Technology
Variable Unique QR Code Printing: Six Problems on the Line and How to Fix Them
Industrial label printer and vision system checking variable unique codes on a packaging line

A unique QR code on every unit is easy to specify and hard to run. On a line moving at speed, the code has to be generated, printed, read back and tied to a record before the pack moves on, and a miss at any step produces stock that looks fine and cannot be verified.

Most failures are not in the design of the code. They sit in the print, the data, or the hand-off between the two.

TL;DR
  • A QR code is a data carrier. What makes it unique is the identifier behind it and the record that identifier points to.
  • Most line problems come down to print grade, duplicate or missing identifiers, and codes that were never read back after printing.

Unique means the record is unique, not the picture

Two QR codes can look completely different and carry the same identifier. Two identical-looking codes can carry different ones. The pattern on the pack is only the carrier: ISO/IEC 18004 defines the QR Code symbology, including its data encoding, error correction and production quality requirements, and says nothing about whether a given code is issued once.

Uniqueness comes from the system that assigns each identifier and the record it points to. That is why a printing problem is often a data problem in disguise. The same thinking sits behind variable data labels, where every label on a roll differs. Regulators are also specifying QR codes on retail packs, as with household insecticides under the 2025 amendment, so the data behind each code matters as much as the print.

Six problems on the line

1. Print quality that changes with the surface

A code that scans on a coated carton may fail on a curved bottle or a rough sleeve. Contrast, edge sharpness and the blank border around the code (the quiet zone) all move with the surface and the print method.

Print quality for 2D barcodes is graded under ISO/IEC 15415, on a scale from 0.0 to 4.0, which is often shown as grades F to A. The overall grade is the lowest grade among the individual parameters, such as symbol contrast, modulation and fixed pattern damage, so one weak parameter sets the result. GS1 sets 1.5 as the minimum overall grade for 2D barcodes in retail applications, as its 2D barcode quality guide explains. Authentication codes are not always GS1 codes, but the same grading gives a shared way to say a print is good enough.

2. Keeping pace with the line

A printer that is accurate at low speed can smear or misplace codes at full speed. Misalignment and timing errors then show up as rework further down the line. Trigger the print from product detection rather than a fixed timer, and test at the real line speed, not the bench speed. Many of the usual culprits are covered in our note on common packaging line problems.

3. Duplicate or missing identifiers

This is the data side. Millions of identifiers have to be generated without repeats, delivered to the printer in the right order, and matched to what was actually printed. The risk is a gap between the two. A printer jam can leave identifiers issued but never applied, and a reprint can leave one applied twice.

Treat an identifier as live only once it has been printed and verified, and retire any that were rejected so they cannot be reused. If cases and pallets carry their own codes too, the parent-child links need the same care; see the glossary entry on parent-child serialisation.

Production and warehouse team in high-visibility vests standing between pallets in a distribution centre

4. Older printing equipment

Legacy printers can struggle with variable data: slow data handling, fading, smudging or drift. The only reliable test is a trial run on the real material, graded against the standard, before volumes scale.

5. Codes that print but do not scan

A code can look sharp and still fail on a phone, a handheld scanner or an older device. Read every code back immediately after printing and compare what was read with the identifier that was meant to be printed. A mismatch should stop that unit, trigger a reprint, and mark the failed identifier as void. Checking later, once packs are in cartons, makes every fix more expensive.

6. Cost

Better printers, verifiers and label materials add cost up front. Set that against rejects, rework and unscannable stock rather than against the cost of a plain label, and start with one line before committing the rest.

Choosing how to print: the trade-offs

There is no single right method. The comparison below is analysis of how each approach usually behaves, so test it against your own substrate and speed.

ApproachWhere it suitsWatch for
Thermal transfer labelsFlat, stable surfaces where a label can be applied after fillingLabel placement errors and a separate application step
Inline printing on the packLines that mark cartons or containers directly at speedSubstrate variation and tight timing with product detection
Pre-printed rollsPackaging made off-site and supplied serialisedKeeping the supplier’s identifier file matched to what arrives

Whichever you pick, a verifier at the end of the print step is the control that matters most, because it tests the code you actually produced.

A four-step control for every unit

Whatever the print method, the line control can follow the same sequence. Print the code. Grade it with a verifier against the target. Read it back and compare it with the identifier that was meant to be printed. Then release the unit, or void the identifier and reprint. Each step is cheap on its own, and together they stop the two failures that cause most field problems: a code that does not scan, and an identifier that does not match its record.

Check the code before it leaves the line

Talk through your line, your substrate and your identifier flow with the Acviss team.

Book a demo

The people side of the line

Variable printing crosses three teams that rarely sit together: production, which owns the printers, IT, which owns the identifier data, and quality, which owns the grade and the reject decision. Programmes tend to stall in the gaps between them.

A common procurement mistake, in our analysis, is buying printers on speed and price without a verifier, then discovering the failure rate only when codes reach the field. Another is leaving the identifier file with IT and the printer with production, so nobody can say which identifiers were applied.

Agree in advance who can stop the line on a failed read, who retires rejected identifiers, and who signs off a change of printer or label stock. Operators also need to recognise a degrading print early, before a whole run is affected.

A code that scans is not a code that proves

Once printing works, a separate question starts. The EUIPO’s anti-counterfeiting technology guide says QR codes by themselves do not offer protection against copies. They can be combined with other authentication technologies, such as holograms, copy detection patterns or unique identifiers, to give automated authentication through a smartphone scan.

A copied code that points to a real identifier will still return a real answer. Spotting that takes review of the scan activity behind it, which is the gap between copy detection and copy prevention. Our piece on product authentication technology covers how the layers fit together.

Where Origin fits on the print line

Origin is Acviss’s supply-chain track-and-trace solution. On the print side, the team configures its products, identifier levels and supply-chain nodes, then generates and commissions serialised identities for the line. Identities can be aggregated into cases and pallets where needed, and scan events are captured at production and at distribution hand-offs, so the identifier printed on the line stays connected to what happens to the pack afterwards.

A scan history shows custody and movement. It does not prove, by itself, that the physical item is genuine, so unusual activity is a lead to investigate rather than a verdict.

The stakes differ by category. Pharma has the strictest rules, and serialisation and aggregation requirements differ between markets, as set out in our guide to FMD and DSCSA aggregation. Food, FMCG and electronics tend to use unique codes for traceability, engagement and parts control. The print problems are the same in each. What differs is what a failed read costs.

Common questions

Does every QR code need to be unique?

Only if you need to tell units apart. A batch-level code is simpler, but it cannot flag a duplicate scan of one unit.

What print grade is good enough?

For retail 2D barcodes, GS1 sets an overall grade of 1.5 as the minimum, but a higher internal target leaves headroom for handling and ageing.

Should the read-back happen at the printer or later?

At the printer. A failed code caught there costs one label. Caught after cartoning, it can cost a pallet.

Start with the read-back

If one change is possible, make it the read-back. Printing is the visible part of the job. Verifying each code against the identifier you meant to print is what makes the rest of the programme trustworthy.

The reason will grow. As retail moves towards 2D codes, covered in our note on the GS1 Digital Link, more partners will scan the same symbol, and a weak print will be found by more readers, in more places, than before. Brands that treat print grade as a governed control now will not have to rework it under pressure later.

Interested to learn more? Get in touch with us

Talk to Acviss about running verified unique identifiers on your production line.

Talk to Acviss
Home » Blog » Variable Unique QR Code Printing: Six Problems on the Line and How to Fix Them
More like this on:Technology
author-avatar

About Arun Krishnan

Arun is a storyteller at heart, with a knack for making complex ideas click. He works at the intersection of technology, content, and communication, turning technical jargon into stories people actually want to read.

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *