your logo all draw from the same 30%. Spend it all on branding and there is nothing left for the real world, which is where the code has to work. In practice the usable ceiling sits nearer 25% of the code area at level H, and below about 22% codes survive being resized, printed small and photographed at an angle. Raising the correction level is not free either, because the redundancy makes the symbol denser and a denser symbol has to be printed larger to stay readable.

One habit prevents most of this. Decode the actual file before it goes to print, rather than scanning your monitor with a modern phone that corrects for a great deal a cheap handheld scanner will not. If a code is already failing and styling is not the cause, our list of fixes for a QR code that will not scan covers the print and camera side, and the QR code decoder will tell you whether the file itself still reads.