Sold as New, Shipped as Used: 6 Checks That Catch Refurbished Stock

๐Ÿง A 4,900-unit rechargeable headlamp order looked flawless on the line. The units were new. The story behind two of the pallets was not: the serials traced back to a production run from five months earlier, the polybags had been resealed, and 6% of the samples showed more than 40 charge cycles in the service menu.

CloudSpects inspector checking a unit serial number against a packing list at a factory bench

๐Ÿง A 4,900-unit rechargeable headlamp order looked flawless on the line. The units were new. The story behind two of the pallets was not: the serials traced back to a production run from five months earlier, the polybags had been resealed, and 6% of the samples showed more than 40 charge cycles in the service menu. That is what B-grade stock looks like at 8 a.m., before a single carton has been opened โ€” and it is exactly the moment a third-party visit is cheap. Importers who book Amazon FBA inspection China after the returns start pay twenty times more for the same evidence.

Refurbished, returned and second-quality units do not arrive labelled. They arrive inside a normal-looking batch, from a supplier who bought them from someone else, on a delivery date that is already tight. โš ๏ธ If your PO only says quantity and colour, nothing in the chain is stopping a returned unit from being sold to you twice.

๐Ÿ“ฆ Why B-grade stock ends up in your container

Every party in the chain has a reason to move a returned unit instead of scrapping it. A local distributor returns 600 units after a promotion ends. A factory has 300 pieces from a cancelled order sitting in the corner. An export agent offers "same specification, better price" on a batch that came back from another buyer's warehouse. None of these units are necessarily broken โ€” many are perfectly functional. They are simply not what your listing, your photographs and your warranty promise describe.

The buyer usually discovers this in November, when the returns arrive faster than the restock. By then the batch is spread across four FBA warehouses and there is no way to isolate the source. The fix is not a stricter supplier promise. It is a check that reads the physical history of the goods while they are still on the factory floor.

๐Ÿ”Ž The 6 checks our inspectors run in this order

#CheckWhat it catches
1Serial and date-code map across every palletAn older production run quietly mixed into a new batch
2Polybag, tape and carton sealing evidenceRepacked goods with fresh tape over old print
3Charge cycles, power-on count and pairing memoryUnits that have already been sold, paired or used
4Cosmetic wear under raking lightScuffed lens rings, worn pins, polished seams from cleaning
5Accessory kit open-count plus total kit weightSpare parts that were never in the box that came back
6Retail print and firmware version against the cartonA carton that describes a build the unit in your hand does not have

Our inspectors run checks 1 and 2 before anything is opened, because the sealing evidence disappears the moment the carton is cut. Then the sample is drawn from sealed stock across pallets rather than from the line-side trolley the factory offers.

๐Ÿงช What the checks look like in practice

On the headlamp order, the serial map took forty minutes and settled the question. One pallet held a continuous range from the current run; two pallets carried a range that ended four months earlier, in the middle of the batch. The carton print dates had been re-stamped. Six percent of the sampled units answered a long-press on the power button with a charge-cycle count between 40 and 90, and the contact pins on 2% showed the polished ring that comes from repeated docking. Four percent of the kits were missing the spare gasket. Individually, each finding had an innocent explanation. Together they described a shipment that had been sold once, returned, and repacked.

The cosmetic check is the one buyers trust least and should trust most. LED lenses, display windows and polished bezels hold micro-scratches that a factory wipe cannot remove, and raking light finds them in seconds. A unit that has sat in someone's kitchen for six weeks looks new in daylight and used under a torch.

๐Ÿ’ธ What a contaminated batch actually costs

Eleven percent of 4,900 units is 539 pieces. At a landed cost near $8.20 that is roughly $4,420 of refunds before removal fees, before the replacement air freight, and before the listing rating absorbs forty one-star reviews that mention the same two words: "used product". The inspection that caught it was two man-days at $169 per man-day, and the factory replaced 520 units while the container was still in the yard. Re-verification closed at 1.2% majors.

โš ๏ธ The arithmetic changes completely once the goods are in FBA. A returned unit in the United States cannot be inspected, repaired or resold at a profit; it can only be refunded, removed or written off. The same unit on the Chinese factory floor is a phone call and a replacement.

โœ… How to write it into your PO and QC brief

Three lines do most of the work. First, "all units from a single production run, serial range declared before packing, no mixed pallets". Second, "carton print date must match unit manufacturing date". Third, "balance payment against third-party inspection report". Add the serial and date-code scan as a specific line item in the inspection brief so it is not left to whoever is on site.

Sampling still does its job. For an order of this size we run AQL 2.5 Level II, which on 4,900 units means 200 pieces with Ac 10 โ€” a filter for the major defects that a serial audit will never see. Run both and the AQL calculator in advance so the sample size is agreed in the PO instead of negotiated on the day. Reports come back typically within 24 hours, with the serial map attached as a spreadsheet.

โ“ FAQs

Can an inspector really tell a refurbished unit from a new one?

Yes, with evidence that lives outside the cosmetic layer: serial number ranges mapped to production dates, sealed-polybag condition, charge-cycle counters in the service menu, contact pin wear and carton print that does not match the unit. Any single signal can be explained away; three that agree cannot.

Will AQL sampling catch a few hundred used units inside a big batch?

Not reliably on its own. AQL 2.5 Level II on 4,900 units means 200 sampled pieces with Ac 10, so an 11% contamination can pass if the used units sit in two pallets. That is why we run a targeted serial and date-code scan on every pallet alongside the sampling.

The supplier says the units were only tested before packing. Is that a problem?

Testing is normal; the questions are how they were tested, what happened to the packaging afterwards, and whether the tested units are the ones in your carton. Re-sealed polybags, a missing spare part and a boot log with dozens of power cycles point to units that were sold once already.

How do I write a serial audit into a purchase order?

Add one line: all units must come from a single production run whose serial range is declared before packing, with no mixed pallets, and the carton print date must match the unit molding date. Then make the balance payment conditional on the inspection report rather than on the sailing date.

Can you check stock that has already been packed and stacked?

That is the normal case at a pre-shipment inspection: cartons are already sealed and palletised when we arrive. We sample from sealed stock, photograph carton marks before opening, and rebuild the pallet so the count is untouched when we leave the floor.

โฐ When to book

Book the visit when production is finished and the goods are packed, roughly five to seven days before the container is due. If the delivery date is already tight and the supplier is talking about a second source, book a during-production check instead โ€” the earlier the same evidence arrives, the cheaper the decision. Across 2,000+ inspections a year, the orders that hurt are never the ones inspected twice; they are the ones inspected after the ship date.

If you want the serial and date-code audit built into your next order, book a pre-shipment inspection or contact our team with the PO and the packing list, and we will confirm the scope and the man-days before you pay a deposit. Full pricing is on the pricing page.

Frequently asked questions

Can an inspector really tell a refurbished unit from a new one?

Yes, with evidence that lives outside the cosmetic layer: serial number ranges mapped to production dates, sealed-polybag condition, charge-cycle counters in the service menu, contact pin wear and carton print that does not match the unit. Any single signal can be explained away; three that agree cannot.

Will AQL sampling catch a few hundred used units inside a big batch?

Not reliably on its own. AQL 2.5 Level II on 4,900 units means 200 sampled pieces with Ac 10, so an 11% contamination can pass if the used units sit in two pallets. That is why we run a targeted serial and date-code scan on every pallet alongside the sampling.

The supplier says the units were only tested before packing. Is that a problem?

Testing is normal; the questions are how they were tested, what happened to the packaging afterwards, and whether the tested units are the ones in your carton. Re-sealed polybags, a missing spare part and a boot log with dozens of power cycles point to units that were sold once already.

How do I write a serial audit into a purchase order?

Add one line: all units must come from a single production run whose serial range is declared before packing, with no mixed pallets, and the carton print date must match the unit molding date. Then make the balance payment conditional on the inspection report rather than on the sailing date.

Can you check stock that has already been packed and stacked?

That is the normal case at a pre-shipment inspection: cartons are already sealed and palletised when we arrive. We sample from sealed stock, photograph carton marks before opening, and rebuild the pallet so the count is untouched when we leave the floor.