Running a small-batch trial order: the sequence, the schedule and the traps

The first order is where most buyer-supplier relationships either start properly or quietly die. This note walks the whole sequence from specification to arrival, with the points where a trial run usually goes wrong and a realistic week-by-week schedule.

The sequence, and where it breaks#

1. Specification freeze. Write down the handle, the head, the modes, the battery, the charger, the packaging and the destination market. Freeze it in writing and date it. Most trial-order disasters are a specification nobody froze, so each party later quotes a different product.
2. Quote with the fixed/variable split. Ask for one-off costs (tooling, plates, artwork, certification) separately from the per-unit price. A single blended number cannot be interrogated, and it is the signature of a party that cannot see its own cost structure.
3. Paid sample, credited against the order. Expect to pay for samples and freight. A free sample is usually a stock unit, not a build of your specification. Agree in writing that the sample cost is credited to the trial order — most suppliers agree, and the ones who will not are telling you something about how the main order will be handled.
4. Certification scope. Decide the destination market before tooling, because the plug, the charger topology and the manual language are bill-of-materials decisions. Certification is per model per market and behaves like tooling: a fixed cost that amortises over the run.
5. Artwork and die-lines. Get the structural die-line before designing. Designing a retail pack on a guessed die-line is the most common cause of reprinting.
6. Order and deposit. A 30/70 structure is normal. Verify the beneficiary account by a second channel — see the payment note below.
7. Pre-production sample approval. Approve a physical sample built on the production line with production tooling, not a hand-made bench sample. Photograph what you approve and keep the photographs with the purchase order; that photograph is your reference in any later dispute over colour or finish.
8. Production, ageing and testing. Ask for the ageing duration and the acceptance criterion in the order acknowledgement, not afterwards.
9. Pre-shipment inspection. Book it before the balance payment. On a first order this is non-negotiable — it converts a subjective argument about quality into a written standard.
10. Documents and balance. Battery paperwork, packing list, invoice; then the balance, then release of the goods.

A realistic first-order schedule#

Ten weeks is a normal trial order for a customised build. If a supplier quotes four weeks for the same scope, ask which of these steps is being compressed. The answer is usually the ageing window or the sample approval, and both of them protect you.

The traps that cost the most#

• Changing the specification mid-run. A colour change after the injection run starts is a new run. A logo change after the plate is cut is a new plate. Every change has a price and a delay, and neither is negotiable once tooling exists.
• Treating MOQ as a per-order number. It is usually per colour and sometimes per SKU. A 3,000-piece order in four colours can fail a per-colour minimum without anyone lying to you.
• Forgetting the packaging cost line. Packaging can be 8–15 % of landed cost in this category. A quotation that does not mention it is incomplete, not cheap.
• Business email compromise. This is the most expensive routine mistake in the trade. A supplier's mailbox is compromised, a “revised banking details” email arrives mid-order with correct letterhead and correct order numbers, and the deposit goes to the wrong account. Always verify a change of bank details through a second, independent channel — a phone call to a number you already had, not a reply inside the same email thread. No legitimate supplier is offended by this.
• Skipping the inspection to save a few hundred dollars. It is the cheapest insurance in the transaction.

What a good trial order should produce#

Not just goods. It should produce a signed specification, a photographed approval sample, a batch record, a test-report set, and a supplier who has now seen your standard. If the second order does not go faster and more cheaply than the first, the trial order did not do its job.