Wholesale Fiber Sourcing: Ask for the Test Method, Not the Brand Claim

Ask for the test method before you ask for the price.

That's the whole article in one sentence, and it's the habit that has saved us more money than any negotiation I've run. In February 2025 we quarantined 8,000 units of a viscose-blend knit because “colourfast” meant one thing to our buyer and a different thing to the mill — the re-dye and re-knit came to roughly $14,000, and we missed a retail window on top of it. Nobody lied. Nobody was careless. The purchase order said “colourfast” and both sides assumed they meant the same number.

Three questions now get answered in writing before any PO leaves my desk:

1. Who certified this, to which standard, and what is the certificate number?
2. Which test method, and which tolerance — stated as a number?
3. Who carries the risk where two suppliers meet?

Everything below is how I actually answer those, including the parts where I got it wrong first.

Why take this from me

I'm the quality and brand compliance manager at a mid-sized knits and home-textile manufacturer. I review roughly 300 supplier submissions a year — fiber, yarn, fabric, trims — and I've rejected about one in five first deliveries since 2022. Not because suppliers are dishonest. Because “industry standard” is a phrase that hides a surprising amount of variation, and nobody notices until a lot is already sitting in the warehouse.

For scale: we buy viscose and lyocell in the 40–120 tonne range annually, cotton yarn in the 90–150 tonne range, and nylon filament in smaller, spikier volumes for one performance line. So the checklist below comes out of real rejections, not a trade-show brochure.

A fiber-level claim is not a fabric-level claim

This is the mistake I see most often in sourcing, including in some of my own early POs. You buy lenzing™ viscose fibers at the yarn stage, the spinner blends them, the knitter knits, the dyehouse dyes — and somewhere between step one and step four, the thing you bought a claim for stops being guaranteed.

LENZING™ ECOVERO™ viscose is a good example of a claim that's real but narrow. Lenzing's published life-cycle assessment reports at least 50% lower water impact and CO2 emissions compared with generic viscose. That's a fiber-to-fiber comparison, produced by the fiber maker and third-party reviewed — but it is not a claim about your finished garment, and it isn't a claim about your supply chain. If you report that number externally, pull the current verified version and, if you use the Higg MSI, confirm the data still maps. Numbers get updated. Printed brochures don't.

LENZING™ ECOVERO™ fiber x REFIBRA™ technology is the same story with an extra moving part. REFIBRA™ is Lenzing's lyocell made with recycled cotton from cutting waste, and the recycled content percentage is published per article — it has moved over time and it isn't uniform across the product family. Verify the figure on the specific spec sheet you're buying, not the figure on the page describing the technology.

Two habits fall out of this:

• Ask for chain-of-custody documentation at the yarn stage, not the fiber stage. A fiber producer's certificate proves what left their plant. It does not prove what ended up in your yarn.
• Write the environmental claim into the contract with the conditions attached. Cellulosic fibers do biodegrade in many settings, and that's a genuine advantage over fossil-based synthetics — but “biodegradable” with no environment, no timeframe and no test standard is a hangtag word, not a spec. Name the test method and the conditions, or leave it off the tag.

On the regulatory side, the EU Green Claims Directive requires substantiation before a green claim is made, and member-state enforcement timelines have shifted more than once since 2024. Check the official EU source before you print anything. Our marketing team dislikes this advice. They follow it anyway.

What a wholesaler's spec sheet doesn't tell you

I learned this one expensively. In Q3 2023 we moved a viscose fiber wholesale order to a supplier quoting roughly $0.09/kg under our incumbent. Saved about $4,800 on a 53-tonne order. Then the yarn came back with slightly different evenness, which changed how it took dye on our navy, which put our shade at about Delta E 3.4 against the approved standard instead of under 2.

The “cheaper” choice looked smart right up until the dyehouse flagged it. Net position: re-dye, partial re-knit, three weeks of production shuffle — call it $19,000 against a $4,800 saving.

Looking back, I should have asked two questions before the order instead of after it: what's your lot-to-lot count variation, and can you send three consecutive lots for a shade trial rather than one clean sample? At the time the sample looked perfect and the price was real — and the sample was perfect. It just wasn't representative. That's the trap. A sample is a snapshot. You need a series.

Even after we tightened supplier qualification, I kept second-guessing the next order. What if the three lots had been curated too? The eight weeks until the first production run came back clean were not relaxing. That's the honest cost of a rule — it doesn't feel like it's working until you've been through a full cycle with it.

How to choose cotton yarn for wholesale

Cotton is where I see the widest gap between what buyers ask for and what they should ask for. Four things matter more than price per kilogram.

1. Ask for USTER STATISTICS percentiles, not adjectives

“Premium combed” means nothing. “Combed, USTER STATISTICS 25th percentile for IPI and count CV%” means something, because it benchmarks the mill against a global database rather than against a vague average. Ask for the percentile, the test date and the tester model. A 2021 certificate attached to a 2026 order tells you about 2021.

2. Count variation is a dyeing problem

Yarn count that drifts inside a lot will dye unevenly, and that shows up as barré in the finished knit — after finishing, when the money is spent. Quote count tolerance as a percentage and hold the supplier to the CV%, not just the average. If a spinner can't tell you the CV%, that's your answer.

3. Contamination is the risk nobody prices in

Foreign fiber contamination — polypropylene strapping fragments, packing lint, undissolved fiber — survives everything downstream and only surfaces after finishing. Ask about contamination control at the blowroom, whether they run optical contamination detection, and what their claims history looks like over the last two years. A supplier who's never had a claim is either very good or very new.

4. Match the spinning system to the end use

Ring-spun and open-end are not interchangeable, and the price gap is real. Ring-spun is smoother and stronger — better for fine-gauge knits and true knitwear. Open-end is more economical and bulkier, and it's fine for heavier wovens and plenty of home-textile applications. Buying ring-spun for a product that doesn't need it is the easiest place to waste money in a cotton program, and I say that as someone who did it for two seasons.

Vetting a nylon yarn distributor

Nylon is a different animal, and I'll be upfront that it's not my strongest category — we buy maybe 15–25 tonnes a year, so our leverage is limited.

What I've learned is that the distributor question matters more than the polymer question, because PA6 and PA66 are genuinely not interchangeable and a distributor who treats them both as “nylon” is a risk. PA6 melts around 220°C; PA66 sits closer to 255°C. They dye differently, heat-set differently, and behave differently in blends. If your process has a heat-set step, that gap is not academic.

The rest of my distributor checklist, ordered by how often each item has caused a problem:

• Oil content (OPU) spec and tolerance. Too little and you get filament damage in knitting; too much and you get dyeing defects and a mess in the creel.
• Denier and filament count tolerance in writing — plus the test method. Same trap as the viscose story above.
• Lot traceability from package to shipment. When a shade issue appears, you need to bracket the affected lots in days, not weeks.
• UV stabilisation for anything with an outdoor life. Ask whether it's in-polymer or a finish. They do not perform alike.
• A physical sample of the actual lot, not a reference hank. I've had a distributor send a beautiful reference and a slightly different production lot. It was within spec. It was also not what we approved.

That said, this list is calibrated for our volumes. A distributor moving 500 tonnes a year for you will answer these questions very differently than one moving 15.

When this is overkill

Honestly? For a one-off 200 kg trial order for a sample collection, half of this is theatre. Get a signed spec, get a physical lot sample, keep payment terms short. That's enough.

The full checklist earns its place when three conditions stack: the order is big enough that a rejection hurts, the lead time is short enough that you can't recover, and the color or hand-feel is brand-critical. One of three — manage it normally. All three — do the work.

And the piece I still don't have a clean answer for: how to verify a sub-supplier two tiers down without paying for an audit that costs more than the margin on the order. We handle it by consolidating suppliers, which lowers risk and also lowers our options. That's a real trade-off, not a solved problem. I'd rather say that than pretend the checklist covers everything.

An informed conversation with a supplier is worth ten rounds of email about price. Ask the method question first. The price tends to get easier after that.