TL;DR: A supplier’s COA is only as useful as the test methods behind it — if they can’t tell you which standard each value was measured against, the document is decorative.
TL;DR: In our incoming inspection protocol, any lot where ECT falls below 126 kN/m (the minimum for a standard C-flute RSC rated for 25 kg gross weight) is quarantined immediately, regardless of what the COA states.
COA Field Requirements That Actually Predict Shipping Performance #
When we qualify a new corrugated carton supplier, the first thing we ask for is a blank COA template before we ever see a filled one. The template tells us immediately whether the supplier is running a real quality system or printing numbers to satisfy a checklist.
A compliant COA for export corrugated cartons must include — at minimum — the following measured parameters, each linked to a named test standard:
| COA Field | Minimum Acceptable Value (C-flute, 25 kg gross) | Test Standard |
|---|---|---|
| Edge Crush Test (ECT) | ≥ 126 kN/m | TAPPI T 811 / GB/T 6546 |
| Bursting Strength | ≥ 1,200 kPa (Mullen) | TAPPI T 810 / GB/T 6545 |
| Box Compression Strength (BCT) | ≥ 3.2 kN (empty, standard stack) | ASTM D642 / GB/T 4857.3 |
| Board Moisture Content | 8–12% | TAPPI T 412 |
| Flute Profile Height (C-flute) | 3.5–3.8 mm | ISO 4046 |
| Basis Weight of Liners (outer/inner) | Per spec (e.g., 200/200 gsm) | TAPPI T 410 |
A COA that lists “BCT: pass” with no numeric value is not a COA. We’ve returned documents like that to suppliers at least six times in the past two years. What you want to see is the actual measured value, the date of test, the batch/lot number, and the name of the tester or QC signatory. If a supplier is using an accredited third-party lab — CNAS-certified in China, or ISO 17025-accredited elsewhere — that should be stated explicitly on the document header.
One field that regularly goes missing on COAs from smaller mills: moisture content at time of shipment. Corrugated board absorbs moisture in transit, particularly on sea freight routes through humid climates (Southeast Asia, Gulf of Mexico ports). A board shipped at 10% moisture arriving at 15–16% will lose roughly 25–30% of its ECT. That’s not a transit problem. That’s a specification gap that needed to be addressed before loading.
We specify moisture content limits in all our export carton purchase orders, and our incoming inspection sheet — what we call the IQ-03 Carton Acceptance Record — flags any lot above 13% for conditional hold pending retest.
What Goes Wrong When Supplier Qualification Is Skipped #
The three failure scenarios we see most often are all traceable to the same root cause: a brand team accepted a price quote without running a qualification round, then treated the first production lot as implicitly approved.
The first scenario involves board substitution after initial sampling. A supplier submits a sample carton using 200/200 gsm double-wall Kraft liner. The brand approves the sample, places a production order, and the factory ships using 180/180 gsm liners to recover margin. The BCT differential between these two constructions is measurable — roughly 0.4–0.6 kN on a standard RSC — and under stack loads of 400 kg (typical for four-high pallet configurations), that gap compounds. The first sign is usually diagonal corner cracking at the bottom layer of the pallet, discovered at the 3PL warehouse. By then the entire shipment is at risk and the MOQ reorder adds 18–25 working days to the supply chain.
What we check to catch this: incoming caliper measurement of liner stock (we pull three sheets per lot), basis weight verification against the COA, and side-by-side BCT testing of the production lot versus the approved sample. If BCT drops more than 8% versus the approved sample, the lot goes on hold under our IQ-03 protocol.
The second scenario is adhesive bond failure at low-temperature destinations. Corrugated cartons destined for cold-chain environments (frozen seafood, pharmaceutical, refrigerated cosmetics) require a dextrin-starch adhesive formulation rated to at least -20°C continuous. Standard corn-starch adhesive, which most commodity corrugated mills run as their default, begins to delaminate the flute tips from the liners at around -5°C under sustained exposure. The failure mode is not visible on the outside of the carton — the outer liner looks fine. The fluting detaches internally, and BCT collapses. This shows up as carton crush at destination, with no obvious external damage and a supplier who will argue the goods were mishandled.
The check: ask for the adhesive material safety data sheet, the formulation type, and the cold-resistance rating before approving a supplier for any cold-chain application. We also run our own internal peel test at -18°C on qualification samples when cold-chain is in scope.
The third scenario, and the one that causes the most downstream damage, is dimensional drift across production batches. A carton that is approved at 305 × 210 × 195 mm (internal) may run at 303 × 208 × 194 mm by the third production run, as die-cut tooling wears. That 2 mm internal length reduction sounds trivial. For a product packed in a formed pulp tray sized to the original cavity, it means the tray no longer fits, or the product shifts inside the carton and the lid buckles. We require suppliers to submit a first-article dimensional report for every new die set, and then re-verify dimensions at 50,000-unit intervals for high-volume runs. The pass/fail tolerance we hold is ±1.5 mm on internal length and width.
Does FSC Certification Actually Change the Board Specification? #
No — FSC chain-of-custody certification governs fibre sourcing and traceability, not board performance. FSC-certified corrugated board can range from low-grade recycled liner to virgin Kraft, and the ECT value depends entirely on the furnish and manufacturing parameters, not the FSC label.
This matters because we occasionally see RFQs where a buyer specifies “FSC certified, ECT ≥ 126 kN/m” as though those two requirements are linked. They are independent. You need to specify and verify both. FSC certification is audited through document trail and chain-of-custody records. ECT is measured on physical board. One does not validate the other.
For ISTA 2A transit testing (the standard most US retail importers require for carton validation), FSC status is irrelevant to the test outcome. The carton either holds up under the vibration and drop sequences or it doesn’t.
Specification Notes for Brand Partners #
When you brief us on export carton requirements, the most useful information you can send upfront is: product gross weight per carton, maximum stack height on pallet, destination climate zone (ambient, refrigerated, or frozen), and whether the cartons will be used in automated fulfilment lines. Those four variables drive every structural decision — flute profile, liner grade, ECT target, and adhesive specification.
The most common gap we see in incoming briefs is the absence of a confirmed pallet pattern. Brands often specify carton dimensions and gross weight, but not how many cartons stack per pallet or the maximum pallet height. Without that, we cannot calculate the dynamic compression load on the bottom layer, which is the number that actually sets the BCT requirement.
On sampling timeline: for standard RSC export cartons with no print, our typical sample lead time is 7–10 working days from confirmed specification. If the brief includes printed outer surfaces with brand artwork, add 5 working days for colour proofing. Structural change requests after the first sample typically add one full sampling cycle, so front-loading the dimensional brief saves more time than any process shortcut on our end.
Frequently Asked Questions #
What AQL level should we specify for incoming carton inspection?
For export cartons going into 3PL or retail fulfilment, we recommend AQL 2.5 for critical dimensional defects (carton won’t close, flap overlap out of tolerance) and AQL 4.0 for minor cosmetic defects. These levels align with ANSI/ASQ Z1.4 sampling tables and give a statistically meaningful sample from lots of 1,200–10,000 units without 100% inspection.
Can we use the same carton specification for both sea freight and air freight shipments?
It depends on your product weight and the humidity exposure during sea transit. Air freight cartons see less cumulative humidity and no vessel motion stacking loads, so you can often downgrade the ECT spec by one grade (from 126 kN/m to 108 kN/m) for air-only lanes without structural risk. If you’re dual-routing the same SKU across both modes, spec to the higher sea freight requirement.
How do we verify a supplier’s BCT results without our own test equipment?
Request the original test report, not just the COA summary. The test report should show the load-displacement curve, the peak load value, the date, and the equipment calibration certificate. Any CNAS-accredited lab in China will produce this as standard output. If a supplier provides only a summary table, ask for the underlying report. If they can’t supply it, treat that as a qualification red flag — a properly run BCT test generates an automatic data record.
Our supplier says their cartons are “Grade A” corrugated. What does that mean technically?
Nothing standardised. “Grade A” is a commercial label, not a regulated classification under any ISO, ASTM, or GB/T standard. The only way to evaluate the actual board grade is to request ECT, BCT, and liner basis weight values measured against named standards. We see this label used loosely across a wide range of actual board performance — from genuinely well-specified material to commodity board that wouldn’t pass a 25 kg gross-weight stack test.
What moisture content should we specify on the purchase order, and does it apply at production or at delivery?
Specify both: ≤12% moisture at time of production (measured per TAPPI T 412), and a transit packaging requirement (poly-wrapped pallet or desiccant slip-sheets) to maintain ≤14% at point of receipt. Corrugated board picks up 1–3% moisture on a 20–25-day sea freight transit in summer routing through Southeast Asia, so the production limit and the delivery limit are not the same number. Setting only one and not the other is where most moisture-related claims originate.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
The TAPPI T 811 vs GB/T 6546 dual-standard thing is worth flagging for anyone sourcing from mainland China suppliers — we’ve had COAs come in citing GB/T 6546 values that tested fine domestically but fell short of the 126 kN/m threshold when we re-ran them in-house using T 811 conditioning protocols. The sample conditioning difference (23°C/50% RH for TAPPI vs the GB/T ambient defaults some labs use) accounts for most of the gap, not the board itself.
The blank COA template request is the real filter — we started doing this with our Yiwu carton supplier in 2022 and within the first exchange knew they were pulling BCT values from a generic mill cert, not lot-specific testing. Took three qualification rounds and about 14 weeks before we had a COA format we’d actually trust for our 250g tin shipments.
The blank COA template request is something we started doing about 18 months ago and it filters out maybe 40% of candidates before we even get to sample testing — a supplier who lists “BCT: pass” with no unit has never once passed our full qualification.
The blank COA template test is real — we did exactly this with a Guangdong supplier last year and their template had BCT listed as a checkbox field (“pass/fail”), no numeric value, no test date. Took three revision cycles to get an actual measured value on there, and when they finally ran the ASTM D642 properly the result came back at 2.9 kN on a box spec’d for 3.2 kN minimum.
Moisture content is the one field we’ve had to add as a hard rejection criterion — we had a Shenzhen supplier hitting every structural number on the COA but shipping cartons at 14–15% moisture, and BCT in our warehouse was running 20–25% below their certified values by the time the shipment cleared customs.
The flute profile height spec (ISO 4046) is one we had to revisit when we moved from virgin kraft liners to 100% recycled content board last year — the recycled material was running slightly thinner at 3.4 mm consistently, which technically failed our own incoming spec even though BCT was holding above 3.2 kN. Took three months of back-and-forth with our Ningbo supplier to get the spec range formally updated and re-validated against actual stack performance data rather than just the inherited virgin-board tolerance.
On the 3.2 kN BCT minimum for empty cartons — do you apply a separate (higher) threshold when the carton is going into a refrigerated transit environment, since cold chain humidity tends to reduce BCT by 20–30% depending on board grade?
Bursting strength is the one we’ve had to push back on most consistently — our Dongguan supplier was reporting 1,380 kPa on COAs for about eight months before we ran parallel in-house testing with a Mullen burst tester and got readings averaging 1,140 kPa on the same lot. TAPPI T 810 vs their in-house method, no accreditation listed anywhere on the document header.