Amazon's 2026 Delivery Rate Rule: What Changed

If you ship your own Amazon orders, this is one of the few metrics that can pull products off the site without warning. For years the penalty was blunt. One bad stretch of shipping and Amazon could shut down every self-fulfilled listing you had.

That changed in 2026. The change is real and it is helpful, but a lot of the advice floating around gets it wrong in one direction or the other. Some sellers think the rule was relaxed. Others think a new stricter rule landed. Neither is right, so here is what actually happened.

Reviewed by the SellerShorts editorial bench. SellerShorts runs an AI tool marketplace for Amazon teams.

Our Amazon Listing Optimizer takes an ASIN and returns a full optimized listing, covering title, bullets, description, and backend keywords, in one run. Push live to Seller Central in one click.

What changed on February 28 2026

Amazon changed the consequence, not the standard. Announced through Amazon Seller Central, the update took effect on February 28 2026 and narrowed the scope of the enforcement action tied to a low On-Time Delivery Rate.

• Before: falling below the OTDR minimum could result in the deactivation of your entire self-fulfilled catalog.
• After: Amazon deactivates only the specific listings contributing most to the low rate.
• Unchanged: the 90% minimum.
• Unchanged: the fact that OTDR covers seller-fulfilled orders only.

In plain terms, Amazon swapped a sledgehammer for a scalpel. The standard you have to hit is exactly the same.

What On-Time Delivery Rate actually measures

OTDR is the share of your seller-fulfilled orders that arrive by the delivery date you promised the buyer. It is a delivery metric, not a shipping metric, which is the part sellers most often get wrong.

• It counts arrival, not dispatch. Handing the parcel to a carrier on time does not protect you if it lands late.
• It only covers orders you fulfill. Anything Amazon ships for you under FBA sits outside this metric.
• The bar is 90%. That means roughly one late order in ten is the edge of the cliff, not a comfortable buffer.
• It is rolling. Recent performance is what shows up, so a rough week matters quickly.

One note on the number itself. The 90% figure is the widely used standard and is what sellers see referenced in Seller Central, but Amazon can adjust performance thresholds, and the detailed policy page sits behind a login. Treat 90% as the working bar and confirm the live figure on your own Account Health page before you plan around it.

Because the promise is set by your own handling time and shipping template, most OTDR failures are self inflicted. You promised something faster than your operation can deliver.

Old rule versus new rule, side by side

Read the last row twice. The change makes the outcome smaller. It does not make the outcome harmless.

Who this affects and who can ignore it

This rule matters to you only if you ship orders yourself. That is a smaller group than it sounds, but it includes several common setups.

• FBM sellers. You pick, pack, and ship, so every order counts toward OTDR.
• Seller Fulfilled Prime sellers. You are shipping yourself against a Prime promise, which is the tightest delivery commitment you can take on.
• Hybrid sellers. If some ASINs are FBA and some are FBM, only the FBM side is measured, but a small FBM tail can still sink the rate because the sample is small.
• Pure FBA sellers. Not affected. Amazon owns the delivery, so the metric is not yours to manage.

The hybrid case deserves attention. Sellers with just a handful of self-fulfilled ASINs often ignore OTDR entirely, then get surprised, because with few orders a couple of late deliveries can move the percentage a long way.

Why this is still a real risk, not a relaxation

The new rule limits the damage, but the listings it takes down tend to be the ones you can least afford to lose.

• Volume drives the metric. The listings contributing most to a low rate are usually your best sellers, because they generate the most orders.
• Deactivation kills the sales signal. A listing that goes dark loses momentum, and rebuilding that is slower than losing it.
• The threshold did not soften. You still have to hold 90%.
• Recovery is not automatic. You cannot assume a listing switches back on the moment your numbers improve.

So the correct read is this. Amazon reduced the chance that one bad month ends your business. It did not reduce the need to run delivery properly.

What actually breaks OTDR

Most sellers who fall under 90% did not have a carrier meltdown. They had a promise problem. Here are the usual causes, roughly in order of how often they show up.

• Handling time set too short. One or two days looks great on the listing, then a weekend or a supplier delay eats the buffer.
• Shipping without tracking. If delivery cannot be confirmed, you lose the benefit of the doubt on timing.
• Cheap untracked economy services. Low cost is worth nothing if the delivery date is unpredictable.
• Ignoring weekends and holidays. Your promise runs on calendar days, your warehouse might not.
• Stock you do not physically hold. Drop shipping or shipping from a third party adds a hop you cannot control.
• Peak season without a wider promise. The same handling time that works in March fails in December.

How to protect your rate

Protecting OTDR is mostly about being honest in your settings. Fix the promise first, then fix the shipping.

• Set handling time to what you can hit on a bad day, not a good one. An extra day in the promise costs far less than a deactivated listing.
• Buy shipping through Amazon with tracking. Tracked, confirmed delivery is the cleanest evidence that you met the promise.
• Match the service to the promise. If you promise a date, use a service that can hold that date.
• Widen promises before peak. Adjust in advance, not after the first wave of late deliveries.
• Watch it weekly. OTDR is rolling, so early is the only useful time to react.
• Move fragile ASINs to FBA. If a product keeps missing dates, letting Amazon fulfill it removes it from the metric entirely.

A simple weekly check

This takes a few minutes and prevents almost every OTDR surprise.

If a listing does get deactivated

Do not panic and do not guess. Work the notification you were sent, because it names the specific issue and the route back.

• Read the notification in full. It tells you which listings were affected and why.
• Fix the cause before you appeal. Appealing while the same late deliveries keep happening wastes the attempt.
• Correct the promise. Longer handling time, better carrier, tracked shipping.
• Follow Amazon's stated reinstatement process. Use the instructions in the notice rather than advice from a forum.
• Consider FBA for that ASIN. If the product structurally cannot meet a self-fulfilled promise, change the fulfillment method instead of fighting the metric.

Conclusion

The 2026 change is a genuine improvement. A single rough patch no longer risks your entire self-fulfilled catalog, and that is a meaningful reduction in downside for small teams shipping their own orders.

But the bar is the same 90% it always was, and the listings Amazon removes are usually the ones making you money. Set promises you can keep, ship with tracking, and look at the number every week. While you are protecting your listings, it is worth making sure they earn their keep. Our Amazon Listing Optimizer rebuilds the whole listing in one run. Next reads: why new Amazon sellers fail, a realistic launch plan for new sellers, and what Amazon Seller Central is.

References

1. Amazon Seller Central: On-Time Delivery Rate policy announcement (February 2026)
2. Amazon: seller fulfillment and shipping guidance