Operations: fulfilment, packaging & compliance
FBA is a product you buy. A third-party warehouse is a relationship you manage. That is the change nobody prepares an Amazon seller for, and it matters more than any line on a rate card.
With FBA there is no account manager, no negotiation and no service conversation. Units go in, the machine works, fees are published. With a 3PL you are a customer of a business run by people, with a rate card that is negotiable, a service level that is whatever you agreed, and an error rate that depends on how well you briefed them. Brands that treat a 3PL like a vending machine get poor service and blame the warehouse.
Here is what actually changes, and what to look for when you pick one.
What you gain and what you give up
The gain is capability and control. Cases, pallets, carton labels, routing guides, kitting, retail-ready display units, relabelling, returns you can actually inspect, inventory you can ship anywhere. None of that exists inside FBA, and the reasons are laid out in 3PL vs FBA.
What you give up is uniformity. Amazon's network absorbs demand spikes, ships from many locations, and never has a bad Tuesday because somebody quit. A single warehouse is one building with a finite team. If your order volume triples in December, somebody there has to have planned for it.
You also give up the Prime badge on anything you fulfil yourself, unless you qualify for the seller-fulfilled programme and can actually meet its delivery obligations from that building.
The realistic end state for most brands selling both channels is both: the warehouse as the stock-holding hub, FBA as a replenished spoke. Getting the flow between them right is a separate discipline, covered in inventory split between Amazon and wholesale.
How the money is structured
The FBA fee schedule is public and roughly a single number per unit for fulfilment, plus storage by volume. A 3PL rate card is a list, and comparing it to FBA by one line is how brands get surprised.
| Line on the rate card | What to ask |
|---|---|
| Receiving | Per pallet, per carton or per hour? Who unloads? |
| Storage | Per pallet position, per bin, per cubic foot? Billed how often? |
| Order pick | Charge per order, per line, per unit, or all three? |
| Packing materials | Included or billed at cost plus? |
| Pallet build | Per pallet, and is shrink wrap extra? |
| Special projects | Hourly rate for relabelling, kitting, rework |
| Account minimum | Monthly floor regardless of activity |
| Shipping | On your carrier account, theirs, or marked up? |
The two lines that quietly decide your bill are the account minimum and the special-projects rate. A small brand can spend more on the minimum than on actual work for months. And every prep task FBA would have done invisibly — a poly bag, a label, an overbox — becomes a billable project line here.
Ask for a sample invoice from a comparable customer with the names removed. A rate card tells you the prices; an invoice tells you which lines actually get used.
What to check before signing
Six questions that separate a warehouse that fits from one that does not.
Have they handled retail accounts. Ask which retailers, whether they produce advance ship notices, and whether they hold electronic connections already. A warehouse that has never built a compliant pallet will learn on your chargebacks. The requirement set is described in EDI for small brands.
Can they prep for Amazon. Poly bagging, labelling, case-packed and individual shipment builds, and booking the inbound. If they cannot, your hub-and-spoke model needs a second party.
What is the order cut-off. The time of day after which an order ships tomorrow. It sounds trivial until a buyer's ship window depends on it.
What accuracy do they commit to. Ask how they measure pick accuracy and what happens when they are wrong. A warehouse that cannot answer does not measure.
How do they handle receiving discrepancies. Your container lands, the count is short by eleven. Who investigates, how quickly, and what do you see.
What does the system show you. Real-time stock, order status, lot and expiry tracking if your category needs it, and an interface your own systems can read.
One more question that is rarely asked and always useful: who will actually answer the phone. Ask for the name of the person who handles your account day to day, and how escalation works when they are away. A warehouse with one overloaded coordinator behaves very differently in December than the tour suggested in June.
The contract terms that matter
Read four clauses properly, because they are the ones that hurt later.
The notice period and exit terms. Moving out of a warehouse is a physical project, and a long notice period with a hostile clause means you can be trapped somewhere that is failing you. Ask explicitly what it costs to have your inventory loaded onto trucks and leave.
The liability limit on lost or damaged stock. Standard warehouse terms usually cap liability at a small amount per pound of goods, which is far below the value of most consumer products. Know the number and insure the gap.
The billing dispute window. If you have ninety days to query an invoice line, someone has to be reading invoices monthly.
And rate review. How often can rates change and with how much notice.
Does a 3PL make Amazon worse
Not if you plan the replenishment properly, and slightly if you do not.
Inbound from a warehouse to a fulfilment centre takes time: pick and pack the shipment, transit, and then Amazon's own receiving, which can add days. That total is your replenishment lead time, and your FBA cover has to be at least that plus a buffer.
Brands that run cover too thin discover the second cost, which is the fee applied when your inventory cover sits below the threshold Amazon considers healthy. Shipping small amounts frequently keeps working capital available and keeps you close to that line, so pick a cadence that is frequent but not hand-to-mouth.
The other thing to avoid is splitting a SKU's identity. Same barcode, same case quantity, same item master in both places. Two versions of one product is how a warehouse ships retail-labelled units into an Amazon shipment and gets them refused.
When staying pure FBA is the right call
Three situations where adding a warehouse is premature.
Every order you ship is a single consumer unit and no buyer has asked for a case pack. The 3PL adds a monthly minimum against no new revenue.
Your product is small, light and fast-moving, so FBA's per-unit economics are excellent and storage is trivial. A slow, bulky SKU is the one that benefits from a pallet position; a fast small one is not.
Or your volume is low enough that you cannot meet a warehouse's minimum without the minimum itself becoming your largest fulfilment cost. In that case, keep fulfilling small wholesale orders from marketplace stock while the channel proves itself — the limits of that approach are in multi-channel fulfilment for wholesale.
Deciding with actual information
The warehouse decision is downstream of a demand question: how much wholesale volume is realistic for your product, and what shape does it arrive in. A category served by distributors buying pallets needs a very different warehouse from one served by boutiques buying two cases.
Paste your listing into WholesalePilot and the preview shows which retailers and distributors stock comparable products, which tells you whether to shop for a warehouse that builds pallets and files ship notices or one that simply picks parcels well.
Then choose on fit rather than on the cheapest pick fee. The difference between rate cards is cents per order. The difference between a warehouse that can serve a retail account and one that cannot is the channel itself.