A retail chain in Phoenix had a 14% tarnish complaint rate and the cause was a 0.02 micron rhodium layer, invisible on arrival and gone by the second month. The order was not lost on price. It was lost on a specification nobody wrote down. This piece looks at industry insights from the buyer's side of the table: what the promise means at the bench, what it costs to get it wrong, and how to write a brief a factory can be held to.
Tarnish is a chemistry problem, not a care problem. Silver reacts with sulphur, and the two things that control how fast are plating thickness and the atmosphere inside the packaging.
Telling a customer to avoid perfume is advice. Shipping a piece with 0.05 to 0.10 micron rhodium and a sulphur-free pouch is a fix.
The default offer in this category is a stock photo, a unit price and a thirty-day promise. That works for one order and stops working on the second, because a stock photo cannot specify cable.
| Comparison | Common practice | What we hold | Why it matters |
|---|---|---|---|
| Cause | Effect | Control | |
| Thin rhodium | tarnish in 30-60 days | 0.05-0.10 micron minimum | |
| Sulphur pouch | accelerated tarnish | vapour-phase inhibitor, no sulphur | |
| Chlorinated water | spotting on gold | rinse and dry before storage | |
| Abrasive cloth | plating wear | soft cloth, no polish on plate |
Pull strength at the clasp is the number we publish per lot. It is the one a buyer can check with a caliper and a loupe in ten minutes, and the one that decides whether a reorder is a formality or a negotiation.
We inspect at AQL 2.5 on the final line and 3 to 5 per cent on setting, sampled per carton rather than per lot. Per-carton sampling is stricter because a defect that clusters in one carton is a process problem, not a random one.
Anything above the AQL is held and reworked before packing. We do not sort after arrival, because sorting after arrival is a cost the buyer pays and a defect the buyer discovers in front of a customer.
At the receiving table: count per SKU against the packing list, weigh three pieces, measure the necklace, check one stone at 10x for seat and tilt, and confirm the plating colour against an approved sample.
Five checks, ten minutes. They catch the defects that are visible on day one and free the buyer to argue about the ones that are not, with evidence rather than impression.
Read the specification as a checklist: cable, curb and box chains at 16/18/20 inch with a lobster or spring clasp rated to 25 N. Each comma is a decision someone made, and each decision has a cheaper alternative that a factory can substitute without telling you.
The substitution is usually invisible on delivery. It shows up as a hue shift in month three, a prong that will not hold after a resize, or a plating that reads slightly warm under store lighting.
We measure pull strength at the clasp on every lot and record the number against the rack. It is the one figure a buyer can independently verify with a caliper, a loupe and ten minutes at the receiving table.
If the number is inside the band, the lot ships. If it is outside, it goes back to the bench rather than into a carton with a note that says the buyer should expect variation.
A returned necklace costs the unit price twice: once in goods, once in handling, plus the freight both ways and the customer's lost confidence. On a necklace the handling is usually larger than the margin.
That is the arithmetic behind every decision on this page. A cent saved on plating is not a saving if it produces a return rate two points higher.
We inspect at AQL 2.5 on the final line and 3 to 5 per cent on setting, sampled per carton rather than per lot. Per-carton sampling is stricter because a defect that clusters in one carton is a process problem, not a random one.
Anything above the AQL is held and reworked before packing. We do not sort after arrival, because sorting after arrival is a cost the buyer pays and a defect the buyer discovers in front of a customer.
The unit price is the visible cost. The invisible ones are returns, size exchanges, replating, markdowns and the labour of photographing a piece that arrived with the wrong tone.
On a typical wholesale programme those five add between eight and fourteen per cent to the real cost of goods. Fixing plating thickness and stone hue band removes most of it before the first season ends.
Capacity is how much a line can make in a month. Availability is how much of that is free in the week you need it. Factories quote the first and deliver the second.
We quote capacity per line and then reserve a slot against your deposit, so the date you are given is a booked date rather than an average.
Testing happens at four points and none of them is the packing table. The casting tree is checked for porosity and short fill, the setting bench for seat and tilt, the plating rack for thickness on a cross-section, and the final line for weight and dimension.
A final inspection alone catches perhaps a third of what goes wrong, because the other two thirds were decided upstream. That is why the record has four columns and not one.
The unit price is the visible cost. The invisible ones are returns, size exchanges, replating, markdowns and the labour of photographing a piece that arrived with the wrong tone.
On a typical wholesale programme those five add between eight and fourteen per cent to the real cost of goods. Fixing plating thickness and stone hue band removes most of it before the first season ends.
Fifty pieces per design is the working minimum on industry insights. A sample runs three to five days, bulk fifteen to twenty days under 500 pieces, and a custom mould twenty-five to thirty days from drawing approval. Payment is 30% deposit with the balance against the bill of lading copy.
If the order is below fifty, the setup cost per piece roughly doubles because the casting tree and the plating rack are fixed costs. That is arithmetic, not a policy to negotiate around.
| Item | Value | Note |
|---|---|---|
| MOQ | 50 pcs / design | sizes and colours can be split |
| Sample | 3-5 days | one piece, approved before the run |
| Bulk lead time | 15-20 days | under 500 pieces |
| Custom lead time | 25-30 days | from drawing approval |
| Payment | 30% / 70% | balance against B/L copy |
| Port | Guangzhou / Shenzhen | export from Wuzhou |
Plate it to at least 0.05 micron rhodium, pack it with a vapour-phase inhibitor, and tell the customer to store it dry. Those three controls cover most of the ninety-day complaint window.
Usually, with a soft cloth and a mild silver dip, as long as the plating has not worn through. Once plating is gone the piece needs replating, not polishing.
No. Gold-filled has a bonded layer measured in microns of gold by weight fraction, and it lasts far longer than a 0.5 micron electroplate, which is what most cheap vermeil actually is.
A single card with four lines: store dry, avoid chlorine, wipe with a soft cloth, and bring the piece in for replating if the colour shifts. Short cards get read.
Yes, if we hold the four inputs: alloy batch, stone parcel, plating tank and setting team. The reorder sheet records all four, so the check happens against a document rather than against a memory.
Photograph the piece with a scale and reference the lot line. Above the agreed AQL we replace and cover freight; inside it we replace the affected units inside ten days.
HOLYCOME by Wuzhou HengliLai Jewelry Co., Ltd. runs casting, setting, polishing and plating on one site in Wuzhou, Guangxi, the artificial gemstone cluster that supplies most of the trade. Monthly capacity is roughly 60,000 pieces across silver and gold lines. Buyers can reach the trade desk at service@holycome.com or on WhatsApp at +86 18078079070, and a factory visit can be arranged with two weeks' notice.
If you are pricing industry insights for a season, send the quantity per design, the stone grade you need and the date you need it on the shelf. We will return a drawing, a landed cost and a production date within three days, and a physical sample before you commit to a run. Samples are charged at cost and credited against the first order.
Related reading: Moissanite vs White Sapphire: Hardness and Fire Compared, Lab-Grown Diamond Certification Explained, GIA vs IGI vs NGTC: Lab Comparison
Wuzhou HengliLai Jewelry Co., Ltd. - Professional gemstone and jewelry manufacturer in Wuzhou, China. OEM/ODM services, factory-direct pricing, low MOQ.
Contact on WhatsApp service@holycome.comWritten by the Holycome manufacturing & quality team. Every figure and claim in this article comes from our own production floor in Wuzhou, China — in-house casting, stone setting and a three-pass quality inspection we have run since 2008.
Materials are tested for nickel, lead and cadmium release by SGS and CTT International; moissanite centre stones ship with their GRA warranty card and report. We also hold an independent SMETA / Sedex ethical-trade audit and a Hong Kong Customs AML registration.
Reviewed for accuracy by our quality team before publication. Last updated .