FBA Storage Options: AWD, GWD and a China Prep Warehouse

Why Holding Location Became a Decision Again

For most of FBA's history there was one place to keep selling stock: the fulfilment centre. Excess inventory sat with your supplier, in a rented container yard, or in a third-party warehouse you paid for by the pallet and visited by email. That arrangement stopped being the only one when Amazon began publishing rate sheets for its own bulk storage programme, and it stopped being stable when the placement fee made destination routing a priced choice rather than a background default.

By 2026 a seller shipping from China is really choosing between three homes for the same inventory: Amazon's bulk warehouse programme inside the United States, the version of that programme Amazon announced from Shenzhen, or a Chinese preparation warehouse that converts stock into compliant FBA cartons on demand. They are not substitutes. They solve different problems, and the wrong choice is expensive in a way that does not show up on any invoice.

Amazon Warehousing and Distribution, Priced as Published

Amazon Warehousing and Distribution (AWD) is bulk storage that feeds fulfilment centres automatically. Its published 2026 rate shape is worth reading carefully, because the numbers explain what the service is for.

Read as a set, the published figures describe a service for sellers who want Amazon to run the replenishment leg and are willing to pay per cubic foot and per box for the privilege. The volume-based storage rate is the part that punishes a wrong forecast: slow-moving SKUs in bulk storage are not waiting cheaply.

Three Constraints That Decide Whether the Programme Fits

Stock already inside FBA cannot be pushed backwards into bulk storage. The flow runs one way. If you have misjudged a replenishment and units are sitting at a fulfilment centre, the bulk programme is not a place you can relocate them to for the season.

The original programme's warehouses are in the United States, not in China. For a seller whose inventory is manufactured within a hundred kilometres of Shenzhen, that means the goods must complete an international leg before the programme's clock even starts. Nothing about the rate sheet changes the geography.

Discount eligibility is a behaviour test, not a volume test. Amazon's discounted storage terms under the Smart Storage framing are tied to automation thresholds, including running automatic replenishment at a coverage level of at least seventy percent and holding roughly seventy days of stock. That is a coherent design — the platform wants predictable, machine-managed inflow — but it means the cheapest published rate belongs to the seller who has already surrendered some forecasting discretion.

Global Warehousing and Distribution: the Shenzhen Variant

On 9 April 2026 Amazon announced a Global Warehousing and Distribution operation in Shenzhen, which is a direct answer to the geography constraint above. Amazon's published positioning is that storage at the Shenzhen location costs up to forty-five percent less than bulk storage in the United States, and that pairing it with the platform's own ocean freight programme can shorten time to shelf by as much as seven days. Those are Amazon's claims about Amazon's service, stated as claims, and any seller comparing options should test both numbers against their own lanes before relying on either.

The structural point survives even if the percentages move: a bulk storage location at the place of manufacture changes what the words "in transit" mean in your working capital. Instead of a container at sea being both your freight and your warehouse, the inventory can be received, racked and counted before the international leg is booked, and released to whichever destination decision you make that week.

The Third Option Was Always Available

Before there was an Amazon bulk programme in Shenzhen, there was a Chinese preparation warehouse — and it solves a different problem than either. Its job is not to be cheap storage. Its job is to convert mixed supplier cartons into compliant FBA shipments on a schedule that matches your sales plan.

• Goods can arrive from several suppliers, in whatever packaging the factory used, and be reconciled against your purchase order rather than against a booking reference.
• Labelling, bagging, bundling and carton marking happen where labour and correction are cheapest, before anything crosses a border.
• The moment of commitment moves later. Stock that has been received and inspected but not yet labelled is easier to redirect than stock already inside a platform's routing logic.
• Storage is priced by the warehouse, and a conversation about a slow month is possible in a way it is not with a published rate card.

That flexibility has a ceiling too: a China-side warehouse cannot make an Amazon placement decision for you, and it cannot shorten a sailing. The honest comparison is about control and timing, not about which row has the smallest number.

A Worksheet Worth Filling In Before You Commit Pallets

Four Questions to Settle First

Ask what your slowest moving twenty percent of SKUs cost per month in each column, not per unit. Ask how much of your volume is genuinely pallet-ready, since the discounted transport leg in the published rate sheet is conditioned on exactly that. Ask when your destination decision has to be final, because that is the sentence that separates automated replenishment from manual routing. And ask what the discount thresholds require of your ordering rhythm — a seventy-day cover target is a purchasing policy, not a warehouse preference.

On the China side we run receiving, inspection and preparation across three networked warehouses totalling 24,000 square metres in Shenzhen, Suzhou and Dongguan, with no minimum order quantity and preparation completed inside one to two working days. We are not a participant in either platform programme and do not move stock between them; what we can do is have your cartons compliant, counted and ready on the day you decide which of the three homes they are going to.