The FBM Seller's Guide to A-to-Z Claims: The Warning Signs Amazon Won't Send You

An A-to-Z Guarantee claim is one of the most stressful things an FBM seller can face. It affects your Order Defect Rate, it can trigger account health warnings, and Amazon's automated systems can grant the claim and deduct the funds before you've even had a chance to respond.

What most sellers don't know is that the majority of damaging A-to-Z outcomes are preceded by a specific, detectable signal — one that gives you a narrow but real window to intervene before Amazon's bot makes the decision.

This article explains how A-to-Z claims actually work, what the highest-risk precursor signal looks like, and what you can do in the hours between that signal and a potential auto-grant.

How A-to-Z Claims Actually Work

The A-to-Z Guarantee exists to protect buyers who don't receive their item or receive an item significantly not as described. When a buyer files an A-to-Z claim, Amazon reviews the situation and can either deny the claim, grant it and charge the seller, or request more information.

What sellers often don't fully appreciate is how much of this process is automated. Amazon's systems evaluate claims based on order data — shipping confirmation, delivery status, carrier tracking — and make grant/deny decisions algorithmically for a significant portion of cases. Human review is the exception, not the rule.

This means that by the time you receive an A-to-Z claim notification, the automated decision may already be made or already be in motion. Responding to a claim notification is often too late to affect the outcome.

The window that matters is before the claim is filed.

The Highest-Risk Pattern: Cancellation on a Shipped Order

Of all the situations that lead to A-to-Z auto-grants, one pattern is both the most common and the most actionable: a buyer requesting cancellation on an order that has already been shipped.

Here's why this is so dangerous. When a buyer requests cancellation and you've already shipped, the correct response from a purely operational standpoint is to tell the buyer the item is already on its way and to refuse the return at the door or initiate a return when it arrives. That's reasonable. But Amazon's systems don't always see it that way.

If the buyer doesn't get the cancellation they want, the next step is often to file an A-to-Z claim. At that point, Amazon's automated review looks at the situation: buyer wanted to cancel, order was shipped, buyer didn't receive a satisfactory resolution. In a meaningful percentage of these cases, the claim is auto-granted.

"Amazon will auto-grant the claim if you don't act first. You have a narrow window — intercept the package or contact the buyer immediately."

The moment a buyer requests cancellation on a shipped order, you are in a race. Not against the buyer. Against Amazon's automated claims system.

What Amazon's API Tells Us

Amazon's SP-API ORDER_CHANGE notification fires when specific order-level events occur. One of the event types is BuyerRequestedChange — this fires when a buyer requests a change to an order, including a cancellation request.

The payload includes the current order status at the time of the request. If that status is Shipped, you have a buyer requesting cancellation on an order that is already in transit. That is the highest-risk combination.

This is the signal DD7 Radar monitors for its ATOZ_RISK alert. When an ORDER_CHANGE event arrives with a buyer cancellation request on a shipped order, the seller is alerted immediately — not in a daily digest, not in a weekly summary, but within minutes of the event. Because this is a situation where hours matter.

Your Action Window

When you receive a cancellation request on a shipped order, you have several realistic options. The value of each depends heavily on how quickly you act.

Contact the carrier to intercept the package. This is only possible if the package is still in transit and the carrier offers intercept services. USPS Package Intercept, UPS Package Intercept, and FedEx Delivery Manager all allow senders to redirect or hold packages for a fee. If you can get the package back before it's delivered, you can process the cancellation cleanly — no delivery, no A-to-Z risk, refund issued voluntarily on your terms.

This option is only available in the first day or two after shipment in most cases. Acting within hours of the cancellation request gives you the best chance.

Contact the buyer directly. Reach out through Amazon's messaging system immediately. Acknowledge the request, explain the order has shipped, and offer a clear resolution path — either a return label once the package arrives, or a voluntary refund if they refuse delivery. A buyer who feels heard and has a clear path forward is less likely to escalate to Amazon.

The tone of this communication matters. You are not arguing with the buyer or defending your right to have shipped the order. You are solving their problem as quickly as possible, before they decide to involve Amazon.

Document everything. Regardless of what happens next, having a clear record of your response — timestamped, professional, solution-oriented — strengthens your position if the situation does escalate to a claim. Amazon's reviewers and its automated systems both weight seller response history.

What You Cannot Prevent

It's worth being honest about the limits here. Some buyers will file an A-to-Z claim regardless of how quickly or well you respond. Some claims will be granted regardless of your documentation. Amazon's automated systems are not perfect and sellers do lose claims they should win.

What early action changes is the probability distribution. A seller who contacts the buyer within two hours of a cancellation request on a shipped order is in a fundamentally different position than a seller who finds out about the request three days later in a digest email. The outcome is not guaranteed either way, but the odds are meaningfully different.

There is also a pattern worth understanding: coordinated buyer abuse. Sellers have documented cases of bad actors placing orders with the intent to cancel after shipment and file A-to-Z claims, specifically to extract free merchandise or damage seller ODR metrics. One seller documented 373 orders placed by fake accounts using this pattern.

"A seller documented a buyer placing 373 fake orders, allowing some to ship, then filing A-to-Z claims."

Detecting this pattern early — a cancellation request on a shipped order being the trigger — is the same signal that catches both the ordinary frustrated buyer and the bad actor. The response is the same in either case: act immediately, contact the buyer, attempt carrier intercept.

At scale, the difference between catching these in real time and catching them in a morning digest is the difference between preventing most of them and losing most of them.

Account Health and ODR

Every granted A-to-Z claim contributes to your Order Defect Rate. Amazon's threshold for ODR is 1%. If your ODR exceeds 1%, your account is at risk of suspension.

At 500 orders per month, a single granted A-to-Z claim moves your ODR by 0.2%. Five claims in a month puts you at 1% and into account health review territory. These are not theoretical numbers — sellers in active forums describe the anxiety of watching their ODR approach the threshold and the consequences of crossing it.

"I wake up and I see a CRITICAL event has occurred causing the account to be at risk of deactivation... I can't be the only one that has chest pains when I see it."

The A-to-Z alert is not just about recovering one order's value. It's about protecting the account that processes all your orders. A single auto-grant on a $300 order is a $300 loss. An account suspension from accumulated ODR damage is an existential event.

A Note on Silent A-to-Z Filers

The ATOZ_RISK alert catches the most common and most actionable A-to-Z precursor — cancellation on a shipped order. It does not catch buyers who skip the cancellation step and file an A-to-Z claim directly without any prior signal.

This is worth being transparent about. Some percentage of A-to-Z claims will arrive without warning, and no monitoring tool can change that. What monitoring changes is your coverage of the cases that do have a detectable precursor — which is the majority of cases.

For silent filers, the best defense is the same as it's always been: ship quickly, use trackable shipping, respond to buyer messages promptly, and keep your performance metrics clean. These reduce both the likelihood of a claim and Amazon's willingness to auto-grant one.

Putting It Together

The FBM sellers who handle A-to-Z risk most effectively are the ones who treat cancellation requests on shipped orders as urgent, time-sensitive situations requiring immediate action — not as routine customer service tickets to handle in the next batch.

That shift in urgency is the main thing early detection enables. The situation doesn't change. The options available to you do.

DD7 Radar fires an immediate ATOZ_RISK alert the moment a buyer requests cancellation on a shipped order — giving you the information while your action window is still fully open. If you're currently finding out about these situations in a daily digest or a weekly review, get started with DD7 Radar today.