The 90% business-hour rule is 22 days out. We measured who is actually exposed.

In 22 days Amazon starts enforcing a delivery metric that most sellers have never had a reason to look at. On September 30, 2026, the Business Hour Delivery Rate becomes a requirement rather than a number on a dashboard: 90% or higher, measured over a rolling 14-day period, on seller-fulfilled shipments to Amazon Business customers.

We covered how the metric works and how to build a control for it when the announcement landed. This note answers a different question, and it is the one that decides whether the deadline matters to you at all: where does seller-fulfilled supply actually sit? We measured it.

First, what Amazon actually published

There are two separate posts, and they are worth reading as two documents rather than one policy summary.

The US announcement introduces the requirement “to seller-fulfilled shipments in our US store.” It sets September 30 as the start, describes the rolling 14-day measurement, and says that if the rate has not improved by October 30, “your seller-fulfilled offers may be deactivated for Amazon Business customers.” FBA and retail offer eligibility are explicitly not affected.

The UK announcement is titled “Update to FBM requirements for Amazon.co.uk” and carries the same threshold and the same two dates. It also bundles in two handling-time changes that the US post does not mention: from 15 July 2026 the account-level default handling time only offers 0-day and 1-day options, and from 1 September 2026 Amazon enables Automated Handling Time on SKUs whose handling time has run a day or more longer than actual performance for over 30 days.

One difference matters more than the rest. On the protected path — Automated Handling Time, Shipping Settings Automation and Amazon Buy Shipping used together — the UK post says those orders “will not count towards the business hour delivery rate requirement.” That is an exclusion from the denominator, not a guarantee that the order passes. If most of your volume runs on that path, your published rate is computed from the minority of orders that do not, which makes the metric noisier, not safer.

Widely circulated summaries state the rule covers the United States, United Kingdom and Germany. We confirmed Amazon’s own posts for the US store and for Amazon.co.uk. We did not find an equivalent first-party announcement for the German store, so we treat Germany as reported-but-unconfirmed below and measure it anyway, because it is the marketplace with the most to lose if the rule extends.

The seller complaint is a measurement problem, not an effort problem

The US thread is unusually specific. One seller writes that they use “the exact combination you recommend” — Shipping Settings Automation, Automated Handling Time and Amazon Buy Shipping — and ship exclusively via UPS, and their Business Hour Delivery Rate is 83%. Another reports an on-time delivery rate above 99% alongside a Business Hour Delivery Rate of 86%, and asks what action is available to them.

That gap is the whole story. On-time delivery asks whether the parcel arrived by the promised date. Business-hour delivery asks whether the first attempt landed between the hours a business keeps. A seller controls the first question and mostly does not control the second.

What we measured

We pulled the top 30 listings by estimated monthly sales for five business-buyer product terms in each of the three marketplaces — office chairs, printer paper, work gloves, label printers and filing cabinets, using localised German terms — and classified each listing by fulfilment type. That is 450 listings, pulled from Helium 10 Product Research on 2026-09-08.

Marketplace	Amazon retail	FBA	Seller-fulfilled	FBM share
US	48	94	8	5.3%
UK	36	73	41	27.3%
Germany	25	73	52	34.7%

The spread is the finding. Seller-fulfilled supply in these categories is roughly six times more common in Germany than in the United States — and the United States is the store where the requirement is confirmed.

Two readings follow, and they point in opposite directions depending on which side of the fulfilment line you sit on.

Exposure is concentrated, not spread

The category detail matters more than the marketplace average, because the seller-fulfilled listings are not distributed evenly.

Product term	US	UK	Germany
Office chair	0.0%	76.7%	93.3%
Filing cabinet	16.7%	3.3%	50.0%
Printer paper	10.0%	40.0%	30.0%
Label printer	0.0%	16.7%	0.0%
Work gloves	0.0%	0.0%	0.0%

Work gloves were 0% seller-fulfilled in all three stores. Office chairs in Germany were 93.3%. The rule does not land on “FBM sellers” as a class — it lands on furniture and bulk consumables, and barely touches small, fast-moving supplies.

The price and weight profile says the same thing. Median seller-fulfilled price ran well above the FBA median in every store: $104.99 against $29.49 in the US, £79.99 against £14.99 in the UK, and €129.99 against €19.99 in Germany. Some of that is a mix effect rather than a fulfilment effect — chairs and cabinets are simply expensive — but it tells you the exposed units are high-value ones.

Weight points the same way. Of the 101 seller-fulfilled listings, 60 reported a weight; 48.3% of those were 20 lb or heavier, rising to 62.9% in Germany. The remaining 41 listings returned no weight or size-tier data at all, which is common for listings that have never been through FBA, so read the weight figures as describing the 60 that reported.

Heavy goods are exactly the ones a seller cannot steer into a delivery window. A 66 lb office chair moves on a freight or two-person service with its own appointment logic. That is the population where an 83% rate comes from, and it is the population Amazon proposes to switch off for business buyers on October 30.

A note on the data

These are top-selling listings by estimated monthly sales, not a census of every offer, and Helium 10’s fulfilment field describes the listing’s current buy-box fulfilment rather than every offer on the ASIN. Thirty listings per term is a sample; treat single-cell figures such as the UK filing cabinet 3.3% as directional. The marketplace totals rest on 150 listings each, which is where we would put weight. Prices are in each marketplace’s own currency and are not converted.

The Operator Read

The deadline is real, but it is narrow. Most sellers reading this have no exposure at all, and the correct action is to confirm that in ten minutes and move on.

• Check the fulfilment line first, not the metric. If your offers in the affected store are FBA, September 30 does nothing to you. Amazon states plainly that FBA and retail offer eligibility are unaffected.
• If you are FBA in a seller-fulfilled-heavy category, this is an opportunity, not a risk. German office chairs are 93.3% seller-fulfilled and UK office chairs 76.7%. If any share of that supply is deactivated for business customers on October 30, the business buy box has fewer competitors in it. Make sure your business pricing and quantity tiers are live before that date, not after.
• If you are seller-fulfilled, sort your orders by weight before you do anything else. Nearly half the exposed listings we measured were 20 lb or heavier. Light parcels on a standard service are not your problem; freight and two-person deliveries are.
• Do not assume the protected path protects you. Read the wording for your store. In the UK post those orders are excluded from the requirement rather than guaranteed to pass it, and sellers already running all three tools report rates in the mid-80s.
• Read your rate now, in Account Health, while it is still only a number. The first time you read it should not be the day it becomes enforceable.
• If the answer is that you should be moving these SKUs to FBA, price that properly. The comparison is fulfilment cost and peak surcharges against the revenue at risk from a business-customer deactivation, not a general preference for one channel. If you have never separated your B2B economics from your consumer economics, our Amazon listing audit is where that work starts.

The honest summary: this rule is confirmed in two stores, it is measured on a metric sellers only partly control, and in the store where it definitely applies only about one listing in twenty is even eligible to fail it. The place to watch is Europe — where seller-fulfilled supply is five to six times denser, and where the same requirement is already live in one of the two marketplaces we could verify.