A lead-cadmium report is only as good as the workflow behind it: the wrong sample, the wrong category or the wrong unit can turn an expensive test into useless paper. This is the six-step workflow experienced buyers run, from frozen specification to archived report, and it works the same for a 500-piece first order as for a rolling programme.
The workflow below is written for importers and brand owners buying ceramic dinnerware, mugs and teapots from China. Its purpose is simple: make sure that when a customer, a platform or a border authority asks "what proves this is compliant?", the answer exists, is current, and refers to the goods actually shipped.
Step 1: Freeze the specification first
Testing without a frozen specification wastes money. Before anything is sampled, decide and write down: the glaze system for each SKU, the decoration technique and its position on each surface (underglaze on food-contact zones, overglaze accents confined to outer walls), and the target markets. The market list decides the panel scope — United States means 21 CFR 109.16 with ASTM C738, the European Union means 84/500/EEC, Germany adds the LFGB limits. A specification that changes mid-programme invalidates everything tested before it, so freeze it and version it.
Step 2: Pull samples the way regulators would
The sample set is where most testing fails before the lab is even involved. Pull from mass production — random cartons off the line, not the pilot set polished for photographs. Cover every decoration colour in the programme, because a red enamel and a blue underglaze are different chemical systems with different risk. Record which cartons and which batch the samples came from, and reference the golden sample so results tie to an identifiable standard. A lab report on unrepresentative pieces proves nothing about the shipment you intend to keep.
Step 3: Define the panel per market
| Market | Standard and limits | Limit basis | Method |
|---|---|---|---|
| United States | FDA 21 CFR 109.16 lead action levels; cadmium per CPG Sec. 545.400 | µg/mL of extract | ASTM C738 |
| European Union | Directive 84/500/EEC lead and cadmium limits | mg/dm² of surface | EC No 636/2004; ISO 6486 series |
| Germany | LFGB / BedGgstV Anlage 14 limits, plus overall migration and sensory scope | mg/L for drinking vessels | LFGB-scope panel |
One submission can report against several frameworks on the same sample set, which is how a three-market programme stays affordable. The panel definition — standards, methods, article categories — goes into the purchase order, not into a verbal agreement.
Step 4: Commission the right laboratory, hold the right document
Use an accredited third-party laboratory; firms such as SGS, Intertek and TUV are commonly engaged for ceramic migration work. Decide who the report is addressed to — a report held only in the factory's name, with no linkage to your order, is awkward evidence — and require the report to identify the articles, the glaze and decoration system, the batch and the golden sample reference. Reports available on request is the phrase to use publicly; the file behind it should be complete.
Step 5: Read the results against category and unit
Check three things before accepting a pass. Category: a mug assessed against plate limits is a fail dressed as a pass — drinkware carries the strictest US level at 0.5 µg/mL, and EU shallow ware carries the 0.07 mg/dm² cadmium limit. Units: µg/mL, mg/L and mg/dm² answer different questions and are not freely comparable. Per-piece results: an average can conceal a single bad piece, and enforcement finds tails, not averages.
Step 6: Retest on change events, archive for the programme's life
| Event | Retest? | Why |
|---|---|---|
| New decoration colour or decal supplier | Yes | Pigment chemistry changes with the colour |
| Glaze, frit or body supplier changes | Yes | The food-contact surface itself changed |
| New kiln, new factory, moved production | Yes | Firing regime affects the glass layer |
| No changes, ongoing programme | Periodic | Cadence agreed with customers keeps the file current |
Archive the declaration of compliance, the reports and the sample seals together, so any question months later can be answered in minutes. Our quality and compliance page describes how we run this workflow with third-party laboratories, reports available on request. If you are scoping a custom programme, see custom and OEM, or start directly at request a quote and we will build the test matrix into the specification.
Frequently asked questions
How many pieces should a test sample set include?+
Follow the laboratory's instruction for the method, but as a specification matter: cover every shape category and every decoration colour in the programme, and pull from mass production rather than pilot pieces. Two shapes and three colours means at least the pieces needed to cover all six combinations.
Who should hold the test report — buyer or factory?+
Both should hold a copy, but the report must be traceable to your order: identify the articles, batch and golden sample, and agree in advance who the commissioning party is. A report the buyer commissioned, addressed appropriately and filed with the shipment, is the strongest evidence.
Is one test enough for repeat orders?+
The results stand for the inputs tested. Reorders are covered while the glaze, decoration system, supplier and firing regime stay unchanged; any change event reopens testing. Periodic retesting on unchanged programmes is the norm for customers who audit their files.
Does a pass in one market carry over to another?+
No. The same clean line can sit far below every limit, but the frameworks differ in categories, units and strictness — German drinking-vessel limits are far tighter than EU container values. Specify the full panel for the markets you intend to sell, in one submission if possible.
