Shipping From China to Amazon FBA in the US: Methods, Costs and What Goes Wrong

The Three (Really Four) Ways to Do It

Three real methods move goods from a Chinese factory to an Amazon fulfillment center, and a fourth — AWD — sits between them. The choice is a straight trade between speed and per-unit cost, and it needs to be made before the purchase order goes in, not after the container is already booked.

Sea freight is the default for anything with steady sell-through: full containers (FCL) if you can fill one, consolidated (LCL) if you can't. Air freight buys back three to four weeks of lead time at several times the per-kilo cost, and it's priced on chargeable weight — the greater of actual weight or dimensional weight — which is where sellers of bulky-but-light products get an unpleasant invoice. Air courier/express (a commercial account with DHL, FedEx or UPS) is for emergencies and samples, not restocks. AWD is Amazon's own inbound buffer: freight lands there in bulk and Amazon releases it to FBA fulfillment centers as listings need it, which smooths some of the receiving risk that comes with dumping a full container straight into FBA.

The Process, Step by Step

Getting a shipment from a factory floor in Yiwu or Shenzhen to an Amazon shelf in the US runs through the same sequence whether the order is 500 units or 50,000:

• Production and inspection — factory completes the order; a third-party inspector checks quantity and quality before it leaves the country.
• Export documentation — commercial invoice, packing list, and for ocean shipments an Importer Security Filing (ISF) with US Customs before the vessel loads.
• Freight booking and export clearance — your forwarder books the container or air waybill and clears the goods out of China.
• Transit — ocean or air, as above.
• US customs clearance — a licensed customs broker files entry, duty and any applicable tariffs get paid based on the product's HTS code.
• FBA prep — poly bagging, labeling, carton marking, done at the factory, at a US prep center, or through a 3PL before Amazon ever sees it.
• Inbound shipment plan and delivery — the shipment gets created in Seller Central, Amazon's carton labels are printed, a delivery appointment is booked, and the truck goes to the fulfillment center or AWD facility Amazon assigns.

Every one of those steps can add a day of slippage, and they compound. A forwarder's quoted transit time almost never includes customs hold, FBA prep, or Amazon's own receiving queue — that gap is the single biggest difference between the date a seller expects and the date units actually go live and sellable.

A Worked Example: Sea vs Air on the Same Order

Say you're importing 2,000 units of a kitchen tool, factory cost $4.50 a unit, packed 50 to a carton, 40 cartons total, roughly 12 CBM. These numbers are illustrative — your lane, season and product will move them — but the math itself doesn't change:

• Sea freight (LCL): a lane running roughly $150–$220 per CBM door to port puts this shipment near $2,200 in ocean freight, plus destination handling and drayage — landed freight cost near $1.30–$1.60 a unit. Transit 30–40 days.
• Air freight: priced on chargeable weight, often $4–$6 per kg once dimensional weight kicks in on a bulky item — landed freight cost can run $4–$7 a unit, three to four times the sea rate. Transit 8–12 days.

The order lands three to four weeks earlier by air, at a freight cost that can add several thousand dollars over sea freight for the same units. That's the real decision every seller is making, explicitly or not: is that much runway before a stockout worth the extra cost on this specific order? Usually the honest answer is a blend — sea for the base reorder, air only for the units needed to bridge the gap.

Where This Goes Wrong

Full Circle has managed more than $500M in revenue across 100+ brands, and the shipping mistake that costs the most isn't a bad freight rate — it's the gap between when a shipment leaves the factory and when it's actually sellable on Amazon. The five patterns we see most often:

• Dimensional weight surprises. A product that's light but bulky gets classified oversize by Amazon's fee schedule, not by its actual weight — sellers price against the standard-size fee and find out only after the first shipment lands.
• Carton labeling errors. Wrong barcode, or a carton count on the label that doesn't match what's inside — Amazon rejects or short-receives the shipment, and units sit in limbo while the listing goes out of stock.
• Treating transit time as total time. The 25-day sea quote doesn't include customs, prep and Amazon's receiving queue on the other end. Reorder points built off transit time alone run out early.
• Chinese New Year and peak-season blind spots. Factories close for weeks around late January/February, and ocean rates spike ahead of Q4. An order placed without padding for either one arrives late for the sales window it was meant for.
• Not checking what actually arrived. Lost or damaged cartons, short shipments, and units received into the wrong fulfillment center happen more often than sellers expect, and Amazon's reimbursement window isn't indefinite.

We've watched a version of the second and third mistakes happen on a client's shipment — a carton count miscoded in transit. The fix was catching the discrepancy against the shipment plan before the reorder point hit zero, not after.

When the Shipment Is Already Late

If a container is delayed and the reorder date has passed, there are three real options, roughly in order of cost: split a partial quantity onto air freight to bridge the gap, pull forward any safety stock sitting in AWD, or accept the stockout and manage the Inventory Performance Index and rank hit deliberately rather than losing both the sales and the ranking recovery afterward.

If the fee that comes back after a shipment lands is higher than expected, don't assume it's correct — check the dimensional weight Amazon used against the carton's actual dimensions, because misclassification happens and it's disputable. And if sales dropped the same week a shipment landed, check the ad account before blaming inventory: a Buy Box loss from an out-of-stock ASIN and a bid or budget change look identical in the topline numbers, and the fix for the second one lives with Dr. PPC, not in the warehouse.