Summary: Buying samples from several China factories only helps if you compare them on the same spec sheet: basis weight, SAP load, rewet, strike-through and dry feel. Pay the sample fee, request a golden sample, and score each on a fixed matrix — that predicts bulk quality far better than a pretty brochure does.
Sourcing your first diapers from China usually starts with a sample box. The mistake I see is people judging samples like a parent at a baby fair — "this one feels soft" — instead of like a buyer with a scorecard. Softness is real, but it's the last thing you should decide on.
Order from at least three, not one
One sample tells you nothing; there's no baseline. Three to five gives you a spread. When you request samples, ask each factory for the same size and the same spec target so the comparison is fair. Our 2026 sourcing guide lists the first questions to send.
What the sample fee actually covers
A sample fee (often refunded against your first order) pays for materials, the line stop to make a small run, and courier. Don't hunt for "free" samples from a factory that won't otherwise talk to you — the serious ones charge because making 200 correct diapers costs them real money. Freight via DHL or FedEx is separate; budget $40–$120 depending on weight and country.
Build a comparison matrix before the box arrives
Decide your columns in advance so you're not swayed by packaging:
| Metric | How to check | Why it matters |
|---|---|---|
| Basis weight (gsm) | Weigh cut core | Material honesty |
| SAP load | Ask + ash test | Real capacity |
| Rewet | Load + blot paper | Skin dryness |
| Strike-through | Drop test | Speed of absorb |
| Dry feel | Touch after load | Comfort |
For the absorbency theory behind these numbers, our absorbency science article is the reference. And the safety analysis covers what "safe" means in practice.
The golden sample is your contract
Once a factory's sample wins, seal one as the golden sample. Every bulk shipment gets compared against it. No golden sample, no comeback when lot 12 feels thinner. Our factory-floor guide shows how pros lock this in, and the 10 critical checks are the pre-sign safeguards.
FAQ
How many factories should I sample from?
Three to five. Fewer gives no baseline; more becomes noise. Keep the spec identical across all so the matrix is fair.
Is a free sample a red flag?
Not always, but a factory that charges a refundable sample fee usually takes the spec more seriously. Judge by the test report, not the price of the box.
What if two samples score equally?
Then decide on consistency and communication: ask for the last three batch reports. The steadier factory wins the long run.
