Bulk jewelry orders drift from the approved sample for six common reasons: the sample was made by the best craftsman, made by a different method, plating drifted, molds wore, materials changed lot, or production was subcontracted. Four controls prevent almost all of it: a golden sample, a numeric specification sheet, a pre-production sample from production tooling, and inspection during and after production. This guide explains each one.
Part of our Custom Jewelry Manufacturing Guide
The sample was perfect. You approved it, paid the deposit, and eight weeks later 500 pieces arrived that are recognisably the same design and unmistakably not the same product. The plating is thinner, the polish is duller, one stone sits crooked. Nothing was substituted and nobody lied — and that is exactly why this keeps happening.
Sample-to-bulk mismatch is the most common serious failure in jewelry manufacturing, and it is almost entirely preventable through process rather than trust. This guide covers why it happens and what to put in place before production starts. Deechy Jewelry Manufacturing & Design runs a 10-step quality control process, and the controls below are the ones any buyer should ask a supplier about.
Why It Happens
Six causes account for nearly every case, and none of them require bad faith.
The sample was made by the best person in the building. Samples are often produced by a senior craftsman with unlimited time. Bulk is produced by a line under a schedule. The sample represents the factory's ceiling, not its average — and you approved the ceiling.
The sample was made differently. A hand-fabricated sample and a cast production piece are different objects even when they look identical. Crisp edges from hand work soften in casting. If your sample was not produced from production tooling, it did not tell you what bulk will look like.
Plating drifts. Plating bath chemistry changes with use. Early pieces in a run and late pieces can differ in thickness and colour unless the bath is monitored and replenished on a schedule.
Molds degrade. Rubber molds soften over successive runs and fine detail disappears gradually. Piece 5,000 is not piece 50 unless the factory tracks mold life and recuts.
Materials came from a different lot. Silver alloy, stones and findings vary between suppliers and batches. Stone colour in particular varies noticeably between lots of the same nominal specification.
Production was subcontracted. The sample was made in-house and the bulk was outsourced when capacity was short. This is the most damaging cause and the hardest to detect after the fact.
The Golden Sample
The foundational control is simple and most brands skip it.
When you approve a sample, produce two identical pieces. Both are physically tagged and sealed with a signed, dated label. You keep one, the factory keeps the other, and it stays on the production floor as the physical reference for the entire life of the product.
This converts every future quality discussion from opinion into comparison. "The polish is worse than we agreed" is arguable. Two physical objects side by side are not.
Photograph both from multiple angles at approval, and store the images with the order file. If the retained sample is ever lost or damaged, the photographs remain.
The Specification Sheet
The second control is writing down what you agreed in numbers rather than adjectives. "Nicely polished" is not a specification. A tolerance is.
| Specify | Example |
|---|---|
| Metal | Sterling silver, nickel-free alloy |
| Piece weight | 4.2g ±5% |
| Dimensions | Stated in mm with tolerance |
| Plating | Rhodium, thickness in microns, nickel-free |
| Finish | High polish, named reference or golden sample |
| Stones | Type, size in mm, colour grade, setting method |
| Findings | Clasp type, nickel-free, jump rings soldered closed |
| Marking | Silver stamp location, logo stamp |
| Packaging | Individual polybag, box specification, labelling |
| Inspection | AQL level and defect classification |
Weight tolerance deserves particular attention. Silver is the largest material cost in most pieces, and a piece running consistently 15% under its specified weight is a piece with less silver in it than you priced. Weight is also the easiest specification to verify — a scale settles it in seconds.
Pre-Production Samples
The single most effective control is requiring a sample made from production tooling, on the production line, before bulk begins.
This is a different object from your original approval sample. The first sample answers "is this the right design?" The pre-production sample answers "is this what the line will actually produce?" — which is the question that matters.
Approve it against the golden sample, in writing, before bulk starts. The delay is typically a few days. The alternative is discovering the answer eight weeks later across 500 units.
Inspection During and After Production
First-article inspection. Check the first pieces off the line, not just the last. Problems caught at piece 20 cost almost nothing; the same problem caught at piece 500 costs the run.
During-production inspection. An inspection at roughly 30 to 50% completion catches drift while there is still time to correct it.
Final inspection to a stated AQL. Agree the acceptable quality limit and the defect classification in advance — what counts as critical, major, and minor. Inspecting to an undefined standard produces an argument rather than a decision.
Verify the invisible things. Plating thickness cannot be assessed by eye, and that is precisely why it is the specification most often quietly missed. Weight can be checked with a scale. Plating thickness needs measurement, and asking for it once tells your supplier you check.
When It Goes Wrong Anyway
Document immediately: photograph the defects against the golden sample, record the proportion of the shipment affected, and quantify the commercial impact.
Then be specific about the remedy you want. Rework, partial credit, replacement on the next order, or a discount to sell the goods as a lower tier — each is reasonable in different circumstances, and a supplier can act on a specific request far better than on a complaint.
How a supplier handles this is the most information you will ever get about them. A factory that acknowledges the problem, explains the cause, and proposes a fix is a factory worth keeping. One that disputes the standard it agreed to has told you what the next order will be like.
If you are planning a custom order, read how custom orders work from brief to bulk, check the golden sample and AQL definitions, and compare finishes across rings and earrings, such as this V-shaped zirconia ring.
Frequently Asked Questions
Why does bulk production differ from the approved jewelry sample?
Common causes include samples being made by a senior craftsman rather than the production line, samples being hand-fabricated while bulk is cast, plating bath chemistry drifting during a run, mold degradation over successive runs, material lot variation, and production being subcontracted when capacity is short.
What is a golden sample?
An approved reference piece produced in duplicate and physically sealed, with one copy held by the buyer and one kept on the factory floor. It converts quality disputes from subjective description into direct physical comparison.
What is a pre-production sample and why does it matter?
A sample made from production tooling on the production line before bulk begins. The original approval sample confirms the design; the pre-production sample confirms what the line will actually produce, which is the question that determines whether bulk matches.
What should a jewelry specification sheet include?
Metal and alloy, piece weight with tolerance, dimensions with tolerance, plating type and micron thickness with barrier layer, finish, stone type and size, findings specification, marking, packaging, and the agreed AQL inspection level with defect classification.
How does Deechy control sample-to-bulk consistency?
Deechy runs a 10-step quality control process covering materials, production stages and final inspection, and bulk production starts only after the buyer approves the sample.
Part of our Custom Jewelry Manufacturing Guide
