The Sampling Calendar: From Proto to TOP

This page is for the founder or ops lead who has a tech pack and a factory and now needs to know what happens between them, in what order, on whose signature, and at what cost. Sampling is where most first-time development budgets are quietly destroyed, because each round looks cheap and the rounds multiply. It sits inside the wider question of sourcing apparel direct from factories, and it depends entirely on the quality of the document you send first: see what a complete tech pack contains.

The eleven garment samples, from mock-up to press sample

Every sample in the sequence answers one question and has one approver. Confusing the questions is what produces the classic failure: a beautiful proto in the wrong fabric that fits nobody and cannot be made on a line.

Sample sequence, purpose and approver. The full development set as used by export factories in India, Bangladesh, Vietnam and China.

Not every style needs all eleven. A repeat tee in an approved colour may need only PP, TOP and a shipment sample. A technical jacket will use every one and add fit rounds.

Why the PP sample is the contractual gate on bulk production

The PP sample is the only sample in the sequence that is simultaneously made on the production line, in the bulk fabric, and with the bulk trims. That combination is the point. A proto proves the design. A fit sample proves the pattern. Neither proves that the mill's actual dyed fabric lot behaves like the sampling-room fabric, that the bulk zipper tape matches the shell, or that the line operators can execute the construction at rate.

Write it into the purchase order in plain words: no fabric is cut for bulk until the brand approves the PP sample in writing. The commercial reason is blunt. Once the marker is laid and the cutter runs, the fabric is committed. A defect discovered after cutting is a rework problem across the whole order rather than a sample problem across one garment, and rework after a failed inspection typically costs 4–9% of order value. Approving PP is the last moment the mistake is still cheap.

The approved PP becomes the sealed sample: signed, sealed, one copy at the factory and one at your desk. Everything downstream, including the TOP sample and any third-party AQL inspection for brand owners, is judged against that physical object rather than against a PDF.

The GPT sample is the one brands skip most often and regret most expensively. It is the sample that goes to a lab for colourfastness to washing and crocking, shrinkage and skew. Skipping it means the first wash test your garment ever gets is performed by a customer.

Using a sample maker Los Angeles brands can visit in person

The garment sampling process above assumes every sample is made where the bulk will be made. Plenty of US founders run the early rounds differently, at an independent sample room close to home. In the LA Fashion District, patternmakers and sample makers occupy the upper storeys above the wholesale showrooms, and a sample room there does development and patternmaking only — it does not run production. That distinction is the whole reason the option exists, and it is set out in what the LA Fashion District actually contains.

The split works, with one hard limit. A domestic sample maker can carry samples 1 to 3 in the table above — mock-up, proto and fit — and the value there is real: you attend the fit session, you handle the garment the same week, and a comment costs an afternoon instead of a courier round trip. What a domestic sample room cannot produce is a valid PP sample, because the PP has to be made on the production line, in the bulk fabric, with the bulk trims. If your bulk is offshore, the PP sample, the sealed sample and the TOP sample are offshore too, whatever you spent on the proto in California.

So treat it as a sequence rather than a choice: develop the pattern locally if iteration speed matters more than sample cost, then re-sample at the production factory from the graded pattern and the tech pack, and budget for both sets of rounds rather than one.

How a garment sampling lead time builds, stage by stage

Lead-time build for a first order, PO to FOB. Stage durations for an export factory in Asia; the ranges widen in Q3 peak season. Every stage down to QC and inspection is inside the PO-to-FOB clock; the shipping line is what transit then adds on the way to a US DC.

The line that surprises people is fabric. Custom dyeing adds 4–8 weeks on its own, which is why colour development has to start the same day as fit development rather than after it. See lab dip and Delta E colour approval for how those weeks are actually spent.

Total garment lead times by complexity: PO to FOB versus PO to DC

Say which clock you are on, because the two in circulation differ by more than a month. PO to FOB runs from purchase order to goods on board at the origin port. PO to DC runs from purchase order to units sellable in your US distribution centre, and it adds ocean transit, customs clearance, drayage and receiving.

First orders land at 12–20 weeks PO to FOB globally, and 10–16 weeks PO to FOB from Bangladesh or India. Ocean transit and clearance add roughly 5–7 weeks on top, so the same first order measured PO to DC is about 15–20 weeks from India and 18–23 weeks from Bangladesh. By complexity, PO to FOB: a basic tee runs 8–12 weeks, a dyed hoodie 10–14, washed jeans 12–18, and jackets 14–20. Bangladesh PO-to-FOB runs 75–120 days on knits and 90–130 days on denim and complex wovens, extending 15–25% in Q3 peak. Vietnam runs 16–20 weeks PO to FOB on a first order and 8–12 weeks on a reorder.

Repeat orders compress to 8–12 weeks PO to FOB because sampling is eliminated. Nothing else changes. The tech pack is frozen, the pattern is graded, the lab dips are approved, the sealed sample exists, and the factory already knows the construction. What remains is fabric booking, bulk production, inspection and transit. That gap between a 16-week first order and a 10-week repeat, both PO to FOB, is the single most valuable thing a brand owns after a successful first production, and it is the number that should drive reorder points under 90-day lead times.

Calendar landmines: Chinese New Year, Tết, Eid and Diwali

Factory calendars are not evenly distributed, and the holidays that matter are the ones that fall between PP approval and bulk cut.

• Chinese New Year (January or February) removes 2–4 effective weeks. Capacity tightens for weeks beforehand as everyone tries to ship before the shutdown.
• Tết in Vietnam removes 1–2 weeks, with bookings needing to be placed 3–4 weeks ahead.
• Eid in Bangladesh occurs twice a year, each costing roughly 1–2 weeks.
• Diwali in India shuts down clusters including Tirupur and Delhi NCR.

Return rates after a long holiday are never 100% on day one, so treat the week after a shutdown as partial capacity rather than full.

What sampling costs: development pricing by garment

First-sample cost by garment type. Indicative export-factory pricing; excludes courier, fabric development and lab dips.

With the 2–4 rounds most styles genuinely need, all-in development lands at $900–$5,000+ per style, and a 30-style collection runs $27,000–$150,000 before a single bulk unit exists. Lab dips add $15–50 per colour, 5–10 business days per round, with 2–3 rounds standard and 4–5 rounds normal for saturated reds, neons and deep navies.

And the number that should govern how many styles you develop: roughly 60% of physical samples never reach production. Development spend is not a production cost, it is a portfolio bet, and more than half of it is written off by design.

Worked example: development budget for a 12-style capsule

Twelve styles, six colours across the range, three rounds each (proto, fit, PP). Sample costs at the midpoint of the ranges above.

That $1,354 per style sits at the low end of the $900–$5,000 band, which is what a disciplined three-round programme on simple product looks like. Add one extra fit round across the two denim styles, at 2 × $600 = $1,200 for the round, and you are at $17,451. Add a fourth round across everything and you are at $21,476, a 32% increase from one lost round. Sample courier and air freight sit on top and are billed separately. Note also that the six colours in this example each carry their own MOQ and dye-lot minimum economics at bulk, which is a separate and larger constraint.

How to run a sampling round efficiently

1. Freeze and version the tech pack before you send it. Send one locked version number with a dated revision log. Mid-round changes are the single largest cause of wasted samples. If a change is unavoidable, issue a new version number and reset the round rather than emailing an amendment.
2. Send colour and fabric development in parallel, not in series. Release lab dips and knit-down requests the same day you release the proto request. Colour takes 5–10 business days per round and custom dyeing adds 4–8 weeks, so it must run alongside fit development.
3. Run the fit session against the spec, not against opinion. Measure every point of measure, record actual against spec and tolerance, and photograph front, back and side on the fit model. Decide fit against the numbers first, then judge aesthetics.
4. Consolidate comments into one dated sheet within 48 hours. One person owns the sheet. Every comment names the point of measure or construction callout it changes, gives the new value, and is marked accept, revise or reject.
5. Require the PP sample in bulk fabric and bulk trims. Reject any PP made in sampling-room fabric or substitute trims, and confirm the bulk fabric lot number and trim supplier references on the sample tag before evaluating it. What counts as the right fabric is defined in how to write a fabric specification.
6. Seal the approved sample and hold a matched pair. Sign and seal it, factory keeps one, you keep one. Every later judgement runs against that physical standard.

Definitions for proto, GPT, sealed sample and points of measure are collected in the Yarnstick glossary of sourcing terms.

Frequently asked questions