Do I need a golden sample before bulk production?

Yes. A golden sample is the physical reference that mass production is measured against. Without a signed reference, any later disagreement about colour, dimension or finish becomes a matter of opinion instead of a comparison against an agreed standard.

How many sampling rounds are usually needed?

Most projects settle in one or two printed rounds: the first confirms structure and print, the second corrects specific deviations. Rounds beyond that usually mean the brief was ambiguous, most often on colour target or internal dimensions.

Can I approve a sample from a photograph?

Not for colour or material. A photograph cannot reproduce light-booth conditions, surface finish or material gauge. Photographs are useful for confirming structure and print position; colour and material have to be judged on the physical sample.

What is the minimum order quantity after sampling?

ChromaPack's standard minimum order quantity is 500 pcs per design and size. Completing a sampling round does not change that threshold.

Who keeps the golden sample?

Both sides should hold a signed copy: the supplier keeps one to produce against, the buyer keeps one to inspect the delivered goods against. Record the sample reference on the purchase order so the link is documented.

Custom Packaging Sampling Playbook: From Sample Request to Approved Production Standard

September 17, 2026 — 15 min read — By ChromaPack Engineering Team

Short answer: a custom packaging sample is not one artefact but five, and each one proves something different. Approve them in order — structural drawing, unprinted structural sample, printed sample, then golden sample — and record what you approved at each gate. Almost every "the bulk order looks wrong" dispute traces back to a gate that was skipped or signed off verbally rather than against a written standard.

What a "sample" actually means: five artefacts, five different jobs

Buyers and suppliers both say "sample", but they usually mean different things. This is the single most common source of confusion in a packaging project, and it is cheap to fix: name the artefact you are asking for.

| Artefact | What it proves | What it cannot prove | Who signs it off | |---|---|---|---| | Artwork proof (PDF) | Spelling, dieline fit, barcode placement, legal panel content | Colour, material, structure behaviour | Brand / marketing | | Dieline / structural drawing | Dimensions, fold lines, bleed, glue areas, panel layout | That the material will perform as drawn | You + supplier engineering | | White structural sample (unprinted) | Fit and function: does the product sit correctly, does the closure hold, does the insert retain | Print quality or colour | You | | Printed sample (pre-production) | Colour, registration, finish, material gauge, overall look and feel | Long-run consistency (it is not from the production run) | You | | Golden sample (signed reference) | The single agreed standard that mass production is measured against | Nothing further — it is the standard | You + supplier, both keep a copy |

The mistake is treating a printed sample as if it were a golden sample. They look identical. They are not: a printed sample is a proposal, a golden sample is a contractual reference.

Stage 0 — Write the sample request brief before you ask for anything

Most wasted sampling rounds come from a vague request, not from a bad factory. Before you ask for a sample, put these nine items in writing:

Product dimensions and fill weight — and whether your dimensions are internal or external.
Product form — solid, powder, liquid, paste, device. This decides barrier requirement more than anything else.
Intended shelf life and distribution — ambient, chilled, frozen; retail shelf or e-commerce parcel.
Material family you want to evaluate, or explicitly "recommend one and justify it".
Print requirement — how many colours, which surfaces are printable, and whether you need variable data.
Colour target — a Pantone reference for brand colours, or "match to supplied artwork".
Finishing — matte, gloss, soft-touch, foil, spot UV, emboss.
Regulatory market — which country the product is sold in, and therefore which food-contact or device rules apply.
Quantity intent — the order size you expect to place after approval.

Item 9 matters more than buyers expect. A supplier quotes and samples differently for 500 pcs than for 50,000, because tooling, plate setup and material purchasing change with volume.

Stage 1 — Structural approval: measure, do not eyeball

The white sample is a dimensional and functional test. Judge it with instruments, not by feel.

Measure and record:

External and internal dimensions at three points per axis. Packaging deforms at glue seams; a single measurement hides that.
Wall thickness and material gauge. If the drawing says 350 g/m² and the sample reads light, the difference will multiply across a production run.
Closure performance: magnetic pull strength, tuck flap retention, zipper engagement, valve function.
Insert fit: does the product move when the closed pack is inverted and gently shaken?
Assembly time: how long to erect one pack. On a 10,000-unit order, ten extra seconds per pack is real labour.

Tolerance is a specification, not a complaint. Ask for the tolerance band on every critical dimension in writing before approval. If you have not agreed a tolerance, you cannot reject a shipment for falling inside it — and you cannot defend a rejection if you do.

Stage 2 — Print and colour approval: the gate that fails most often

Print approval is where projects slip, because colour is subjective unless you make it objective.

What to check on a printed sample:

Registration — are the colour plates aligned, particularly on small type and fine rules?
Colour against target — compare under standard D50 or D65 lighting, not office lighting and not on a phone screen.
Surface finish uniformity — lamination haze, foil edge definition, spot UV registration to the printed layer.
Material gauge and hand-feel against the agreed specification.
Reproduction of fine detail — small reversed type, thin serif fonts, fine barcodes.

Two practical rules:

A screen proof is not a press proof. RGB screens and CMYK presses have different gamuts. What appears as one brand blue on screen may be out of gamut on press. If the colour is critical, approve a physical proof of it.

Brand colours should be specified as Pantone, with a documented CMYK fallback for process-printed work. This is what makes reorders match — and reorder drift is the complaint buyers actually feel six months later.

Stage 3 — Functional and compliance validation

Structural and print approval do not cover performance. Add the checks that match your product:

| Product type | Test to request | What you are looking for | |---|---|---| | Coffee / roasted food | Valve function and reseal test | One-way degassing works; aroma barrier intact after opening and reclosing | | Powders and dry food | Seal integrity and leak test | Continuous seal, no channel leaks, no weak spots at the fin | | E-commerce / parcel | Drop and vibration behaviour | Product survives the last mile, not just the pallet | | Medical devices | Sterile barrier integrity per ISO 11607 | Seal strength and visual seal continuity documentation | | Retail gift and rigid | Compression and corner strength | The box looks new after stacking and handling |

Ask for the test reports, not just the test. A finished report with sample identification, method and result is a document you can show your own customer or auditor. "We tested it" is not.

Stage 4 — Golden sample sign-off

The golden sample is the point where the project becomes enforceable. Do this properly:

Approve against the written standard — specification sheet, colour target, tolerance band, material, finish.
Sign and date the sample, on the sample itself where possible, and record a reference number.
Put the reference number on the purchase order. Without this, the link between the approved sample and the bulk order exists only in email.
Keep one copy each. The supplier produces against theirs; you inspect against yours.
Photograph the sample under controlled light and store it with the approval record — useful when a dispute happens months later and the physical sample has been handled to death.

The five sampling mistakes that cost the most

| Mistake | What it costs | Prevention | |---|---|---| | Approving colour from a phone screen | Entire print run of the wrong shade | Approve a physical proof; specify Pantone with CMYK fallback | | No written tolerance | Unenforceable rejection; disputed invoice | Agree the tolerance band before sample approval | | Verbal approval by email thread | No enforceable standard; rework at your cost | Signed golden sample with a PO reference number | | Sampling without stating order quantity | Sample optimised for a different scale of production | State expected order size in the brief | | Skipping the unprinted structural round | Print approved before fit is proven; expensive re-sampling | Approve structure first, then print |

Sampling cost, MOQ and the sample charge

Two commercial questions come up every time.

Does the sample charge come back? Some suppliers deduct sample and tooling charges from the bulk invoice once the order is placed. Ask for this in the quotation, in writing, rather than accepting it verbally at sampling stage — it is a commercial term, not a technical one.

Does sampling change the MOQ? No. Sampling is an approval step, not a volume commitment. ChromaPack's standard minimum order quantity is 500 pcs per design and size, whether or not a sampling round took place.

What actually drives the sampling timeline

Sample turnaround is driven by three things, in this order:

Your brief quality. A complete brief with dimensions, colour target and material direction is the single biggest lever. Vague briefs produce samples that answer the wrong question.
Whether new tooling is required. A structure that needs a new cutting die or a new print cylinder takes longer than one produced on existing tooling.
The product family. Flexible packaging settles faster than rigid structures because rigid boxes involve board, wrap, insert and assembly as separate steps.

Published sample turnaround for our own product lines runs from 3–5 days for flexible packaging to 5–7 days for rigid and cosmetic box structures, measured from approved artwork and dieline. Medical sterile projects follow the applicable ISO 11607 validation steps before a sample can be approved at all.

The honest summary: you control the first factor, and it is the largest one.

Where to go next

Packaging sampling guide — the wider sampling process
Golden sample approval checklist — the sign-off document
Custom packaging quote information — what to send for a fast, comparable quote
Export logistics playbook — what happens after the golden sample is signed

Custom Packaging Sampling Playbook: From Sample Request to Approved Production Standard

Custom Packaging Sampling Playbook: From Sample Request to Approved Production Standard

What a "sample" actually means: five artefacts, five different jobs

Stage 0 — Write the sample request brief before you ask for anything

Stage 1 — Structural approval: measure, do not eyeball

Stage 2 — Print and colour approval: the gate that fails most often

Stage 3 — Functional and compliance validation

Stage 4 — Golden sample sign-off

The five sampling mistakes that cost the most

Sampling cost, MOQ and the sample charge

What actually drives the sampling timeline

Where to go next

Buyer questions on this stage