TL;DR: Artwork validation failures caught after plate-making cost 3–5× more to fix than those caught at PDF preflight — the protocol order matters as much as the protocol itself.
TL;DR: Our internal AF-QC02 artwork release checklist flags 23 discrete pass/fail criteria before any file is forwarded to prepress, reducing plate remake rate to under 4% across our folding carton lines.
What Actually Gets Tested — and in What Sequence #
Most artwork QC discussions focus on what to check. The sequence is what determines whether defects are caught cheaply or expensively.
Our AF-QC02 release protocol runs in four gates. Gate 1 is automated preflight at file receipt — before any human review. We use a rules-based preflight profile that checks minimum image resolution (300 dpi at 100% reproduction size for process print, 1200 dpi for barcode elements), PDF/X-4 or PDF/X-1a compliance per ISO 15930-4, embedded versus linked font status, overprint settings on black text below 14pt, and bleed confirmation at ≥3mm on all live edges.
Gate 1 failures are returned to the brand partner within 4 business hours with a structured error report. No human prepress time is spent on a file that hasn’t passed Gate 1.
Gate 2 is colour space and separation validation. This runs after Gate 1 clearance. Spot colour calls are cross-referenced against our active Pantone Matching System library — we work from the current Pantone GP1601B coated guide, refreshed annually. Any spot colour with a delta-E greater than 2.0 versus the declared Pantone reference on our press substrate triggers a query to the brand partner before proceeding. RGB objects remaining in a CMYK-designated file fail Gate 2 automatically, regardless of whether they visually appear correct on screen.
Gate 3 is structural registration validation — overlaying artwork against the confirmed dieline in our prepress software. Knockout gaps, bleed encroachment into structural fold zones, and text or logo elements within the 4mm safe zone boundary all generate flagged reports. We measure fold-zone clearance to ±0.5mm because our sheet-fed offset presses hold a registration tolerance of ±0.2mm but downstream diecutting adds up to ±0.3mm cumulative variance.
Gate 4 is a physical colour proof sign-off against ISO 12647-2 for offset print, using a calibrated contract proof output on substrate that matches the target stock. Our proof devices are calibrated to G7 Grayscale methodology, certified annually by an independent G7 Expert.
Asking Suppliers the Right Questions About Their Validation Process #
When you’re evaluating a packaging supplier’s artwork QC capability, ask for their preflight profile export — not just a statement that they run preflight. A credible supplier can export the rules file from their preflight tool (Enfocus PitStop, Callas pdfToolbox, or equivalent) and show you exactly which parameters are checked at what thresholds.
Ask specifically: “What is your minimum acceptable image resolution for flexographic print at 133 lpi screen ruling?” A supplier running flexo at 133 lpi should specify 300–350 dpi at final size. If they cite a single blanket figure without referencing screen ruling, that gap in their answer tells you their preflight rules aren’t calibrated to their actual press configuration.
Ask about their colour measurement equipment and calibration schedule. Spectrophotometers used for proof verification should be calibrated per ASTM E308 illuminant conditions, with instrument calibration logged at minimum daily before first measurement. We log calibration under our internal Form QI-11 before every proof verification session. A supplier who can’t tell you their calibration interval, or who conflates screen-based soft-proofing with a physical calibrated proof for colour-critical sign-off, is not running a print-production-grade validation workflow.
Response time to your question matters here too. A supplier with a documented protocol can answer these questions within one business day. Delays or vague answers suggest the process exists informally rather than as a controlled procedure.
Resolution, Proof Format, and the Cost-Performance Trade-off #
Contract proofing on calibrated output is the highest-fidelity validation method before plates are made. It also adds cost and 1–2 working days to the artwork approval cycle.
For commodity packaging — secondary cartons, plain shipper boxes, single-colour utility labels — a calibrated soft proof on a colour-managed monitor, verified against an ICC profile matched to the target press and substrate, is a reasonable alternative. The colour accuracy of a well-managed soft proof on a wide-gamut display is measurably close to a physical contract proof for solid spot colour work. The risk increases with complex photographic imagery, skin tones, or gradient-heavy brand assets.
| Validation Method | Colour Accuracy | Cost Relative to Plate-Make | Best Application |
|---|---|---|---|
| Automated preflight only | No colour validation | Negligible | Repeat/reorder jobs, no colour changes |
| Calibrated soft proof (ICC) | ±2–4 delta-E typical | Low | Commodity cartons, spot-colour-only jobs |
| Physical contract proof (G7) | ±1.0–1.5 delta-E | Moderate | Premium brand packaging, new colour builds |
| Press proof on production substrate | ±0.5 delta-E | High (full setup cost) | Brand-new substrates, critical metallic/special ink |
Colour accuracy ranges are based on controlled measurement conditions; real-world variance increases with substrate optical brightener content and ambient lighting at sign-off.
The counterargument for physical proofing: if a job repeats unchanged across 8+ production runs, the validated press profile from the first run makes physical reproofing redundant. We hold approved press profiles for active repeat SKUs for 24 months. If a brand partner returns within that window with identical substrate and ink specification, we can release to plates with Gate 1–3 validation only and skip the physical proof cycle — saving 1.5 working days on average.
Barcode Validation: The Specification Most Artwork Briefs Under-Specify #
Barcode verification is where we see the largest proportion of correctable errors arriving in files — and the highest downstream cost if those errors reach print.
A scannable-looking barcode is not a verified barcode. Print quality verification per ISO/IEC 15416 (linear barcodes) or ISO/IEC 15415 (2D symbols) grades barcode print quality on a scale of A (4.0) to F (0.0). Most retail and logistics requirements specify a minimum grade of 1.5 (grade C) for GS1-128 shipping labels, while direct-to-consumer product barcodes for grocery and retail typically require grade 2.5 (B) or above.
The resolution requirement for barcode elements is not the same as for photographic imagery. A standard EAN-13 at 100% magnification (37.29mm × 26.26mm at nominal) requires the barcode module width to be reproduced at the defined X-dimension — for EAN-13, the nominal X-dimension is 0.330mm, with a tolerance of ±0.101mm per GS1 General Specifications. At 1200 dpi image resolution, each pixel represents approximately 0.021mm, giving adequate resolution to hold that tolerance. At 300 dpi, each pixel is 0.085mm — technically sufficient but with no margin for any RIP rounding error or plate dot gain. Our spec for barcodes embedded in artwork files is 1200 dpi minimum as a vector-originated rasterisation or as native vector, never as a 300 dpi raster element.
We run verification scans on every barcode-bearing job using a Zebra DS9908 calibrated to ISO/IEC 15416 measurement aperture conditions. Verification reports are archived by job number and available to brand partners on request. If a barcode grades below 2.0 on first press proof sheets, we quarantine that press sheet batch and re-examine plate exposure before approving production run continuation.
One area we’re still tracking: how UV-cured inkjet variable data printing on top of offset-printed barcodes interacts with verification grade under different scanner illuminant conditions. Our dataset covers 11 SKUs across 14 months — not enough volume yet to set a firm internal spec. We’ll have a better baseline by end of Q3 2025.
Specification Notes for Brand Partners #
When you brief us on a new artwork file, the minimum information we need to open a job in our AF-QC02 workflow: confirmed PDF/X standard (X-1a or X-4), declared Pantone references with coated or uncoated suffix, substrate specification (for colour profile matching), and final dieline version number. Without the dieline version, Gate 3 cannot run — we’ve had samples go through three iterations because the artwork was registered to an earlier dieline revision.
The most common brief gap we see is undeclared special finishes in the artwork file. If a spot UV varnish, foil, or emboss area is represented as a Pantone spot colour rather than a correctly named and separated fifth layer, Gate 2 will flag it as a colour match query rather than a structural finish, which routes it through the wrong approval path. Name your finish layers clearly: “SpotUV_Flood”, “HotFoil_Gold”, “Emboss_Logo”. This single convention saves one full sample iteration in our experience.
Our standard artwork review and sample development timeline is 10–12 working days from approved artwork to first physical sample, assuming Gate 1 clears within 48 hours of file submission. Complex jobs with multiple special finishes run 14–16 working days. Rush processing at 7 working days is available for orders above 10,000 units with a confirmed PO.
What is your minimum image resolution requirement for packaging artwork?
300 dpi at 100% final reproduction size for all photographic or tonal imagery in process print. Barcode elements require 1200 dpi or native vector. Below 300 dpi on tonal imagery, dot gain at typical offset screen rulings (150 lpi) will visibly degrade highlight detail.
Do you require a physical contract proof for every job?
No. Repeat orders with no colour changes and an active press profile on file can release to plates after Gate 1–3 automated checks only. Physical G7-calibrated contract proofs are required for new SKUs, new substrates, or any job where photographic imagery or skin tone accuracy is brand-critical.
What barcode grade do you verify against, and what happens if it fails?
We verify against ISO/IEC 15416 and target a minimum grade of 2.5 for retail product barcodes. If press proof sheets grade below 2.0, we hold the batch and investigate plate exposure before continuing. A barcode that passes in the file but fails at press verification almost always traces back to plate dot gain compounding on a barcode that was specified at minimum magnification — we typically recommend 110–115% magnification for retail EAN/UPC to build in tolerance.
How do you handle Pantone colours that have been discontinued or reformulated?
We maintain a current Pantone GP1601B coated reference guide refreshed annually. If a brand partner specifies a Pantone number that has been reformulated in recent editions, we flag the delta-E shift versus our current guide and provide a delta-E measurement between old and new formula before proceeding. For shifts above 2.0, we request brand confirmation before mixing ink. For shifts below 1.0, we proceed under standard tolerance.
What file format do you accept for artwork submission?
PDF/X-1a or PDF/X-4 per ISO 15930 are our preferred submission formats. Native files (AI, INDD) are accepted for edit purposes only, not as production-release artwork. All fonts must be embedded. Linked images must be flattened into the PDF at full resolution before submission — we do not accept PDFs with low-resolution previews substituting for placed images.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
