The email is three sentences long. "Hi — we're showing a discrepancy on your last inbound. We can't locate 340 units. Looking into it."
What happens over the next two weeks decides whether you recover most of that money or none of it. Almost all of it is determined by things you either did or did not do before the email arrived.
This is the part of the prep-center relationship nobody plans for. You can vet an operator properly and understand exactly what they charge and still end up here, because inventory goes missing at competent facilities too. The difference between a good operator and a bad one is not whether it happens. It is whether they tell you, and what the paperwork says when they do.
One caveat before anything else: this is a practical playbook, not legal advice. Your contract governs, and a lawyer who reads it will beat any article on the internet. What follows is what to do, in what order, and which questions actually decide the outcome.
First, find out where the goods disappeared
Sellers skip this step and start negotiating with the prep center immediately. That is a mistake, because there are at least six places inventory goes missing and only two of them are the prep center's problem.
The supplier short-shipped. The carton count matched, the piece count did not. Your claim is against the supplier, and your evidence is the receiving count.
It went missing in transit. Freight arrived short or damaged. That is a cargo claim against the carrier, governed by completely different rules with much shorter deadlines — see the cargo claims playbook.
It was never received correctly. Goods arrived, the count was wrong at intake, and the error propagated.
It was lost inside the facility. Received correctly, then misplaced, mis-picked, or shipped to someone else's account.
It was damaged in handling. Physically present, commercially worthless.
Amazon received short. It left the prep center and did not arrive at the fulfillment center — which is either a carrier claim or an Amazon reimbursement, not a 3PL claim.
The single most useful document here is the receiving report with its timestamp and count. If the operator recorded 1,200 units at intake and can only account for 860 now, the loss is theirs and the conversation is short. If intake never produced a piece count, you are about to discover that you cannot prove the units were ever there.
That is the reason the vetting checklist makes such a point of asking whether receiving includes a count. It is not a service question. It is an evidence question.
The first 48 hours: build the file before you write the email
Do this before you start negotiating, and do it in writing.
Ask for the inventory ledger, not a summary. You want every transaction against the SKU: received, adjusted, picked, shipped, damaged, disposed. A number in an email is not evidence. A ledger shows where the count broke.
Ask for the receiving documents. Signed delivery receipt or proof of delivery, the intake count, and any photos taken at receipt. Good operators photograph inbound pallets by default.
Ask what changed. Cycle count results, any adjustments made and who made them, and whether the SKU was moved between locations in the warehouse.
Put a preservation request in writing. Ask them to retain CCTV covering the receiving and storage dates. Warehouse video is usually overwritten on a 14 to 30 day loop. If you ask on day 40, there is nothing to retain. This one sentence, sent early, is worth more than anything else in this section.
Pin the value. Your commercial invoice from the supplier, the landed cost per unit, and what those units were selling for. You will need all three, because what you get paid on is almost never retail.
Keep it factual and unemotional. Every message you send is a document that may be read later by an adjuster.
What a prep center actually owes you
Here is where expectations and reality separate.
They are a bailee, not an insurer
A warehouse holding your goods is a bailee. In the US this generally means they owe reasonable care, and they are liable when the loss results from their failure to exercise it. They are not automatically on the hook for every loss the way an insurer would be.
Practically, this means the argument is about negligence. "The goods were in your building and now they are not" is a strong opening position, because the operator has custody and therefore the burden of explaining. But it is a position, not an automatic win.
The liability cap is the whole game
This is the sentence in your contract that decides the number, and most sellers have never read it.
Warehouse contracts almost always cap liability, and the caps are brutal. Common forms:
Per pound — for example $0.50 per pound. A pallet of small electronics weighing 400 pounds caps at $200, no matter that the goods cost you $38,000.
Per unit — a flat few dollars per unit lost.
A multiple of fees — liability capped at one, two, or three months of what you paid them. If you spend $900 a month with an operator, that is your ceiling.
Time-barred — claims must be filed within 30, 60, or 90 days of the event, and are void afterward.
Read that clause today, before you need it. If you spend $900 a month and hold $60,000 of inventory in the building, the gap between those two numbers is your actual exposure, and no amount of arguing after a loss will close it.
Many contracts also allow you to declare a higher value for an additional charge. Almost nobody does it, because almost nobody reads the clause.
Four different pockets
When goods are lost, the money can come from four places, and they are not interchangeable.
The operator's own funds. Small losses, especially where they want to keep your business, get credited without a formal claim. This is the fastest path and the reason a calm, well-documented first email matters.
Their warehouse legal liability policy. This covers goods in their care, custody and control, and it responds only to the extent they are legally liable — which loops straight back to the cap. General liability does not cover your inventory; it covers someone slipping in their parking lot.
Your own cargo or stock-throughput policy. This is the pocket most sellers do not have and should. It insures your goods on your terms, at your declared value, regardless of the operator's cap or fault. For inventory of any size it is the single highest-leverage thing on this list, and the premium is small next to the gap it closes.
Amazon. Only for losses that happen after Amazon takes custody. Not applicable here, but see below for the sequencing trap.
Filing the claim
Send a formal written claim even if conversations are friendly. It starts the clock, creates a record, and is usually a contractual precondition for anything else.
Include:
Identification of the shipment and SKUs, with dates
The quantity lost or damaged, and how you arrived at it
The value claimed, with your basis for it — attach the commercial invoice
Copies of the receiving documents and ledger you already requested
Photos, if damage rather than disappearance
A specific demand and a deadline for response
On value: expect to be paid your cost, not your selling price. Recovering lost profit is a much harder argument and most contracts exclude consequential damages explicitly. Claim your landed cost, including the freight and duty you already spent getting the goods there, because that is money genuinely out the door. Do not inflate. One unsupportable number makes every other number in your claim negotiable.
The double-recovery trap
If part of the loss involves units that left the prep center and never arrived at Amazon, you may have both an Amazon reimbursement case and a claim against the operator or carrier. Sequence matters, and so does honesty.
Work out who had custody at the moment of loss and claim once, against that party. Filing with Amazon and the operator for the same units, and collecting twice, is the kind of thing that gets an Amazon account reviewed. It is also the kind of thing operators discover, because adjusters talk.
Where the split is genuinely unclear, say so in the claim rather than filing two full claims and letting them sort it out.
The lien: why "we're pulling our inventory out" can fail
This is the detail that surprises people most, so plan for it before you make threats.
Warehouse operators generally have a lien on goods in their possession for unpaid charges. If you stop paying invoices because you are in a dispute, the operator may be entitled to hold your remaining inventory until the balance is settled. You can end up with the original loss plus everything else you own sitting in a building you can no longer access.
The practical rule: keep paying undisputed invoices while you fight about the disputed amount. Pay under written protest if you like, but pay. Withholding payment feels like leverage and usually hands leverage to the other side.
If you are leaving, retrieve the goods first and settle after. Arrange the outbound, get it on a truck, then negotiate.
Catching it earlier next time
Most losses are discovered late, and late discovery is what destroys claims. Deadlines run, video is overwritten, staff turn over.
Reconcile monthly. Units received minus units shipped minus units on hand should equal zero. When it does not, ask that month, not at year end.
Ask for cycle counts in the contract. Quarterly on your top SKUs, with the results sent to you. An operator who counts regularly finds their own errors, which is exactly what you want.
Watch the shrinkage rate, not the incident. One missing unit is noise. A consistent quarter-percent drift is a process problem and worth raising while it is still small.
Photograph high-value inbounds. Ask for pallet photos on receipt as standing practice.
What to change in the next contract
You will not renegotiate a warehouse's standard terms wholesale. You can usually move three things, and they are the three that matter:
A declared-value option, with the rate to raise the cap written down.
A realistic claim window — 90 days rather than 30, since discovery is the bottleneck.
A retention clause for receiving records and video — 90 days, in writing.
And the one that does not require their agreement at all: buy your own cargo policy covering goods at third-party locations. It makes the cap someone else's problem.
The vetting checklist tells you whether an operator can do the work. The pricing breakdown tells you what it costs. This is the third question, and it is the one that decides how much of a bad month you personally absorb: when this operator loses your goods, what exactly are they on the hook for?
Find out now, while it is a clause in a document and not a number in an email.
