If you ask three Chinese factories for a quote on custom porcelain dinnerware, you get three different MOQs and three different explanations, and none of them match what the retailer asked for. The gap is not price. It is that “custom” means at least three different things, each with its own tooling, its own minimum and its own failure mode. This guide covers how MOQs are actually calculated, where artisan production stops being economic, and what a usable specification looks like before you pay a deposit.
What “Custom” Actually Covers
Custom porcelain splits into three layers. Shape is the first: a new profile needs a mould, a plaster prototype and a firing trial, because porcelain shrinks 12-15 percent between the wet and fired states, so the mould is cut oversize to compensate. Glaze is the second: a colour match needs a dip test and a firing at 1,280-1,300 degrees, and dark colours behave differently over white and grey bodies. Decoration is the third: an under-glaze decal fires with the glaze, while an on-glaze decal sits on top and needs a second, cooler firing. Most buyers who ask for “custom” mean decoration only, which is the cheapest layer and the one with the lowest minimum. If a supplier quotes one MOQ for all three, ask which layer they priced; the answer usually explains a number that looked strange.
How MOQ Is Really Calculated
MOQ is not one number, it is the maximum of three. Mould MOQ covers tooling and firing trials, and lands at 500 pieces per shape for a new profile, or 300 if a stock mould already exists. Glaze MOQ is set by the dip tank: a colour change means emptying, cleaning and refilling, so factories batch colours and ask for 500 pieces per colour. Decoration MOQ is driven by the decal printer, who charges per sheet and will not run a short batch, which is why four-colour artwork can push the minimum to 1,000 pieces. That is why a 300-piece order with four colours and a new shape is impossible, while the same 300 pieces in white, in an existing shape, ships without argument. Negotiate the layer, not the number.
Where Artisan Production Stops Being Economic
Hand-thrown and jiggered porcelain is beautiful and inconsistent by design. Wall thickness varies by 1-2 mm, capacity by 5-8 percent between pieces, and the glaze breaks differently on every item. That is acceptable for a restaurant that wants character and orders 200 covers, and unacceptable for a retailer that needs 2,000 identical units on a shelf. The economic line sits at roughly 1,000 pieces per shape. Below it, semi-hand production (jiggered body, sprayed glaze) keeps character at a workable cost; above it, pressing or casting with a controlled glaze is the only way to hold tolerances. If a supplier promises hand-thrown quality at 5,000 pieces, ask to see the last three shipments; the pieces will not match each other.
The Spec Sheet That Prevents Rework
A factory cannot guess. Send one sheet with: capacity in millilitres measured to the rim; rim diameter with a tolerance of plus or minus 3 mm; wall thickness target (4-6 mm for dinnerware, 4-5 mm for mugs); the exact glaze reference with a physical sample if colour matters; decoration placement and size in millimetres; packing type; and the test standard the goods must pass. Without a capacity figure measured the same way on both sides, a 280 ml bowl becomes a 250 ml bowl and the run is wrong. Without a glaze sample, “white” becomes “warm white” and the retailer rejects it against their own shelf. The sheet is boring, and it is the cheapest document in the project.
Sampling and the Real Timeline
A realistic schedule: mould drawing and approval 5-7 days; plaster mould and trial firing 10-15 days; samples shipped 5-7 days; buyer review 3-5 days; production 25-35 days for 5,000 pieces; sea freight to Europe or the US 30-38 days. That is 12-16 weeks from approval to arrival, assuming no rework. The step buyers underestimate is the trial firing: the first pieces out of a new mould shrink differently from the tenth, so a factory that skips the second trial sends samples that match the drawing and production that does not. Budget one rework cycle, or order from an existing mould if the launch date cannot absorb it.
What Actually Goes Wrong
The failures repeat. Colour drift between batches, because the glaze was mixed by eye instead of by weight. Capacity shortfall, because the mould wore after 8,000 shots. Decal misalignment on curved surfaces, because the artwork was flattened in the file. Chipping at the rim, because the packing used single-wall cartons for a 2,300 km rail leg. And the quiet one: a glaze that meets food-contact limits on the sample but not on production, because the colourant supplier changed. Each is detectable before shipment with a 30-piece pre-shipment inspection that measures capacity, checks glaze against a retained sample, and drops a carton from 80 cm. That inspection costs less than one rejected container.
Choosing Between Two Quotes
Two quotes for the same drawing, 20 percent apart, usually differ in three places: the glaze (imported colourant versus local), the firing (one pass versus a re-fire for colour depth), and the packing (double-wall versus single). None of those appear on a one-line quotation. Ask each supplier to break the price into body, glaze, decoration, packing and freight, and to name the colourant brand. The cheaper quote is often the honest one, but only when the breakdown proves it. If a supplier cannot break down the price, they are either buying from a trader or guessing.
Write the Specification, Then Compare Quotes
Custom porcelain is not hard because factories are difficult; it is hard because the request is usually vague. Define the layer you are customising, accept the MOQ that layer implies, and put capacity, tolerance, glaze reference and packing on one page. The quotes you get back will finally be comparable, and the first shipment will look like the sample. If you are starting from a stock shape, see our porcelain dinnerware range for capacities and glaze options.