Quality Control for Vietnam Quilling Cards: The Paper, Glue and Colourfastness Specs Wholesale Buyers Should Write Into Every PO (2026 Guide)

A gift buyer in Dubai opens the first carton of a 12,000-piece quilling card order and finds nothing obviously wrong. The cards look like the sample. Six weeks later the retailer calls: forty per cent of the cards on the display stand have a coiled rose that has lifted at one edge, three designs have faded from coral to a washed-out pink where the stand sat near a window, and the envelopes no longer close flat because the relief has been compressed in transit. None of this was visible at the inspection. All of it was preventable with four lines of specification on the purchase order.

Quilling cards are the hardest handcraft category to inspect well, because almost every defect that matters is a latent defect. A wooden cutting board that will crack tells you something on day one — you can measure moisture content. A quilling card that will delaminate, fade or flatten looks perfect on the day it is packed. The quality is in the paper, the adhesive, the dye and the packing engineering, and none of those are visible to a person holding the finished card.

This guide sets out the technical specification wholesale and private-label buyers should be writing into quilling card purchase orders in 2026: what to specify for paper and strip stock, how to test colourfastness and adhesive integrity, what dimensional tolerances are realistic for a handmade product, how to build a defensible AQL plan when every unit is genuinely unique, and how to structure inspection so that it catches what actually goes wrong.

In-process QC on a Vietnam quilling card line: colour matching, dimensional checks and glue-integrity inspection before the cards are sleeved and packed.

Why “It Looked Fine at Inspection” Is Not a Defence

Most quilling card inspections are visual and cosmetic. An inspector opens cartons, counts pieces, checks that the design matches the golden sample, looks for glue smears and creased corners, and signs off. That process catches workmanship defects — the visible ones — and misses the four failure modes that generate almost all downstream claims:

Adhesive failure. A coil that is tacked rather than properly bonded holds fine for weeks and lifts once the card is handled repeatedly or stored in heat.
Colour migration and fading. Dyed paper strips that were never lightfastness-tested lose saturation under retail lighting in a matter of weeks.
Relief compression. Cards packed without a rigid spacer arrive with the three-dimensional element crushed to half its intended height.
Moisture-driven warp. Cards made and packed at 75–85% relative humidity in northern Vietnam, then shipped into a Gulf or European winter, cup and curl as the base board loses moisture unevenly.

None of the four is a workmanship problem. All four are specification problems. The inspector did their job; the purchase order did not do its job. That distinction matters commercially, because a supplier who was never asked for a lightfastness standard is not contractually in breach when the cards fade — and you will discover that when you try to claim.

What Actually Fails on a Handmade Paper Card

Before writing specifications it is worth understanding the physical construction. A quilling card is typically four layers: a folded base card (the greeting card itself), a mounted panel or aperture frame, a backing sheet on which the quilled artwork is built, and the quilled elements themselves — narrow paper strips rolled, shaped and glued on edge to the backing. Each interface is a potential failure point.

Layer / component	What fails	Root cause	Where to catch it
Base card (270–350 gsm board)	Cupping, warp, hinge cracking at the fold	Wrong grain direction; moisture cycling; board too heavy to score cleanly	Incoming material QC + creasing setup
Mounted panel / aperture frame	Panel lifts at corners; visible adhesive bleed	Wrong adhesive for coated stock; insufficient dwell under pressure	In-process, first-off approval
Backing sheet	Bubbling, wrinkling around dense quill clusters	Water-based adhesive applied too wet on light stock	In-process
Quilled elements	Coils uncoil, lift, or shed; loss of relief height	Edge-glue starvation; strip too light; compression in packing	In-process + drop/press test at final
Dyed strip colour	Fading, bronzing, colour shift between batches	No lightfastness spec; unmatched dye lots	Incoming material QC + lot control
Envelope and sleeve	Envelope will not close; sleeve creased; card rubs	Relief height exceeds envelope gusset; no spacer	Packing validation before mass production

Read that table as a QC allocation map. Roughly two-thirds of quilling card failures are decided before a single coil is rolled — at incoming material inspection and at packing design. Only about a third are decided at the artisan’s bench. Buyers who put all of their inspection budget into final random inspection are inspecting the smallest share of their risk.

Paper and Board Specification: The Four Numbers That Matter

Most quilling card enquiries specify the artwork in enormous detail and the paper not at all. Reverse that priority. Four material parameters determine whether the card survives a year on a shelf.

Parameter	What to specify	Why it matters
Base card grammage	270–350 gsm for A6/A5 formats; 350 gsm where a heavy quill cluster is mounted	Under 250 gsm the card will not stand on a display stand and cups under the weight of the artwork
Grain direction	Grain parallel to the fold, stated explicitly on the PO	Cross-grain folding cracks the coating at the hinge — the single most common “cosmetic” complaint
Quilling strip weight	120–160 gsm for structural coils; 100–120 gsm for fringed or decorative elements	Light strip collapses under its own relief; heavy strip will not hold a tight inner coil
Strip width tolerance	±0.2 mm on nominal width (typically 3 mm, 5 mm or 10 mm)	Width variation is what makes a “handmade” card look untidy rather than artisanal

Add two qualitative requirements. First, specify acid-free or alkaline-buffered stock (pH 7.0 or above) for anything sold as a keepsake or wedding card; groundwood-content paper yellows visibly within eighteen months and buyers in the premium gift channel will notice. Second, if your market requires it, specify FSC Mix or FSC Recycled chain-of-custody on both the board and the strip, and ask for the certificate number and the transaction-level claim, not just a logo on a brochure. A supplier who can show you the certified paper on the invoice is a different proposition from one who says “our paper is FSC”.

One practical note on sourcing. Vietnamese card manufacturers buy base board and dyed strip from a mix of domestic mills and imported stock. Ask which, and ask for the mill name. It is not an intrusive question — it is the difference between a repeatable colour and a colour that shifts every time the factory changes supplier on price.

Colourfastness: The Test Almost Nobody Runs

Quilling cards spend their commercial life on a display stand, frequently near a shop window or under continuous LED retail lighting. Dyed paper strips in the coral, magenta, violet and turquoise families are the most fugitive; deep reds and blues are usually more stable. If you have never asked your supplier for a lightfastness figure, you do not know which of your designs will survive a season.

The standard reference is the Blue Wool Scale (ISO 105-B02 / ASTM D5383 in the graphic-arts context), which runs from 1 (very poor) to 8 (excellent). For quilling cards a workable commercial specification is:

Blue Wool 4 or better for any colour used in a design that will be sold through open retail display.
Blue Wool 5 or better for pale tints and pastels, which show fading soonest because there is less pigment to lose.
Blue Wool 3 acceptable only for cards sold sleeved and boxed, in a closed gift set, or for seasonal designs with a defined eight-week retail window.

If a full laboratory test is disproportionate for your order size — and for a first order it often is — run a practical alternative. Ask the factory to produce a colour reference card carrying every strip colour in the range, cut it in half, seal one half in an opaque envelope and tape the other to a south-facing window at your own office. Compare after four weeks and again after twelve. It is not a certified test, but it will tell you which three colours in your palette are the problem, and it costs nothing.

Separately, specify colour consistency across batches, which is a different problem from fading. Dyed paper is produced in lots, and lots vary. The contractual language you want is: strips for the entire order quantity to be drawn from a single dye lot per colour; where that is impossible, the supplier notifies before production and submits a physical lot comparison for approval. Without that clause you will receive an order in which cartons 1–40 and cartons 41–80 are visibly different shades of the same “coral”, and you will have no basis to reject.

Adhesive Integrity: Where the Warranty Claims Come From

Every coil in a quilling card is bonded on its narrow edge — a contact area of a few square millimetres. Adhesive choice and application discipline therefore matter more here than in almost any other paper product.

Specify three things. First, the adhesive type: a PVA-based, acid-free, non-yellowing craft adhesive, applied by fine applicator rather than brush. Cheap general-purpose white glue yellows at the bond line within a year and is visible against pale paper. Second, the dwell and cure regime: quilled panels to be held flat under light weight for a stated minimum cure period before the next operation, and cards not to be sleeved until fully cured. Rushing cure to hit a shipping date is the most common cause of a whole shipment lifting. Third, a physical acceptance test.

The acceptance test that works in practice is a simple inversion-and-tap test, and it belongs in your PO because it converts a subjective judgement into a pass/fail:

Hold the finished card vertically, artwork facing down, and tap the reverse of the panel firmly three times with a fingertip.
Pass: no element detaches, no coil shifts position, no loose paper dust falls.
Fail: any element detaches or visibly moves. One failure in the sample is a major defect, not a minor one.
Run the test on every unit in the inspection sample, not on a sub-sample — it takes about four seconds per card.

Add a heat condition if your destination market warrants it. Cards destined for the Gulf, South and Southeast Asia should also be tested after conditioning at 40 °C for 24 hours, because a marginal bond that holds at 25 °C will release in a warehouse in Jebel Ali or Jeddah in August. This one clause has probably prevented more claims than any other single line we write into a specification.

Dimensional Tolerance on a Handmade Product

Buyers arriving from a printed-card background often specify machine-print tolerances and are then surprised when the factory quietly ignores them. A hand-assembled card cannot hold ±0.5 mm. Specifying an unachievable tolerance is worse than specifying none, because it teaches your supplier that your specification is decorative.

Dimension	Realistic tolerance	Notes
Trimmed card size (folded)	±1.0 mm	Guillotine-trimmed board; tighter is achievable but adds cost
Aperture / window position	±1.0 mm	Die-cut; check the die has not worn mid-run
Artwork placement on panel	±2.0 mm from nominal centre	Hand-mounted; ±1 mm is possible but slows the line materially
Relief height of quilled element	Nominal ±1.5 mm, and a stated maximum	The maximum matters more than the nominal — it drives envelope and packing design
Envelope internal size	Card size +3 to +5 mm on both axes	Too tight and the card scuffs the artwork every time it is removed
Squareness of trim	≤1.0 mm out of square over the long edge	Visible on a display stand when cards are stacked

Note the fourth line. Maximum relief height is the single most useful dimension on the whole specification, and it is the one buyers most often omit. It determines envelope depth, sleeve size, the spacer required in the carton, the number of cards per inner box, and therefore your cubic metre count and your freight cost. Fix it early and every downstream decision falls into place; leave it open and you will be redesigning the packing after the first shipment.

Relief Compression: Why Cards Arrive Flat

A quilling card is, structurally, a box of air held up by paper coils standing on edge. Stack forty of them face-to-face in a carton and the ones at the bottom carry the weight of the ones above. Add a pallet stacked two high and a hot container, and the coils take a permanent set.

The fix is packing engineering, and it should be validated before mass production, not after the first claim. Specify:

A rigid or semi-rigid spacer — a crash-lock board frame, a die-cut card collar, or a clear rigid sleeve — that carries the compressive load around the artwork rather than through it.
A maximum stack count per inner box, derived from the actual relief height rather than from a round number.
Cards packed on edge in the master carton where format allows, rather than face-to-face flat, so that the load path runs through the board rather than the coils.
A maximum carton stack height on the pallet, marked on the carton, and a “do not double-stack” instruction where the relief exceeds roughly 8 mm.
A validated drop test: master carton dropped from 90 cm on one corner, three edges and one face, then opened and every card checked against the tap test above.

Ask for photographs of the drop test being performed on your actual packing, with your actual cards. Any competent factory will do this for a new programme; a factory that will not is telling you something useful.

Moisture, Humidity and the Hanoi-to-Dubai Problem

Paper is hygroscopic. It reaches equilibrium with the ambient relative humidity, and it changes dimension as it does so — more across the grain than along it. A card assembled in a Vietnamese workshop at 75–80% RH and shipped to a Gulf warehouse at 25–35% RH, or into a heated European interior in January, will lose moisture and shrink unevenly. Because the quilled artwork is bonded to only one face of the backing sheet, the card cups toward the artwork.

Three controls address this, and they are cheap:

Conditioning before packing. Finished cards held in a conditioned area at 55–60% RH for a stated period before sleeving, so that they are packed nearer to the mid-point of their service range rather than at the wet extreme.
Desiccant in the master carton. Silica gel sachets sized to carton volume, with the sachet count specified on the PO and verified at inspection. This is standard practice for ocean freight out of Vietnam and costs a few cents per carton.
Sealed inner packaging. A poly or cellophane sleeve per card, or a poly liner in the inner box, slows moisture exchange during the container journey where the humidity swings hardest.

If your programme is going into a genuinely dry market — inland Middle East, or a heated northern-European retail environment — say so at the enquiry stage. It changes the backing-sheet weight the factory should be using and the conditioning regime, and it is much cheaper to design for than to remedy.

The Golden Sample: What It Must Physically Contain

For quilling cards a single approved sample card is not enough, because the sample tells you what “good” looks like but not what the boundaries of acceptance are. A usable golden sample set for this product contains five things:

The approved card itself, signed and dated by both parties, sealed against light so it does not fade and become useless as a colour reference.
A colour reference strip card carrying every dyed paper colour in the range, with the dye lot recorded.
A limit sample for the worst acceptable card — the one at the edge of tolerance on artwork placement and coil neatness. Without this, “handmade variation” is argued case by case on every shipment.
A reject sample showing the defects that will be rejected: adhesive bleed, lifted coil, creased hinge, compressed relief.
The approved retail pack — envelope, sleeve, spacer, barcode label and inner box — assembled exactly as it will ship.

Hold one set, leave one set at the factory, and give one set to your third-party inspection agency. The cost is three sets of cards and an afternoon of somebody’s time, and it removes the single largest source of dispute in this category: the argument about whether a variation is charm or a defect.

Building an AQL Plan for a Genuinely Handmade Product

AQL sampling (ISO 2859-1 / ANSI Z1.4) works perfectly well for quilling cards, provided the defect classification is written for a handcraft product rather than copied from an electronics checklist. The mistake buyers make is not the sample size — it is classifying normal handmade variation as a minor defect, which produces inspection reports full of noise and trains everybody to ignore them.

A workable classification looks like this. General Inspection Level II, single sampling, normal severity:

Class	AQL	Examples for quilling cards
Critical	0	Wrong design or wrong quantity in carton; missing barcode or wrong retail price; sharp staple or pin left in the card; any element that detaches on normal handling in the retail pack
Major	2.5	Coil lifted or detached on tap test; visible adhesive bleed on the face; hinge cracked at the fold; artwork placement out by more than 2 mm; relief compressed below the stated minimum; envelope will not close; visible colour mismatch between cards in the same inner box
Minor	4.0	Slight glue sheen visible at an oblique angle only; coil spacing variation within the limit sample; trim within tolerance but visibly uneven; light surface dust
Not a defect	—	Individual coil geometry differing between cards where the design and limit sample are met; minor variation in the depth of a fringed element; artisan-to-artisan variation in hand-shaped petals

That last row is the one to negotiate explicitly. If you do not write down what is not a defect, every inspector will apply their own judgement, and a quilling card programme can fail inspection on characteristics that are the entire commercial point of buying handmade.

For a first order, tighten the plan: run at Level II but with a reduced AQL of 1.5 on majors, and commit to relaxing to 2.5 after three consecutive clean shipments. That gives the factory a visible incentive and gives you protection where your risk is highest.

In-Process QC Beats Final Inspection in This Category

Final random inspection on a handcraft product tells you whether to accept or reject a container that is already finished. On a quilling card programme, that is a poor place to discover a problem: rework means dismantling and rebuilding cards by hand, which is slower than making new ones, and a rejected lot two weeks before a seasonal on-shelf date is a commercial disaster rather than a quality event.

Push your inspection budget forward. A sensible structure for a quilling programme:

Incoming material check. Board grammage and grain, strip weight and width, dye lot recorded against colour reference. One hour, at the start of production.
First-off approval. The first twenty cards off the line compared against the golden sample and the limit sample, before the line is released to run. Photographed and shared with the buyer the same day.
During-production inspection at 20–30% completion. Early enough that a systematic error — wrong adhesive, wrong strip width, artwork drifting — can be corrected on the remaining 70%.
Packing validation. Drop test and stack test on the actual carton, before the bulk is packed.
Final random inspection. AQL sample, including the tap test on every card in the sample and a heat-conditioned sub-sample where the destination warrants it.

The during-production inspection is the one buyers most often skip and the one that pays for itself most reliably. Catching a strip-width error at 25% completion costs you a conversation; catching it at 100% costs you the order.

Third-Party Inspection: What to Ask For

If you are using an independent agency — SGS, Bureau Veritas, Intertek, TÜV or a regional equivalent — the quality of the report depends almost entirely on the brief you give them. A generic “AQL II, 2.5/4.0, general appearance and quantity” brief on a quilling card order will produce a generic report.

Include in the inspection brief:

The tap test protocol, stated as a numbered procedure with the pass/fail criteria above.
Relief height measurement on a stated number of pieces, with the nominal and maximum figures.
A carton drop test with the specific drop heights and faces.
Colour comparison against the sealed colour reference under a stated light source (D65 daylight is the usual choice) rather than under warehouse fluorescents.
Verification of desiccant sachet count and inner packaging integrity.
Photographs of the limit-sample comparison for any borderline judgement, so that you can adjudicate rather than accept the inspector’s opinion by default.

Nominating your own agency and making its determination binding is normal practice and a reasonable ask. A supplier that resists a buyer-nominated inspection on a handcraft product is worth a second look; a supplier that offers it before you ask is telling you something about how they run.

Packing and Retail Presentation Are Part of Quality

For gift and stationery buyers, the quilling card and its presentation are the same product. A structurally perfect card in a scuffed sleeve with a crooked barcode label will be marked down at the shelf edge exactly as if the card itself were defective. Bring packing into the QC specification rather than treating it as logistics:

Sleeve or bag material and gauge, and whether it is resealable — resealable matters for open-display retail where customers handle stock.
Barcode placement to a stated position and orientation, and a scan verification on a sample at final inspection. A barcode that will not scan at the till is a critical defect in most retail contracts.
Backing board or insert colour and finish, since it shows through the sleeve behind the card.
Inner box count, orientation and labelling, so that the retailer’s replenishment staff can restock a display without opening every box.
Any market-specific labelling: country of origin marking, importer details, recycling symbols, and — for the EU market — packaging material identification where required.

A QC Clause You Can Adapt for Your Purchase Order

The following is a condensed, usable version of the specification described above. It is deliberately short enough to sit inside a PO rather than in a separate quality manual that nobody reads.

Materials. Base card 300 gsm minimum, acid-free (pH ≥ 7.0), grain parallel to fold. Quilling strip 120–160 gsm structural, width tolerance ±0.2 mm. All dyed strip to Blue Wool 4 minimum; pastels Blue Wool 5 minimum. Single dye lot per colour for the full order quantity unless notified and approved in advance in writing. Adhesive: acid-free non-yellowing PVA, fine applicator.

Dimensions. Trimmed card ±1.0 mm. Artwork placement ±2.0 mm from nominal. Relief height nominal as per approved sample, maximum [X] mm. Envelope internal dimensions card +3 to +5 mm both axes.

Testing. Every card in the inspection sample to pass the tap test (card inverted, panel tapped three times; no element to detach or shift). Sub-sample of [n] cards to pass the tap test after 24 hours at 40 °C. Master carton to pass a 90 cm drop test on one corner, three edges and one face, with all cards then passing the tap test.

Conditioning and packing. Cards conditioned at 55–60% RH before sleeving. Silica gel [n] g per master carton. Maximum [n] cards per inner box. Cartons marked “do not double stack”. Packing to be validated and photographed before bulk packing.

Inspection. ISO 2859-1 Level II, single sampling, normal. Critical 0 / Major 1.5 / Minor 4.0 for first three shipments, thereafter Major 2.5. Handmade variation within the approved limit sample is not a defect. Buyer may nominate an independent inspection agency whose determination is binding.

Six Warning Signs in a Supplier’s Answer on Quality

“All our paper is imported, very good quality.” No mill name, no grammage, no grain direction. If they cannot name the stock they cannot control the stock.
No answer on lightfastness. A supplier who has never been asked about Blue Wool has never had a buyer who cared, which tells you about the rest of their customer base.
Resistance to a limit sample. “Handmade, so each one is different” is true and is also the beginning of the conversation, not the end of it.
No in-process inspection. If the only check is at the end, systematic errors run to completion.
Packing designed after the card. If nobody has measured the relief height and worked out the carton, the packing will be improvised at the last minute.
Refusal of a buyer-nominated third-party inspection. There is no good reason for this on a handcraft order.

None of these is disqualifying on its own. Three of them together, on a first order, is a reason to run a smaller trial before committing a season’s volume.

What This Costs, and Why It Is Cheap

Buyers sometimes resist this level of specification on the grounds that it will add cost to a low-unit-value product. In practice, the specification above adds very little to the ex-works price. Acid-free board and non-yellowing adhesive are a marginal upgrade on standard stock. Conditioning is a scheduling decision, not a capital one. Desiccant is a few cents per carton. A rigid spacer is the only meaningful addition, and it typically costs less than the freight saved by packing to a validated stack count instead of guessing.

Set that against the alternative. A 10% claim rate on a 20,000-piece quilling card programme is 2,000 cards, plus the retailer’s handling charge, plus the markdown on the remaining stock, plus the conversation about next season. The specification is not a cost centre; it is the cheapest insurance available on a product whose defects only appear after you have been paid.

Frequently Asked Questions

How many quilling cards can a factory realistically produce per day?

Output depends entirely on design complexity. A simple single-flower design runs at perhaps 25–40 finished cards per artisan per day; a dense multi-element design with fringed flowers and layered foliage may run at 8–15. This is why lead time on quilling cards is driven by design choice far more than by order quantity, and why a buyer who simplifies one element of a design can sometimes halve the lead time without any visible loss of quality.

Can quilling cards be made to a specific Pantone reference?

Approximately, not exactly. Dyed paper strip is bought from a limited standing palette rather than mixed to order, so the practical process is to match your Pantone reference to the nearest available strip colours and approve a physical sample. If a colour is central to your brand, ask early whether a custom dye run is viable — it usually requires a substantially higher minimum quantity and a longer lead time.

What is a realistic MOQ for a private-label quilling card range?

For an existing design with your own artwork on the insert and your own packaging, minimums are usually modest. For a fully custom design developed from your artwork, expect a higher per-design minimum plus a development sample charge, because the factory has to prototype the coil structure and train the line on it. Splitting a modest total across many custom designs is the most common reason a quote comes back higher than a buyer expected.

Do quilling cards need any product safety testing?

Standard adult greeting cards generally do not fall under toy safety regimes. However, if the card is marketed to children, sold as part of a children’s activity set, or carries small detachable embellishments, it may fall within scope of EN 71 in the EU or equivalent regulations elsewhere, and small-parts requirements become relevant. Clarify the intended market positioning before production, not after.

Working With Viet Farm Vision

Viet Farm Vision manufactures and exports handcrafted quilling cards, wooden kitchenware and houseware, and compostable and disposable tableware from Vietnam for buyers across the Middle East, Asia, Europe and Africa. On quilling card programmes we work to a written material specification rather than to a photograph, we record dye lots against a sealed colour reference, we run first-off and during-production checks and share the photographs the same day, and we validate the packing with a drop test before we pack the bulk. We will tell you the maximum relief height of a design before you commit to an envelope, and we will tell you when a colour in your palette is likely to fade rather than discovering it with you six weeks after the cards reach the shelf.

Buyers are welcome to nominate their own third-party inspection agency and to make its determination binding on us. If you are developing a quilling card range for a gift, stationery, wedding or corporate programme — or re-specifying an existing range after a quality problem — contact our export team at info@vietfarmvision.com or visit vietfarmvision.com to discuss materials, tolerances and an inspection plan appropriate to your market.