How Should the Gold Sample Be Used on a Reorder?
A gold sample is not a souvenir of the approval meeting. It is the physical standard every later run is judged against, and it only works if two exist, both are sealed, one is never opened, and the comparison is written down. This article covers all four, plus when a gold sample has to be replaced.
['MOQ 500 pcs per style and colourway; 300 pcs open to discussion on a repeat.', 'Sampling 6-10 working days from a signed-off tech pack.', 'Bulk production 35-50 days from written sample approval.', 'Inspection at AQL 2.5, ISO 2859-1, General Inspection Level II.', 'T/T 30/70, FOB Xiamen; CIF and DDP available.', 'Two sealed gold samples: one working, one archive, both from the approved run.', 'A gold sample is a standard, not a preference - it is replaced only by agreement.']
Almost every reorder argument ends the same way: someone produces a sample and asks whether the new run matches it. Where that sample is a sealed gold sample from the approved run, the question has an answer. Where it is whatever unit happens to be on the desk, the question is decoration.
The gold sample is the physical half of a tolerance system. The tolerance table is the number; the gold sample is the thing the number refers to. A band without a reference cannot be read, and a reference without a band cannot be enforced. Programmes that have one and not the other spend a season discovering it.
What Is a Gold Sample and What Is It Not?
A gold sample is a unit from the approved run, agreed in writing by both parties, sealed, dated and stored as the reference for later runs. It is the answer to what did we approve. It is not the nicest unit anyone could make, not a sales sample, and not a photograph.
It is also not the same as a reference panel. A panel is a piece of material; a gold sample is a finished bag. Both are needed, because a run can be made from the correct materials and still be assembled differently, and a material comparison alone would pass it.
One more distinction matters in practice. The gold sample is the standard for how the bag is made; the approved drawing is the standard for what the bag is. They answer different questions and neither replaces the other. A run can satisfy the drawing and differ from the gold sample, which is why the drawing alone has never settled a reorder dispute.
Where a programme runs several colourways of one style, each colourway needs its own gold sample. Construction can be shared across colourways on paper, but the physical standard cannot, because shade and trim are exactly the properties that differ between them and exactly the properties a reorder is judged on.
How Many Gold Samples Should Exist and Where Do They Live?
Four, ideally: one working and one archive on each side, all from the same approved run. Two is the minimum - one held by the buyer, one by the production side - because a single sample makes every disagreement a question about whether the holder looked after it properly.
The working sample comes out at every comparison and ages slowly. The archive sample is opened only when the working sample is suspected of having aged or been damaged. Keep them in a dark drawer, not on a shelf in daylight and not in a sample room where they are handled by visitors.
How Is a Gold Sample Sealed, Dated and Stored?
Sealed means tagged and closed so that opening it is visible: a tie, a seal or a signed label across the opening. Dated means the approval date, the run number and the revision are written on the tag. Stored means flat, dry, dark and away from anything that could crush or stain it.
| Handling rule | Why it matters | What happens when it is ignored |
|---|---|---|
| Store flat and dark | Light fades fabric and foam yellows under it | The reference drifts and later runs are judged against a faded unit |
| Tag with run and revision | States which approved run the sample came from | Nobody can say which revision the sample represents |
| Keep two, one unopened | Gives a check on the working sample | A damaged working sample silently becomes the standard |
| Do not use as a sales sample | Handling changes shape, surface and trim | The reference no longer matches what was approved |
| Record every opening | Shows how much handling the sample has had | A change in the sample is mistaken for a change in the run |
| Store away from adhesives | Fumes can mark some coatings and linings | Surface marks read as a production defect |
What Happens When the Gold Sample Ages?
Everything ages: fabric fades, foam yellows, plating dulls, elastic relaxes, and a printed panel shifts slightly. A gold sample that has aged produces a false pass, because the new run is compared against a unit that no longer describes the approval. That is the most dangerous failure mode in the whole system.
The control is the archive sample plus a replacement rule. Compare the working sample against the archive sample once a year; where they differ, the working sample has aged and should be replaced from the archive or from a current run that both parties agree matches the record.
How Is a Reorder Compared Against the Gold Sample?
At a fixed station, under a named light source, with the gold sample and the production unit oriented the same way, at a fixed distance, by a named grader, on a sheet. Those six conditions are what make the comparison repeatable; remove any one and the result depends on who happened to be holding the bag.
Compare the whole bag rather than one panel: overall silhouette, panel alignment, seam symmetry, trim position, hardware finish, artwork placement and hand feel. A run can match on body shade and fail on panel alignment, and a comparison that reads only the shade will pass it.
Take the units for the comparison from the inspection sample rather than from a hand-picked carton. A unit chosen because it looks presentable has already been selected on the outcome being measured, which defeats the exercise. Under ISO 2859-1 at General Inspection Level II and AQL 2.5, the sample size for the lot is known in advance, so there is no reason to improvise.
Photograph the comparison once, from the same angle, with the run number on the sheet. A photograph is not the standard, but it is useful evidence of what was on the table that day, and it is the most economical way to answer a question raised three months later about a run that has already shipped.
What Is Compared by Eye and What by Instrument?
By eye: silhouette, panel alignment, symmetry, seam appearance, trim position, artwork placement and hand feel. By instrument: shade, dimensions, thickness, stitch density, and any property with a written band. The split matters because the eye is good at pattern and poor at magnitude.
| Property | Judged by | Against what | Recorded as |
|---|---|---|---|
| Silhouette and shape | Eye, at a fixed distance | The gold sample beside it | A pass or fail line on the sheet |
| Panel alignment | Eye, plus a rule at named points | The gold sample, and the drawing | A measurement in millimetres |
| Body shade | Instrument, checked by eye | The retained panel and the gold sample | A colour difference and a gray scale grade |
| Dimensions | Steel rule or calliper | The measured values from the approved run | A reading at each named point |
| Stitch density | Count over a fixed length | The count on the gold sample | A count at each named position |
| Hardware finish | Eye under a named light, checked against a retained piece | The retained hardware reference | A pass or fail line and a code |
| Hand feel | Touch, against a retained hand panel | The retained panel from the approved run | Agreement recorded on the sheet |
How Is the Gold Sample Used at First-Piece Sign-Off?
It sits on the table. The first pieces off the line are laid beside it and read before volume starts, with the same sheet used at final inspection. That is the entire mechanism, and it catches the two things a later inspection cannot: a machine that was left set from the previous style, and an operator who has not made this style before.
Where the first pieces differ from the gold sample, the fix is a setting in most cases and a material in a few. Distinguishing them at that moment is worth more than any later check, because the cost of a correction scales with the number of units already made.
Who Holds the Gold Sample and Who Approves a New One?
Both parties hold one, and the holder is named on the spec file. Neither side may replace a gold sample unilaterally: a new gold sample is created by agreement, from a named run, on a dated record, and the old one is retained rather than discarded so the chain stays readable.
That last point is counter-intuitive but important. Discarding the previous gold sample breaks the chain between run one and the current run, which is exactly the comparison a programme needs when a slow drift is suspected. Keep the old one, marked as superseded, in the archive box.
When Should a Gold Sample Be Replaced?
Four triggers: the working sample has aged when checked against the archive; a class two or higher change has been accepted and a new reference is needed; the style has been re-sampled after a construction change; or more than the agreed period - a year is typical - has passed since the sample was sealed.
Four triggers for a new gold sample
- The working sample measures differently from the archive sample on any property with a band.
- A class two or higher change has been accepted, and the outcome was a revised reference.
- The style has been re-sampled after a construction, material or assembly change.
- The agreed retention period has passed and the sample has been in regular use.
A new gold sample should be taken from a production run rather than from a fresh sample, wherever possible. A fresh sample is made carefully by the people who know the style; a production unit is what the customer will receive. Using a production unit keeps the standard honest.
Seal one sample per construction as well as per colourway. Where a style is made in two constructions, a single gold sample cannot describe both, and a comparison made against the wrong one produces a fail that has nothing to do with the run.
Record the reason on the new tag. A gold sample that carries the reason it was created tells the next reader whether it represents a changed product or simply a fresh copy of an unchanged one, which is the question that matters most when a drift is being traced across several years of runs.
What Does the Gold Sample Record Contain?
One record per sample: the style and colourway, the run number it came from, the revision, the date sealed, who sealed it, where it is stored, whether it is working or archive, the date of each annual check against the archive, and the date it was superseded if it was.
Add one field that is easy to forget: which runs the sample has been used to judge. A record that shows the sample was used on runs two, three and four tells a reader which comparisons share one standard, and it is the first thing to check when a slow drift is suspected across several runs.
Keep the record with the sample rather than in a folder elsewhere. A card in the archive box, updated when the box is opened, survives staff changes and office moves in a way that a spreadsheet on someone machine does not.
- Style and colourway
- As written on the spec file, with the revision number
- Source run
- The run number and date the sample was taken from
- Sealed by and date
- Named on both sides, on the tag and on the record
- Working or archive
- Which role this sample serves
- Storage location
- Where it is kept, and the conditions
- Annual check
- Date and result of the comparison against the archive sample
- Superseded
- Date and reason, with the replacement run number
A gold sample that comes out of the drawer every week is a working tool. One that comes out once a year and has never been compared against its archive is a story.
Send us the gold sample rule
Tell us how many sealed references you hold and where they live. We coordinate production against a working and an archive sample per style, bulk in 35-50 days from written sample approval, inspected at AQL 2.5 under ISO 2859-1.
Frequently Asked Questions
What is a gold sample?
A unit from the approved run, agreed in writing by both parties, sealed, dated and stored as the physical standard for later runs. It is the reference a tolerance band points at, and it is not the nicest unit anyone could make or a sales sample.
How many gold samples should exist?
Four is ideal and two is the minimum: one working and one archive on each side, all from the same approved run. Two matter because a single sample makes every disagreement a question about whether the holder looked after it properly.
How should a gold sample be stored?
Flat, dry and dark, away from daylight, adhesives and anything that could crush or stain it. Light fades fabric and yellows foam, and a faded reference produces a false pass when later runs are compared against it.
What happens when a gold sample fades?
It produces a false pass, because the new run is being judged against a unit that no longer describes the approval. Compare the working sample against the archive sample once a year and replace the working sample where the two differ.
How is a reorder compared against the gold sample?
At a fixed station, under a named light source, with both units oriented the same way, at a fixed distance, by a named grader, on a sheet. Compare the whole bag - silhouette, alignment, trim position, hardware finish, artwork and hand feel - not one panel.
What is judged by eye and what by instrument?
By eye: silhouette, panel alignment, symmetry, seam appearance, trim position, artwork placement and hand feel. By instrument: shade, dimensions, thickness, stitch density and any property with a written band. The eye reads pattern well and magnitude poorly.
How is the gold sample used at first-piece sign-off?
It sits on the table beside the first pieces off the line, read before volume starts on the same sheet used at final inspection. That catches a machine left set from the previous style and an operator who has not made this style before.
Who may approve a new gold sample?
Both parties, by agreement, from a named run, on a dated record. Neither side may replace a gold sample unilaterally, and the superseded sample should be retained in the archive box so the chain between run one and the current run stays readable.
When should a gold sample be replaced?
On four triggers: the working sample differs from the archive on any bounded property; a class two or higher change was accepted with a revised reference; the style was re-sampled after a construction change; or the agreed retention period has passed.
Should a new gold sample come from a fresh sample or a production run?
From a production run wherever possible. A fresh sample is made carefully by the people who know the style; a production unit is what the customer receives. Using a production unit keeps the standard honest about what the line actually holds.
Is a photograph enough as a reference?
No. A photograph cannot be handled, cannot be measured at a named point with a calliper, and its colour depends on the camera, the screen and the light it was taken under. Photographs are useful supporting evidence and useless as the standard itself.
What is the difference between a gold sample and a reference panel?
A panel is a piece of material and a gold sample is a finished bag. Both are needed, because a run can be made from the correct materials and still be assembled differently, and a material-only comparison would pass it.
People Also Ask
What is a gold sample in bag production?
A sealed, dated unit from the approved run, held by both parties as the physical standard for later runs. It is the reference that a written tolerance band points at, and it is replaced only by agreement and on a dated record.
How do you store a gold sample properly?
Flat, dry and dark, tagged with the run and revision, one working and one archive copy, opened only for a comparison and never used as a sales sample. Record each opening and check the working sample against the archive annually.
How often should a gold sample be replaced?
On four triggers: ageing detected against the archive, an accepted class two or higher change, a re-sample after a construction change, or the end of the agreed retention period - a year is typical.
Can a photo replace a gold sample?
No. A photograph cannot be handled or measured, and its colour depends on the camera, the screen and the light. Photographs support a comparison; they cannot be the standard that the comparison is made against.
Who should keep the gold sample?
Both parties, ideally with a working and an archive copy each, from the same approved run. A single holder makes every later disagreement a question about whether that sample was looked after properly.
Sources and further reading
Every figure on this page is either our own programme data or attributed below. Open the originals before you write them into a specification.
- AATCC evaluation procedure 1 — gray scale for colour changeAmerican Association of Textile Chemists and Colorists
- ISO 2859-1:1999 — sampling procedures for inspection by attributesInternational Organization for Standardization
- ISO 105 series — tests for colour fastness of textilesInternational Organization for Standardization