The sample is not the fabric. I tell every new buyer this, and most of them nod and then learn it the hard way anyway. A swatch of Albini fabric shows you what a mill can produce. The spec sheet—agreed in writing, before the deposit—is what decides whether the 4,000 meters arriving in six weeks will match what your design team signed off on.
If you're sourcing denim fabric wholesale, building a suiting fabric private label program, or writing a satin fabric specification guide your mill will actually follow, the number that matters is not price per meter. It's the tolerance you both signed off on. Everything else—the heritage, the showroom, the relationship—is downstream of that.
Here's the short version: specify weight tolerance, color tolerance, and shrinkage before you approve anything. If a vendor won't put numbers in the contract, that's your answer.
Why I say this with some confidence
I manage quality and brand compliance for a mid-size apparel brand. I review every fabric lot before it hits production—around 180 lots a year, maybe 200, I'd have to pull the log. Denim, poplin, linen, satin. In 2024 I rejected 14% of first deliveries. Almost none of those were bad fabric in isolation. They were fabric that didn't match the order.
I only started caring about written tolerances after I ignored one. Back in 2022 we had a rush denim order for a spring capsule. The swatch was perfect. The mill had been solid for two years, so I approved the lot without pushing on weight variance in the contract. Delivery came in 9% lighter than the sample. That's not subtle—you feel it in the hand. Every pair of jeans would have hung thinner than what our design team approved. We couldn't use the lot.
The mill wouldn't take it back. Not because they were being difficult, but because "weight variance" wasn't in the agreement. We ate roughly $22,000 in rework, air freight for a replacement, and a two-week delay on a launch that was already tight.
I still kick myself for that one. Twenty minutes of paperwork would have either given us grounds to dispute the lot or—more likely—forced the mill to self-check before shipping. Either way, no $22,000 hole.
What I actually check, by fabric type
Most spec sheets cover the same categories. What changes is where the risk lives.
Denim fabric wholesale
- Weight: ±5% in oz/yd² or GSM is a reasonable default. Tighter than that and you'll pay for it; looser and the hand feel drifts noticeably between lots.
- Shrinkage: specify residual shrinkage after wash, not raw. We ask for under 3% warp and weft on most programs. Indigo often runs higher because of the dye process—know what you're buying before you commit to a wash recipe.
- Color: indigo lots vary more than buyers expect. Two lots from the same mill can be visibly different if the dye bath shifts between runs, and indigo shifts more than most.
Suiting fabric private label
Suiting is less forgiving than denim on consistency because it's usually sold as a set. A jacket and trouser made from different dye lots that don't match is a customer service problem you don't want to deal with at scale.
- Yarn count and construction: not "similar to sample"—the same count. Small changes ripple through drape and long-term durability in ways that don't show up until the garment is worn.
- Color continuity across lots: if you're building a program, ask how the mill manages lot continuity. This is where a vertical setup matters. Albini 1876 has been running its own dye houses since the 19th century, so lot control is part of the process rather than a bolt-on service.
- Hand feel: hard to spec, but you can spec the finish—mercerized, sanforized, and so on. Honestly, most "hand feel" disagreements I've mediated turn out to be finish disagreements wearing a costume.
Satin fabric specification guide
Satin is where I see the most confusion, because the word covers a lot of ground—silk satin, polyester satin, cotton sateen, and blends in between. The spec sheet has to name the fiber and the weave, not just "satin."
- Momme or GSM: momme is essentially the silk equivalent of GSM. A 19mm and a 22mm might look the same on a hanger and behave completely differently on a body.
- Weave density: threads per inch, or you'll get a satin that looks fine flat and pulls apart at the seam under any real tension.
- Color tolerance: satin reflects light differently than matte fabrics. A Delta E that looks fine on poplin can look visibly off on satin under the same lighting.
Color is where most buyers get burned
Across fabric types, color is the number one reason I reject lots. It's also the spec most buyers write vaguely—"match to approved sample" without a tolerance attached.
Industry standard accepts a Delta E under 2 for brand-critical colors. A Delta E of 2 to 4 is noticeable to a trained observer. Above 4 is visible to most people under normal lighting. Pantone publishes the framework that mill QC departments work against, and a mill should be reporting a number, not a thumbs up.
What I've learned: a vendor who sends you the Delta E reading alongside the shipment is usually a vendor who checks it in the first place. The number matters, but the habit matters more.
The question I ask before price
Before I ask "what does it cost," I ask "what's not included." Weight variance fees, test reports, lab dips, minimums per color—these all show up later if you don't ask now. The quote that looks 10% higher but lists everything usually costs less in the end than the quote that's silent on the details. Not because the cheap vendor is dishonest. Because you can't budget for what isn't written down, and surprises always cost more than planning.
If you've ever signed off on a quote and then watched the final invoice come in 18% higher, you know the feeling.
When tight specs are the wrong move
Two cases where pushing for tolerances backfires.
Small runs. If you're ordering 200 meters, the mill isn't going to spin up a separate dye lot for you. Pushing for ±2% weight tolerance on a minimum order is going to get you either laughed at or quoted a number that makes you wish you hadn't asked. For small volumes, work with what's in stock and accept the variance—or pay the minimum for a true custom run and get the tight specs.
Certain artisanal or couture programs. Some designers want variation. A hand-woven linen with inconsistent slubs is the point, not a defect. I've watched buyers write specs that would have rejected genuinely beautiful fabric because they didn't stop to think about what they were actually trying to buy.
I should mention that none of this replaces seeing the fabric. Numbers catch drift. Your hand catches everything else.
