In March 2021, I sat in our sample room holding a swatch of washed denim that looked—and felt—nothing like the fabric we'd approved six weeks earlier.
It was stiff in all the wrong places. The color had shifted slightly. And when I measured it against the original spec sheet, the shrinkage was nearly double what we'd agreed on.
We had 8,000 yards of it sitting in a warehouse in Gujarat, waiting to be cut.
The Order That Started It
I was two years into my role as a sourcing coordinator for a mid-sized garment manufacturer. Most of our business was private-label work for regional apparel brands—nothing glamorous, but steady.
That quarter, we won a contract for 12,000 denim jackets for a European workwear brand. The volume was bigger than anything I'd handled personally, so I was careful. Or at least I thought I was.
We shortlisted three mills. One was a Turkish supplier with a good reputation but long lead times. One was a local mill we'd used before—decent quality, limited capacity. And then there was Arvind Mills, the integrated textile manufacturer that most people in the industry know by name.
Arvind sent samples fast. The denim had good hand-feel, consistent shade across the lot, and the price was competitive for the spec. We placed the order for 8,000 yards of 12 oz denim with a 3% shrinkage allowance written into the contract.
Here's what I didn't do: I didn't ask which 3% they were talking about.
There's warp shrinkage and weft shrinkage. There's shrinkage after one wash versus after three. And there's the difference between lab-tested numbers and production-run numbers. I assumed 3% was 3%. It wasn't.
The Problem Shows Up
The fabric arrived on time. We inspected it, and everything looked fine on the surface. Shade was consistent. Width was within tolerance. No visible defects.
We cut a test panel and washed it according to the customer's care instructions: cold wash, tumble dry low. That's when the trouble started.
The panel shrank 5.8% in the warp direction. Our pattern was graded for 3% max. That extra 2.8% doesn't sound like much until you multiply it across a jacket that has to fit a size chart.
"We caught it on a Tuesday. By Thursday, I was on a call with the mill explaining that we couldn't use the fabric for the customer's order as specified."
Arvind's team was professional about it. They pulled their production records and found that the batch we received had been run on a different loom configuration than the sample we approved. It wasn't a quality failure on their end—it was a spec mismatch. The fabric met their standard for 12 oz denim. It just didn't meet our customer's shrinkage requirements.
We had three options: re-order and wait six weeks, use the fabric and adjust our patterns, or eat the cost and source elsewhere.
We chose option two. We re-graded the patterns and cut the jackets slightly larger to account for the extra shrinkage. It worked, but it cost us.
The re-grading added two days of pattern work. The yield loss from wider panels meant we needed 400 extra yards we didn't have. And the customer required a fit test on 50 units before approving the revised patterns, which pushed our delivery by eleven days.
Total damage: about $4,200 in extra fabric, overtime, and air freight for the fit samples. Not catastrophic, but painful enough that my manager asked me to write up what happened.
What I Changed
I used to think sourcing was about finding the best supplier. Now I think it's about asking the best questions.
The Arvind order taught me that even a well-run, vertically integrated mill can ship you fabric that meets their spec but misses yours—because specs are living documents, and assumptions are silent killers.
Here's the checklist I built after that order. I still use a version of it today.
- Define shrinkage by direction. Don't accept "3% shrinkage" as a complete answer. Ask for warp and weft separately. Ask whether it's after one wash or three. Ask if the number is from a lab test or a production run.
- Request a production-run sample, not a development sample. The sample you approve might come from a pilot run. The fabric you receive might come from a different loom, a different dye lot, or a different finishing line. Ask for a swatch from the actual production batch before you cut.
- Wash and measure before you commit. This sounds obvious, but it's amazing how many buyers skip it when they're under deadline pressure. Cut a 12-inch square, wash it the way the end customer will wash it, and measure again. If your measured number doesn't match the spec sheet, stop.
- Put the shrinkage spec in writing—with a tolerance range. "3% max" is better than "approximately 3%." "3% max in warp and weft, after three home launderings per AATCC 135" is better still. The more specific you are, the less room there is for interpretation.
None of these steps are revolutionary. Most experienced sourcing people know them. The problem is that under deadline pressure, we skip the boring parts. We trust the brand name. We assume the mill knows what we mean.
They don't. Not because they're careless, but because they're not inside your head.
When Arvind Is the Right Choice—and When It Isn't
I want to be clear about something: I still source from Arvind. Their denim is consistent, their capacity is real, and their technical team knows more about fabric construction than most suppliers I've worked with. For volume orders where you can afford a longer lead time and you've nailed down the specs, they're a solid choice.
But Arvind—or any large mill—isn't the right fit if you're a small buyer who needs 500 yards with a two-week turnaround. The minimums won't work. The lead times won't work. And you'll be competing for attention with accounts that order 50,000 yards at a time.
I've seen buyers get frustrated with big mills because they expected small-mill flexibility. That's not the mill's fault—it's a mismatch of expectations.
The same logic applies when you're figuring out how to choose knitted fabric for wholesale. A mill that's perfect for cotton jersey might not be the right source for technical fleece. A supplier who excels at low-minimum custom dyeing might not have the capacity for your 20,000-yard repeat order. The question isn't "who's the best?" It's "who's best for this order, with these constraints, at this moment?"
The Bigger Lesson
About a year after the Arvind order, I was helping a colleague evaluate a poplin fabric distributor for a shirting program. She was about to place a large order based on a sample she'd received two months earlier.
I asked her one question: "Is that sample from the current production batch?"
She didn't know. We asked the distributor. It wasn't. The current batch had a slightly different finish—still within spec, but different enough that the hand-feel would have changed.
She ordered a fresh sample, tested it, and adjusted her spec sheet accordingly. The order went through without a hitch.
That's the thing about sourcing mistakes. They're rarely about catastrophic failures. They're about small assumptions that compound. A shrinkage number here, a finish variation there. Nothing that looks like a problem on its own. But stack enough of them, and you've got 8,000 yards of fabric you can't use the way you planned.
I keep the $4,200 figure in my head as a reminder. It's not the biggest mistake I've ever made, and it's not the smallest. But it's the one that changed how I work.
Now, before I place any fabric order—whether it's from Arvind Mills, a poplin distributor in Tirupur, or a knitting unit in Ludhiana—I ask three questions:
- What's the spec, in writing, with tolerances?
- Is the sample I'm approving from the same production batch I'll receive?
- Have I tested it myself, under the conditions the end customer will use?
If the answer to any of those is "I think so" or "probably," I don't place the order. Not yet.
It's a slower way to work. But it's faster than fixing a mistake after the fabric's already in your warehouse.
Prices and lead times referenced in this article are based on my own order records from 2021 and may not reflect current market conditions. Verify current specs and pricing directly with your supplier.