Polartec material note

The 9-Point Pre-Purchase Checklist I Run Before Approving Any B2B Order

2026-11-13 · Jonas Richter

Who this checklist is for

I'm the person who signs off—or sends back—incoming orders before they reach our customers. In 2024 that was 214 unique SKUs across four categories: apparel fabric, hospitality linens, industrial mesh, and guest amenities. I rejected 12% of first deliveries last year. Not because suppliers were careless, necessarily. Mostly because the spec sheet said one thing and the physical goods said another.

What follows is the checklist I actually run. Nine steps, in order. About 40 minutes per new SKU. If you're approving deliveries, managing supplier quality, or writing purchase specs, this is for you.

The 9-point checklist

1. Confirm material identity before you confirm price

Price is the last thing you negotiate. Fiber identity comes first.

In March 2024, a private-label client asked us to source a women's wool coat comparable to a loro piana wool coat women's style—same hand, same drape, roughly 60% of the cost. We found a mill that claimed 100% merino. The third-party lab report came back 78% merino, 22% polyester. The "savings" would have shown up as pilling within six weeks.

Checkpoint: Request a fiber composition report from an independent lab, not the supplier's own spec sheet. As of January 2025, that runs roughly $85–140 per sample (based on quotes from three US labs; verify current pricing).

2. Touch the goods, not just the spreadsheet

The loro piana cashmere polo sweater we benchmarked in Q1 2024 had a hand-feel no spec sheet captured. GSM, micron count, twist—all within range on paper. But the "premium" alternative felt flat. Turned out the finishing process was different, and finishing doesn't show up in a fiber test.

Checkpoint: 30-second touch test against a known reference sample, in good light, with the sample at room temperature. Not in a cold warehouse.

3. Weigh and measure a random sample—not the top layer

We didn't have a formal sampling process until 2023. Cost us when a hotel linen suppliers delivery arrived with 300-thread-count sheets that measured closer to 240 after one wash. The top three sheets in the box were fine. The rest weren't.

What I mean is that thread count alone isn't a spec—it's a marketing number. The real specs are weight per square meter, yarn size, and shrinkage after the first wash. A 300-count sheet at 110 GSM is a different product from a 300-count sheet at 145 GSM, even though both can legally carry the same label in most markets.

Checkpoint: Pull 5% of the shipment, minimum five units, from different layers. Weigh, measure, wash one, measure again. Shrinkage tolerance for hospitality linens is typically 3–5% per the supplier's own TDS—hold them to it.

4. Verify industrial specs separately

Fiberglass mesh fabric looks identical across grades. What varies is alkali content, mesh count per inch, and coating weight. We once approved a batch at 8×8 mesh that should have been 9×9. On a 2,000-square-meter order, that's a structural problem, not a cosmetic one.

Checkpoint: Match the batch against ASTM D5034-21 or ISO 13934-1:2013 tensile requirements, depending on the application. Ask for the mill's test certificate, then verify with your own spot test. Don't accept "meets industry standard" as an answer—ask which standard, which clause, which test method.

5. Check the label, not just the product

Even guest amenities need this step. We stock a hair serum—the instructions for how to use chi silk infusion are printed on the back of the bottle. If that copy doesn't match the technical data sheet the supplier provided, one of them is wrong. On a 5,000-unit order, wrong instructions mean wrong usage, which means complaints.

Checkpoint: Compare label copy, TDS, and the marketing claim sheet side by side. Any mismatch is a red flag, even if the product itself passes physical inspection.

6. Run a blind comparison with your own team

The numbers said go with Vendor B—same specs, 15% cheaper. My gut said their responsiveness felt off. So we ordered samples from both and ran a blind test with six team members.

Five of six picked the "expensive" option as more professional. The cost difference was $1.40 per unit. On an 8,000-unit run, that's $11,200 for measurably better perception. I went with the gut—this time, the data agreed.

That said, this test works best for products where perception matters: apparel, packaging, anything a customer touches. For a fiberglass mesh order, a blind test is pointless—nobody's judging the hand-feel of a structural reinforcement.

Checkpoint: Blind test, at least five people, no brand labels visible. Ask one question: which one feels more premium? Don't explain the cost difference first.

7. Ask for the supplier's own rejection rate

Never expected a mid-tier mill to have a better QC process than the premium one we'd been using. Turns out their internal rejection rate was 3.2% versus 6.8%. The premium vendor was coasting on reputation.

We've since moved about 60% of our volume to them—maybe 55%, I'd have to check the PO log—and the defect rate on incoming goods dropped from 9.4% to 4.1% over two quarters.

Checkpoint: Ask for the last six months of internal rejection data. If they can't produce it, that tells you something.

8. Put the spec in the contract, not the email

We didn't have a formal approval chain for spec changes. Cost us when a "minor" substitution showed up in production—same color name, different dye lot. The redo was $22,000 and a three-week delay.

Checkpoint: Any change to fiber content, dye lot, mesh count, or label copy requires written approval from quality, not just purchasing.

9. Schedule the re-check before the first delivery

This is the step most teams skip. A pre-production sample review, 10 days before the first shipment, catches more issues than any incoming inspection. 5 minutes of verification beats 5 days of correction.

Checkpoint: Put it on the calendar before the PO is signed. If I remember correctly, we've caught about 40% of issues at this stage—before goods ever shipped.

Three mistakes that cost us the most

Skipping the sample review to save on freight. Saved $80 by skipping expedited shipping on a pre-production sample. Ended up spending $400 on a rush reorder when the sample missed our approval window.

Letting the supplier define "industry standard." When a vendor says a measurement is "within industry standard," ask which standard, which clause, and which test method. If they can't answer in one email, it's not a standard—it's a phrase.

Trusting the first batch. First batches get extra attention at the mill. Batches two through ten are where consistency lives or dies. Re-check at least once per quarter, not just at onboarding.

What this checklist can't do

It won't catch everything. Some issues only show up after 20 wash cycles, or after a season of UV exposure, or after the customer's own finishing process. And it adds roughly 40 minutes per SKU—which, on a 200-SKU year, is about a week of work.

Worth it. The alternative is what we used to do: find the problem after it shipped. That's not quality control. That's quality archaeology.