Why Do QR Codes Still Work When Partly Damaged? The black-and-white squares do not store only the bare minimum of information. A QR code also contains carefully calculated error-correction data that can help a scanner rebuild pieces that are missing or unreadable.
That is why a code may survive a scratch, a fold, a splash of coffee, or even a small logo placed over the middle. It looks like digital sorcery, but it is really organised redundancy and some very clever mathematics.
A QR Code Contains More Than the Message
A QR code divides information into small units called codewords. Some codewords represent the original data, such as a web address. Others are added specifically to help detect and correct errors.
The added information is produced using Reed–Solomon error correction. Instead of simply making several identical copies of the message, the system calculates extra values from the original data. If some codewords cannot be read, the scanner can use the surviving data and those correction values to work out what is missing.
A rough analogy is receiving a sentence with several letters obscured, plus a set of clues about the words. Enough clues let you reconstruct the lost letters without having seen a second complete copy.
The Four Error-Correction Levels
QR codes can be created with four standard error-correction levels: L, M, Q, and H. According to DENSO WAVE, the company behind the QR code, these can restore roughly 7%, 15%, 25%, and 30% of codewords respectively.
Those percentages are useful guides, not promises that any random chunk of the image can disappear. Damage concentrated in crucial areas may cause failure sooner than scattered marks. The scanner also needs enough of the code to locate, orient, and interpret the grid correctly.
Higher correction levels use more space. If the same message is encoded at level H instead of level L, the pattern usually becomes denser or the QR code must become physically larger. Resilience is bought with additional squares.
What the Big Corner Squares Do
The three large square targets in the corners are finder patterns. They help a camera locate the code and determine its position and rotation. Smaller alignment patterns help compensate when the code is viewed at an angle or printed on a surface that is not perfectly flat.
Timing patterns tell the scanner how the grid is spaced. Other areas record the format and version information needed to decode it. These structures explain why a QR code cannot be treated as a completely uniform patchwork: some squares perform particular jobs.
If a large finder pattern is destroyed, a scanner may struggle to recognise that a QR code is present at all, even if much of the data area remains intact.
Why Logos Can Sit in the Middle
Branded QR codes often place a logo over the centre. This deliberately hides part of the symbol, but designers choose a sufficiently high error-correction level and keep the obstruction within a safe area.
It is still possible to overdo it. A giant logo, low contrast, rounded decorative modules, or a crowded background can consume the code’s safety margin. A code that works on the designer’s large monitor may fail when printed small, viewed through a scratched camera lens, or scanned in poor light.
Damage Is Not the Only Problem
Error correction can recover lost codewords, but it cannot rescue every bad design. QR codes need a clear quiet zone around the outside so a scanner can separate the symbol from its surroundings. Strong contrast helps, and the complete code must occupy enough pixels in the camera image.
Glare, motion blur, extreme perspective, tiny printing, and patterned backgrounds may prevent the scanner from identifying individual modules. In those cases, the data may technically be present but the camera cannot capture it cleanly enough to begin decoding.
Why More Damage Sometimes Still Scans
Two codes with similar-looking damage can behave very differently. One may use a higher error-correction level. One scratch may cover a relatively forgiving data area, while another removes a finder pattern or wipes out too many codewords from the same region.
Different camera apps also vary in how well they correct perspective, sharpen an image, and try alternative interpretations. Successful scanning therefore depends on the encoded resilience, the location of the damage, and the quality of the captured image.
The Bottom Line
QR codes survive partial damage because they include mathematical recovery information alongside the original message. Reed–Solomon error correction can reconstruct a limited amount of missing data, while finder and alignment patterns help the scanner understand the symbol. It is robust engineering, not an indestructibility cheat code.
FAQ
How much of a QR code can be damaged?
The nominal recovery ranges are about 7% to 30% of codewords, depending on the chosen correction level. Actual success also depends on where the damage occurs.
Why do some damaged QR codes fail immediately?
Damage to finder patterns, timing information, or a concentrated block of codewords can be more serious than scattered marks across ordinary data areas.
Does a QR code need its white border?
Yes. The quiet zone helps the scanner distinguish the code from nearby images and text. Cropping it too closely can make an otherwise perfect code unreliable.
Sources
Error correction feature :: https://www.qrcode.com/en/about/error_correction.html
What is a QR Code? :: https://www.qrcode.com/en/about/
QR Code Standardization :: https://www.qrcode.com/en/about/standards.html