A data matrix grade is a measurement, not an opinion about legibility
Codes are graded on eight measured parameters. A code that scans reliably today can still grade poorly, and that gap is what customers audit.
Manufacturers marking traceability codes often assume that a code which reads on their scanner is acceptable. Customers auditing the process use a different criterion: the graded quality of the code against ISO/IEC standards, measured with calibrated equipment under specified lighting.
What gets measured
Grading assesses several parameters independently, each scored and the lowest determining the overall grade. They include contrast between light and dark modules, modulation, the condition of the fixed pattern that locates the code, axial and grid non-uniformity, unused error correction, and print growth.
Each captures a different failure mode. Contrast covers marking depth or oxide development. Grid non-uniformity covers scanner field distortion and part movement. Unused error correction indicates how much margin remains before damaged modules make the code unreadable.
Why a readable code can grade badly
Modern scanners are tolerant. Reed-Solomon error correction reconstructs data even when a substantial fraction of modules are damaged or misread, so a code can scan perfectly while consuming most of its error correction budget.
That is the crux. A code at grade A has margin: it will still read after handling, wear, partial contamination or a less capable scanner. A code that scans today with no margin left will fail later in the product's life, which is exactly what traceability requirements are meant to prevent.
What causes low grades in practice
Marking parameters. Insufficient contrast, or overheating that blurs module edges.
Surface condition. Reflective or textured surfaces scatter light and reduce measured contrast, sometimes regardless of marking quality.
Field distortion. Codes marked near the edge of an uncorrected scan field show grid non-uniformity.
Part movement. Marking on the fly with imperfect encoder synchronisation distorts module positions.
What production does about it
Inline verification with a grading reader, recording the grade per part rather than sampling. Where grade drifts, the trend is visible before parts fail, and the cause is usually traceable to one of the four factors above.
That converts marking from an operation that either works or does not into a measured process with a control chart — which is what customers requesting evidence actually want.
This article was produced by the LasersNews AI desk and reviewed by our editors.
Related reading
Traceability rules quietly reshape what a marking system has to do
Permanent part identification is no longer a finishing step — it is a data problem that happens to involve a…
UV marking found its place where heat is the enemy
At 355 nm the mark comes from breaking molecular bonds rather than melting. For plastics, glass and thin coat…
Traceability rules are now writing the marking specification
UDI, aerospace part marking and battery passports each demand a code that survives the product's life. Perman…