How to Confirm Samples Before a Bulk Order (Without Missing Anything)

Most quality disputes between lighting buyers and factories are not caused by bad production — they are caused by samples that were never properly confirmed. This is the protocol we put in front of buyers before any bulk order: what to test, what to write down, and how to turn an approved sample into the reference that governs the whole production run.

What does 'sample confirmed' actually mean?

Short answer: a named sample, signed or referenced in writing by both sides, with its acceptance criteria — and the bulk order states that production must match it. Anything less ('the sample looked fine, go ahead') leaves every later disagreement with no referee.

The protocol below takes about two weeks end to end and exists in two rounds: an evaluation round where you learn what to change, and a confirmation round where the golden sample is defined. Skipping the second round is the single most expensive shortcut in lighting procurement.

Round 1 — the evaluation sample

Order the catalogue version first, even if you plan OEM changes. You are testing the platform, not the branding. Check it in the environment it will actually live in — a hotel lamp under a nightstand, a retail lamp on a mock shelf — and grade it against the criteria below.

Round 2 — the golden sample

Once the platform passes, lock the configuration. The golden sample carries everything production will be judged against: finish and shade material, CCT and output at the socket, cable length and plug type, markings and labels, and — if you are branding — the actual colour box, manual and logo application. Photograph it from six angles and file the photos with the reference.

Both sides then reference this one object. On our side the golden sample is what final inspection judges against, and what IQC verifies incoming materials against. On your side it is what your inspector (if you appoint one) carries as the acceptance basis. One physical reference, shared by everyone, ends most arguments before they start.

What to write down (the one-page confirmation sheet)

A confirmation sheet does not need to be a contract; it needs to be unambiguous. One page covers it:

• Model code and catalogue reference (e.g. T92 — touch table lamp, 35 cm, nickel matt)
• Light data: CCT (with tolerance), luminous flux, dimming behaviour if any
• Finish and materials, per part
• Cable length, plug type, voltage for the destination market
• Packaging: box type, artwork version, master-carton quantity
• Markings: logos, labels, ratings text and their positions
• The sentence: 'Bulk production must match golden sample ref. ___ dated ___'

Timing: how long each round should take

Catalogue samples: 5–10 days to dispatch — the constraint is finish availability, and on our range the sample cost is credited against the first production order. OEM-finished samples (your colour box, your finish): 2–3 weeks, dominated by shade and packaging-material sourcing and artwork approval. A full ODM prototype: 4–8 weeks depending on design complexity — that timeline is broken down stage by stage on the OEM page.

What should not take time is the feedback loop on your side. Decide internally who signs off and give them a deadline; the most common sample-stage delay is an approval queue, not the factory. We flag it because it is the one delay that no factory can fix.

One more timing rule that saves whole weeks: parallelise the rounds. Order evaluation samples from your three shortlisted suppliers in the same week, not sequentially, and send the change-request for round two to all of them simultaneously. Buyers who run sample rounds in series routinely spend two months where one would do — and then blame the season for a launch that was late before it started. The calendar is part of the protocol, not an afterthought to it.

From golden sample to first bulk check

When production starts, do not wait for shipment to check again. Ask for a pre-production sample from the first run — the first units off the line, made with production materials and production tooling — and compare it against the golden sample side by side. On repeat orders, the check is faster: the golden sample already exists, and incoming materials are verified against it at IQC.

For large or phased programmes, the same logic scales: one golden sample per finish and packaging version, referenced on every PO, and third-party inspection sampling against it before container loading. That is the entire system — it fits on a page and it prevents almost every dispute we have ever watched happen.

Five sample-stage mistakes that cause later disputes

We watch these repeat across buyers who are otherwise thorough — none is technical, all are avoidable:

• Approving by committee delay. Samples sit in an internal approval queue for weeks, then everything is rushed. Name one decision-maker with a deadline before samples ship.
• Approving the wrong thing. The sample that impresses is often the platform you did not price. Confirm the exact model code, finish and configuration you intend to order — not 'something like this'.
• Testing in the wrong room. A bedside lamp judged on an office desk under ceiling lights reads wrong. Test where it will live, at the hour it will be used.
• Verbal tweaks. 'Make the shade a touch warmer, you know what I mean' — no, the production line does not. Every change is a new sample with a new reference, or it does not exist.
• Skipping the packaging check. The colour box is part of the product: artwork version, carton quantity, barcode and manual language all belong on the confirmation sheet, because all of them go wrong in bulk otherwise.

Quick answers