Textile QC for Small Importers: Burn Tests, Colorfastness and GSM Checks

A $40 sample order tells you whether a $400 reorder is safe — but only if you actually test the sample. Most small importers do a look-and-feel check and call it QC. The factories know this, which is why the "sample quality vs production quality" gap exists at all. Closing it requires five tests you can run on a kitchen table with household items, none of which the supplier can fake from the other side of an ocean.

This is the receiving-QC workflow for home textiles: fiber verification, colorfastness, weight, construction, and hardware. Fifteen minutes per sample, and each test maps to a specific return-and-claim scenario you are protecting against.

Test 1: The burn test — is the fiber what the tag says?

The single most common wholesale fraud in textiles is fiber substitution: polyester sold as cotton-linen, rayon sold as silk, acrylic sold as wool blends. A lighter, a pair of tweezers, a sink, and ten seconds expose it.

Cut a snippet from a seam allowance or hem (never the visible face), hold it with tweezers over the sink, and light it:

Fiber	Approach behavior	Burning	Ash/residue	Smell
Cotton	Ignites on contact	Steady, yellow flame	Soft grey ash, crumbles	Burning paper
Linen	Ignites slowly	Steady flame	Fine light ash	Burning grass/paper
Silk	Shrinks from flame	Slow, sputters	Black crushable bead	Burning hair
Wool	Shrinks, self-extinguishes	Slow	Black crushable bead	Strong burning hair
Polyester	Melts and shrinks	Drips while burning	Hard black bead, cannot crush	Sweet chemical
Acrylic	Melts, ignites fast	Drips, black smoke	Hard irregular bead	Acrid fishy
Rayon/viscose	Ignites fast like cotton	Fast, bright	Soft grey ash	Burning paper — smells like cotton, but burns faster and drips ash

The two decisions the burn test makes for you:

1. Hard bead = synthetic present. If the quote said "cotton-linen blend" and you crush a hard black bead between your fingers (it won't crush), you have polyester — renegotiate or walk. A cotton-linen blend leaves only soft ash.
2. Rayon vs cotton: both smell papery, but rayon burns noticeably faster with a brighter flame and leaves no ember. Rayon is cheaper and weaker wet — a "cotton" tablecloth that fails when wet is rayon.

For blends, run the test twice from different areas of the garment — substitution scams sometimes use correct fabric at the hem tag location. The fabric-tier vocabulary for your listings ("silk-feel polyester," "coated cotton-linen") is only honest if you have burned the sample; the labeling stakes are covered in the FTC textile labeling guide.

Test 2: Colorfastness — will it bleed on the customer's table?

Color bleeding is the number-one review-killer for colored linens: a navy runner that turns a white tablecloth pale blue in one wash ends the SKU. Three checks, escalating severity:

1. Rub test (dry): white cotton cloth, firm pressure, 10 strokes on the face. Any color transfer = surface dye not fixed — reject.
2. Rub test (wet): same with a damp white cloth. Minor transfer on deep reds/navies is tolerable at this stage; visible transfer is a fail.
3. Wash test: sample swatch + a strip of white cotton fabric in warm water with a drop of detergent, agitated 15 minutes (a sealed jar shaken by hand works). The white strip tells the truth: any tint on the white strip means the dye will migrate in the customer's machine.

Dark, saturated colorways on cotton-linen fail most often. If a sample fails the wash test but the color is essential to the line, ask the factory for the reactive-dye version (fiber-bonded, not surface-pigmented) — it costs $0.30–0.80 more per unit and survives laundering. A care card telling customers "wash separately, cold, first three washes" manages what remains — the same honest-horizon card logic as the jewelry tarnish cards.

Test 3: GSM check — is the weight what you paid for?

Grams per square meter is the density spec that separates a $3.50 tablecloth from a $6.50 one, and it is the easiest spec for a factory to shave invisibly. Verify with a kitchen scale and scissors:

1. Cut a rectangle exactly 20×25 cm (= 1/20 m²). A book and a ruler make a clean cutting guide.
2. Weigh it in grams on a 1 g-resolution scale (a $12 coffee scale is fine).
3. GSM = weight × 20.

Product	Quoted GSM class	Acceptable measured range
Voile/sheer runner	60–90	±10
Standard cotton-linen cloth	180–230	±12
Thickened faux-linen	230–280	±15
PVC-coated cloth	280–420 (with coating)	±20
Pillow cover canvas-tier	200–260	±12

A measured weight more than one band below quote is a price renegotiation, not a shrug — you are paying per kilogram of textile in freight, and the drape your listing promised lives in that GSM. Weight-class merchandising detail is in the tablecloth GSM guide.

Test 4: Construction — seams, hems, pattern match

Sixty seconds per sample with your hands:

• Seam pull: grip the fabric on both sides of a seam, pull firmly. Gapping stitches or thread that saws through fabric = low SPI (stitches per inch). Count stitches over 2.5 cm: 10+ is decent, 14+ is good, under 8 is a return-risk product.
• Hem evenness: measure hem depth at four points; variance over 5 mm shows in photos and on shelves.
• Pattern/plaid match: on stripes and checks, corners should align across seams — mismatch screams mass-market and caps your retail price.
• Grain straightness: hold the cloth up by one corner; a trapezoid instead of a square means off-grain cutting, and it will hang crooked on every customer's table forever.
• Tassel and fringe anchor: pull each tassel firmly once. Any that come off in your hand come off in the customer's washing machine.

Test 5: Hardware and coatings

• Zippers (pillow covers): three full open-close cycles; teeth that catch or sliders that wobble fail before the customer's second wash.
• Coatings (PVC/wipe-clean cloths): fold the sample tightly for 60 seconds, unfold — cracking or whitening at the fold line means a coating that will craze within a season. Wipe with a drop of cooking oil and a damp cloth: if the oil ghosts, the "oil-proof" claim dies in reviews. The wipe-clean category claims to verify are listed in the wipe-clean tablecloths guide.
• Metal (grommets, rings, clasps): 24 hours in a sealed bag with a damp paper towel — any orange spotting means unsealed steel that will rust-stain the fabric.
• Elastic (fitted edges, scrunchies): stretch to full extension five times; permanent elongation over 10% means the elastic is undercured and will sag.

The receiving workflow: production batches too

Samples pass; now apply the same five tests to every production carton, on a sample basis:

1. Open one carton per shipment immediately — before it goes to shelf or storage
2. Pull 2–3 units per SKU (10% of a 30-unit carton is fine at small scale)
3. Run burn + rub + weight + construction on each; hardware checks on one unit per SKU
4. Photograph failures against the approved sample — side-by-side photos are your claim evidence
5. File the claim within 7 days with photos, counts, and the PO reference; suppliers who know you test at receiving behave differently on the next run

That last point is the real ROI: visible QC discipline raises your average incoming quality across every future order, because the factory's packing line knows the carton gets opened and weighed.

The 15-minute sample QC routine

1. Burn snippet — fiber truth (hard bead = synthetic)
2. Wet rub + jar wash — colorfast truth (white strip stays white)
3. Cut 20×25 cm, weigh ×20 — GSM truth (within one band of quote)
4. Seam pull, stitch count, grain check — construction truth
5. Zipper cycles, fold test, damp-bag metal test — hardware truth

Pass all five, reorder with confidence. Fail any one, renegotiate with evidence — the sample you just tested is the cheapest negotiation leverage a small importer will ever own.