Module 4: Sampling & Negotiation
Samples are where paper meets reality. This module climbs the sample ladder — proto, fit, pre-production — reviews each round against the tech pack, negotiates the whole deal (not just unit price), and tests the product the way customers will abuse it.
The sample ladder: proto, fit, pre-production
Sampling is a ladder, and each rung answers a different question. Skipping rungs does not save money; it asks expensive questions of the most expensive rung — bulk production.
Proto sample — “can this be made at all, roughly like this?” Often built in available materials and colors, it proves construction and silhouette. Judge shape and construction only; founders who reject a proto for its substitute fabric are grading the wrong rung. Fit / size sample — “does it fit and measure to the chart?” Made closer to spec, in at least the base size, sometimes the full size run. You measure every point against Page 3 of the pack and mark corrections on the pack itself — the revision log starts earning its keep here.
Pre-production (PP) sample — “is this exactly what bulk will be?” Final materials, final artwork, final packaging, made on the production setup. This is the rung that matters most: the approved PP becomes the golden sample, the physical standard your inspection compares bulk against. Approve it in writing, keep one sealed copy yourself, and leave one with the factory. Production / shipment sample — pulled from the actual bulk run before shipping; your last cheap look at reality.
Budget honestly for the ladder: samples are usually charged at one-and-a-half to three times unit cost, plus courier fees, and two to four rounds is normal for a first product. A factory that promises no revisions will be needed is selling, not sampling. Put sample costs and round-trip times into the launch calendar now — sampling is where launch dates are actually set, four to ten weeks before anyone mentions freight.
Your proto sample arrives in a substitute grey fabric because your custom-dyed yarn is not ready. The construction and shape are right. What is the correct review?
Reviewing a sample like a factory does
Scenario: the PP samples land on your desk on a Friday, and they look great on the hanger. By Monday, three are gifts for friends and the order is approved. This is the most common and most expensive review process in product development, and it finds nothing.
Review on paper, in this order. Measure first, before you form an impression: every Page 3 dimension, recorded next to its tolerance, on at least two units. Numbers do not get charmed by nice packaging. Compare to the pack page by page — BOM materials (feel, weight, composition if labeled), construction points, artwork placement measured from its fixed points, label wording letter by letter. Then wear it, wash it, use it — the functional review: does it do the job it promises, on a body, in motion?
Write comments the factory way: one list, numbered, each item tied to a page of the pack (“Page 3: cuff width 7.3 cm, spec 8.0 ± 0.4 — correct”), each marked must fix or note. Photograph every defect next to a ruler. Vague feedback — “make it feel more premium” — returns vague revisions and burns a two-week round trip. Precise feedback returns one clean round.
Decide per item what happens next: correct and re-sample (critical specs, always), correct in bulk with a photo/video check before production (minor, verifiable), or accept and amend the pack (when the sample is honestly better than your spec — it happens, and the revision log is where the pack learns). What you may not do is approve silently and hope bulk guesses your reservations.
Your turn: Write your sample review sheet: the measurements you will take, the pack pages you will check in order, and the must-fix / note labels you will use.
Saved.
A sample measures 7.3 cm on a dimension specced at 8.0 ± 0.4 cm. The product otherwise looks and feels excellent. Correct call?
Negotiate the whole deal, not just the price
Unit price is one line of a deal that has at least seven. Founders who negotiate only the price win a cheaper product on terrible terms; founders who negotiate the deal often keep the price and win everything else. The lines: MOQ structure, price breaks at the next quantities, payment schedule, sample and revision costs, lead time with a ship date, defect terms (what counts, who pays, remake or credit), and packaging inclusions. Each is money wearing a different coat.
The strongest lever you own is information, not pressure. You hold three to five quotes on the same spec — a factory that knows you are comparing (it does) needs a reason to sharpen. Ask with a target and a trade: “At $3.40 we can start at 1,000 and reorder quarterly; at $3.75 we must start at 500. Which structure do you prefer?” You are not bluffing about a number — you are offering a shape of business, and factories routinely trade price for volume certainty, simpler specs, or faster payment.
Cost engineering beats cost squeezing. Ask the Module 3 question again, pointed: “Which line of my spec costs the most, and what would you change?” A half-cushion sole, a stock yarn color one shade off your custom dye, a belly band instead of a box — specialists find savings inside the spec that keep the promise and cut the price, because they know where money hides in their own process. A factory that simply agrees to any price, conversely, will recover its margin where you cannot see it: thinner material, looser tolerances, slower line priority. A price cut with no visible source is a spec cut in disguise.
Close in writing. The final agreement — spec revision, price, MOQ, terms, defect handling, ship date — goes in one document both sides sign or confirm, and it becomes part of the development file next to the pack. Every later disagreement in this business is settled by documents; make sure yours exist while everyone is still friendly.
A factory instantly agrees to your requested 20% price cut, no spec change discussed. Best interpretation?
Testing: wash, wear, and break it on purpose
A sample that survives your desk review has proven it can be made. Testing proves it can be used — for as long as your promise implies. The method is unglamorous: use the product the way your harshest reasonable customer will, count the cycles, and write down when it fails.
For a grip sock, the protocol writes itself from the promise. Wash it fifty times the way the care label says — grip adhesion, elasticity, color, and shrinkage measured against a fresh pair every ten washes. Wear-test in the actual activity: reformer sessions, counted, noting where the grip pattern misses the contact points. Stretch, snag, and pull at seams. Our working standard came straight from the commonest customer complaint in the category — grips peeling within a month — which at three washes a week is roughly twelve washes. A product tested to fifty can promise a month with a straight face and deliver a season.
Match the test to the critical specs you starred in Module 2. Every critical spec gets a test with a number: not “grip seems strong” but “grip intact and functional after 50 washes”; not “sizing feels right” but “foot length inside tolerance after washing, on all sizes”. Untested critical specs are just hopes with bold formatting.
When a test fails, you have found the cheapest possible version of that failure — the whole point. The fix goes into the pack (a grip spec change, a material upgrade, a care instruction rewrite), the sample round repeats on that point, and the cost is a courier fee. The same failure at bulk scale is a recall, a review-page disaster, and a warehouse of evidence. Break it yourself, early, on purpose.
Your grip passes 12 washes, then starts peeling. Your care label and ads imply a season of regular use (about 36+ washes). What do you do?