Cashmere Procurement RFP: 8 Sections Every B2B Buyer Should Include (Free Template)

A request for proposal that does not specify fiber diameter, down content, count, gauge, and AQL will return eight quotations that cannot be compared. A B2B cashmere buyer who then picks the lowest one has not saved money — they have shifted the variance into quality claims three weeks after loading.

This is the 8-section RFP structure we use, with the specific cashmere specification language that goes in each section, the four questions that most RFPs forget to ask, and the three ways mills respond to an RFP that is written properly versus one that is not.

Why cashmere needs a stricter RFP than most commodities

A buyer sourcing a commodity can often specify it in one line. “5000 kg 100% cashmere yarn” is not a specification. It leaves five commercially decisive variables open, each of which moves the price by 15-40%:

Variable	Range that changes the price	Why it matters commercially
Average fiber diameter	14.5μm / 15.5μm / 16.5μm / 19.5μm	This is the single largest quality axis in cashmere. 14.5μm sells for 1.6-2.1x the price of 16.5μm.
Down content	85% / 90% / 95% / 98%	Sets the hand-feel and the pilling grade. A 95% down yarn cannot be re-labelled 98%.
Yarn count	2/26, 2/36, 2/48, 2/60 Nm	Determines gauge and therefore the application. Wrong count means the knitted fabric is wrong.
Dyeing method	Piece-dyed / yarn-dyed / undyed	Piece-dyed accepts custom colors cheaply. Yarn-dyed does not.
Certification scope	OEKO-TEX 100 / GCS / RWS / none	Certification cost per kg varies by a factor of 3-5.

An RFP that omits these will get quotations that vary by 2-3x in price and are not comparable. The mill is not guessing — each mill is quoting its most standard product within the vague envelope, and the products are genuinely different.

Section 1: Scope and volume

State the total quantity and the delivery schedule explicitly, including whether the quantity is a single delivery or split.

Total: 4,000 kg 100% cashmere yarn, 2/26 Nm, worstled. Delivery: 3,000 kg in one lot, 1,000 kg in one lot 8 weeks later. Quote the split shipment as a single order price and as a per-lot price.

The single-lot vs split-lot distinction matters because most mills price a 1,000 kg lot at a higher per-kg rate than 4,000 kg. Getting both numbers in the RFP lets you compare mills on the basis you will actually order.

Section 2: Fiber specification

This is the section that decides whether quotations are comparable.

Fiber: 100% cashmere. Average fiber diameter 14.5-15.5μm, measured per ISO 18135. Down content 95% minimum per GB/T 25262 or ASTM D4716. Coarse guard hair content below 2% by weight. Cashmere content declared on the mill test certificate, which must be a third-party laboratory certificate naming the test method and sample size.

What most RFPs forget: the sample size and the test method. “14.5μm” without a method is not testable — the micron reading on a different method can be a full micron lower. If the method is not named in the RFP, each mill tests to its preferred method and the readings are not comparable.

Section 3: Construction

Yarn construction, twist, and finish determine performance downstream.

Count: 2/26 Nm worstled. Twist: Z-twist, 480-520 T/m. Finish: none (undyed for piece-dyeing) or specify oiled. Bobbin: 1,000-2,000 g net weight, paper bobbin.

If the yarn is for piece-dyed knitwear, “no finish” is correct. If it is for a mill that wants to commission-dye it, an oiled yarn with a specific oil content is correct. Getting this wrong in the RFP means some mills quote an oiled yarn and others quote un-oiled, and the prices differ by 8-12% for reasons that have nothing to do with quality.

Section 4: Color

Specify the number of colors and the color-matching requirement.

Colors: 8 colors, Pantone TCX references to be supplied. Color matching: ΔE ≤ 1.0 against the TCX reference, measured on a spectrophotometer, across 6 positions per colorway. Lab dips required before bulk production on all 8 colors.

The ΔE tolerance and the number of positions matter. A mill that quotes ΔE ≤ 2.0 without stating positions is quoting a visual match, not an instrumental match. For a brand with retail color consistency requirements, this is the difference between a color that survives the season and a return shipment.

Section 5: Certification

Name the certificate and the scope. Do not write “certified cashmere”.

Certification: OEKO-TEX Standard 100, Class II, covering all 8 colorways and the base yarn. Certificate must be current at time of shipment and must name the mill and the processing facility. GCS (Good Cashmere Sense) certification required with chain-of-custody documentation from combing through spinning.

What most RFPs forget: whether the certificate covers the mill or the brand. An OEKO-TEX certificate held by the mill covers the mill’s product. A brand-specific certificate issued in the buyer’s name has to be requested separately and takes 3-6 weeks.

Section 6: Quality standard and inspection

AQL 2.5 on the finished bobbins per ISO 2859-1, with 5 critical / 10 major / 15 minor defect classification. Third-party pre-shipment inspection to be arranged by the buyer, with 5 business days notice. Fiber diameter verification per Section 2 required at inspection with 10 bobbins sampled.

The inspection notice period matters operationally. A mill that only accepts 3 days notice cannot accommodate a third-party inspector travelling to Ordos.

Section 7: Commercial terms

Payment: [state the method — see our LC vs TT guide for the tradeoff analysis]. Quotation validity: 30 days. Price basis: [state the Incoterm — see our FOB vs CIF guide]. Lead time: from deposit to loading, stated separately from the production lead time.

Separating “production lead time” from “deposit to loading” matters. Most mills quote the second, which includes the payment processing time. A buyer planning around a quoted 45-day lead time who then adds 15 days of L/C issuance and presentation has a schedule slip they did not budget for.

Section 8: Submission format

State exactly what you want back and in what format.

Please submit: (1) unit price per kg by colorway, (2) total order price, (3) production lead time and total deposit-to-loading lead time, (4) fiber diameter and down content with the test method used, (5) certification names and certificate numbers, (6) a sample of the actual yarn with mill test documentation, (7) your AQL standard and defect classification, (8) two B2B cashmere buyer references in the EU or North America that we may contact.

Item 8 is the one most RFPs leave out and most buyers wish they had included. A reference that a mill has provided is a selected reference. A reference obtained independently, at a phone number the buyer looked up, is an actual check.

The four questions worth adding to any cashmere RFP

“What percentage of your last 12 months of cashmere yarn production used fiber in the 14.5-15.5μm band?” A mill that answers “most” without a number is blending rather than sorting. A mill that answers 60% has a genuine sorting operation.
“Can you supply a named third-party lab test certificate for the base yarn before dyeing?” This distinguishes mills that sort from mills that relabel.
“What is your moisture barrier liner specification for container stuffing?” “Dry container” is not a specification. Ask for the liner brand and the RH indicator card.
“If we specify 14.5μm and the delivered fiber tests at 15.5μm, what is your rework or replacement obligation in writing?” This is the only question that tests whether the mill reads the specification at all, and the answer tells you what to write into the contract.

How mills respond to a properly specified RFP

We have read several hundred cashmere RFPs. The pattern is consistent: a well-specified RFP reduces the number of quotations by about half, because roughly half the mills in the market cannot meet the specification as written and decline rather than quote a substitute.

That is the point. The half that decline were going to waste your review cycle. The half that quote are quoting the same thing, so the comparison is real.