Yiwu vs factory: which is better for low-MOQ testing?

Question

Accepted Answer

Short Answer: Yiwu for Testing, Factory Once a Product Proves Itself

Direct Answer

For testing, Yiwu wins clearly. One carton of 50-200 units per SKU against a factory's usual 1,000-5,000 unit minimum means you can validate many products for the cash a single factory run would absorb. Move a SKU to a factory or 1688 once it sells and you need volume pricing or real customisation.

The Core Trade-Off in One Table

Neither channel is better in the abstract. They answer different questions: Yiwu answers whether a product sells, a factory answers how cheaply you can make it once you know.

Dimension	Yiwu booth	Direct factory
Minimum per SKU	50-200 units (one carton)	1,000-5,000 units
Unit price	Higher, includes a trader layer	Lower at volume
Customisation depth	Print, colour, simple packaging	Tooling, materials, full spec
Speed to first order	Days	Weeks once tooling is involved
Comparing alternatives	Dozens side by side in one aisle	One supplier at a time
Cash at risk on an unproven SKU	Low	High

A buyer comparing several similar products displayed at a market booth — Yiwu lets you compare dozens of near-identical products in one morning.

Why Testing Breadth Beats Testing Depth

The expensive mistake is not buying the wrong quantity. It is buying a large quantity of a product nobody has yet paid for. Yiwu lets you fail cheaply across many SKUs, which is statistically how winners are found.

Yiwu: 20 SKUs tested~$4,000

Factory: 1 SKU tested~$4,100

Yiwu: units per SKU50-200

Factory: units per SKU1,000-5,000

What You Give Up by Testing at Yiwu

An honest comparison has to name the costs. There are three, and all of them are acceptable during a test phase and unacceptable at scale.

A higher unit priceYou are paying a trader layer on roughly 70% of booths.

Shallow customisationLogo and colour are easy; a new mould or material is not.

A ~20% reorder markupReordering through a trader costs more than going to the source.

But near-zero inventory riskWhich is worth all three during validation.

A production line running inside a mid-size consumer goods factory — Factories win on unit price once a SKU has proven demand.

The Signals That a SKU Is Ready to Move

Moving too early wastes tooling money on a product that is still unproven. Moving too late leaves margin on the table every cycle. These are the practical triggers.

1

Two or three clean sell-through cycles

Demand is repeatable, not a one-off spike.

2

You are reordering 1,000+ units at a time

Volume has reached the point where factory minimums stop being a barrier.

3

You need a spec change the booth cannot make

New material, new mould, or a certification requirement.

4

Margin pressure is real

A 10-30% unit saving now moves the P&L rather than rounding it.

Case: The Hybrid That Worked

Buyer: Canadian pet-accessory brand, second year of trading

The client tested 18 SKUs across District 1 and District 2 in a single trip. Three products carried the account. We moved those three to a factory in the surrounding towns for a tooled version with their own packaging, while the long tail stayed on Yiwu stock so the catalogue kept breadth without inventory risk.

OutcomeUnit cost on the three lead SKUs fell 23%, while the remaining fifteen continued to be bought by the carton.

A small carton beside a stack of bulk pallets in a warehouse — Test by the carton, scale by the pallet once demand repeats.

Cost Context for the Decision

50-200Yiwu units per SKUOne carton, stock items

1,000-5,000Factory MOQ per SKUTypical direct minimum

10-30%Unit saving at the factoryOnce volume justifies it

~20%Trader reorder markupThe cost of staying at Yiwu at scale

How RND SOURCING Runs the Yiwu-to-Factory Path

Because we sit in Yiwu and also work the factory towns around it, we can run both halves of this path for the same client. The test order is bought by the carton across booths and consolidated into one shipment. When a SKU proves out, we trace it back to the actual producer, request an EXW quote, verify the business licence on the national GSXT register and arrange a workshop video before any tooling is paid for. We do not own factories, so the recommendation to move a SKU is based on the client's numbers rather than our own capacity.

Map my Yiwu-to-factory path

RND SOURCING — Yiwu, Zhejiang, China. Written by our sourcing desk from first-hand market work.