Error Correction
Error correction is the mathematical redundancy — typically Reed-Solomon coding — built into 2D symbologies that lets a scanner reconstruct missing or damaged portions of the data.
What You Need to Know
QR Code offers four levels (L, M, Q, H) that reserve roughly 7%, 15%, 25%, and 30% of the symbol for recovery data. A QR Code at level H can lose up to 30% of its area — a logo punched through the middle, a torn corner, a scuffed surface — and still decode perfectly. Data Matrix and Aztec apply the same family of algorithms with symbology-specific structures.
Choosing a level is a capacity-versus-robustness trade: higher correction means fewer data cells for the same symbol size, so the symbol grows or holds less. Marketing overlays tempt designers to bury high-correction QR Codes under logos and gradients; this works until print gain and low contrast consume the margin. Error correction never substitutes for contrast and quiet zones — it rescues damage, not design.
Where Error Correction Applies
The dominant 2D matrix barcode: up to 7,089 digits in a square symbol with selectable error correction.
Generate QR Code →A compact 2D matrix code with fixed Reed-Solomon error correction — the standard for pharma serialization and tiny parts.
Generate Data Matrix →A compact matrix code with a bullseye finder and no quiet zone — popular on rail tickets.
Generate Aztec Code →A stacked-linear 2D code carrying up to 1,850 text characters — the workhorse of IDs, boarding passes, and transport documents.
Generate PDF417 →