Amazon Business Hits $60B and the Buyer Is Now a Bot: What UK Sellers Must Do Next

Amazon Business crossed $60 billion in annualized gross sales in Q2 2026, according to Digital Commerce 360, adding 1.8 million new organisational customers in the first half of this year alone. That's up from $35 billion as recently as December 2025. The platform now serves over 11 million organisations worldwide, including 97 of the Fortune 100. By any measure, the numbers reframe B2B ecommerce from a sideline to the main event.

But the headline figure is almost the least interesting part of the story.

The Real Disruption: Agentic AI in the Buying Seat

Running alongside the sales milestone is something with much deeper structural implications for sellers. Agentic AI - software capable of autonomously researching, shortlisting, and in some cases transacting on behalf of a buyer - is moving from enterprise pilots into live procurement workflows. Digital Commerce 360, citing Deloitte analysis published in July 2026, has put the industry on notice: B2B ecommerce operators need to restructure how their product information is built, surfaced, and made machine-readable. Right now. Not after the technology matures.

The core problem is fundamental. Most supplier catalogues were designed for human eyes: product titles written for keyword browsing, pricing locked inside ERP systems unreachable by API, stock availability refreshed on slow batch cycles. Agentic AI agents parse structured data differently. They query attributes programmatically, compare specs automatically, and execute decisions based on buying-organisation rules, without a human touching it at each step. If your data doesn't answer a machine query cleanly, you simply don't appear.

As MarketScale reported this week, the practical implication is blunt: your first buyer interaction may be with a machine that has already filtered out most of your competitors. Winning that first filter requires machine-readable product data, competitive pricing logic, and platform presence. It does not require a great sales pitch.

What This Means for UK Sellers

Amazon UK is already the third-largest Amazon marketplace globally, with an estimated £45 billion in annual GMV and more than 280,000 third-party sellers active on the platform. Amazon Business sits on top of that infrastructure, free to access for registered sellers, offering tiered quantity pricing, automated VAT invoicing, and quote workflows now opening to more third-party sellers following the Reshape 2026 conference announcements.

The VAT piece matters for UK operators specifically. Amazon Business in the UK handles VAT calculation and generates compliant invoices automatically, removing a real administrative burden for smaller sellers trying to pitch into NHS trusts, local authorities, or universities that now make up a meaningful slice of B2B procurement spend. That compliance layer used to be a genuine barrier; it mostly isn't anymore.

The agentic AI shift creates a new, less forgiving one. Many UK SME sellers built their Amazon presence around consumer-facing product descriptions: punchy copy, lifestyle language, benefit-led bullets. That approach worked fine when a procurement manager was skimming the page. It won't work when an AI agent is parsing a structured attribute field looking for ISO certifications, pack configurations, and API-accessible stock counts.

The Catalogue Audit Nobody Is Doing

Deloitte's guidance, as covered by Digital Commerce 360, boils down to a clear sequence: clean and enrich product data schemas, ensure pricing and availability are accessible via API rather than buried in PDFs, and audit how products appear on the platforms where agentic agents are most likely to operate. Companies waiting for agentic AI to become mainstream before acting are already behind. The window to fix catalogue infrastructure is open now, not after the technology is fully deployed.

For smaller sellers this is uncomfortable territory. Catalogue hygiene sounds like enterprise IT busywork. It isn't. At $60 billion in annualized B2B spend flowing through a single platform, the completeness and accuracy of product attributes directly determines visibility to an enormous buyer pool. Incomplete product data is, as MarketScale put it, a direct revenue leak.

Consider where brand-building energy has gone. According to AIBD analysis of IPO (TMD) data, UK Class 25 trademark filings, the registration class covering clothing and apparel goods, totalled just 1,896 in 2026-Q3, down 64.2% versus the prior period. That collapse in new brand filings is a separate story, but it rhymes with a broader pattern: operators are chasing distribution channels at speed while foundational work, from trademark protection to catalogue data architecture, gets deferred. You can't afford both misses simultaneously.

The Numbers That Reset Expectations

Amazon Business grew from $35 billion to $60 billion in annualized gross sales between December 2025 and Q2 2026, a move that should recalibrate every B2B seller's channel priority list. The platform's buyer base spans small businesses through to enterprise accounts, government entities, hospitals, and schools. Its catalogue expansion ran 30% in the first half of 2026 alone.

For any UK brand or distributor with products that could legitimately serve a business buyer, whether office consumables, workwear, catering supplies, tech accessories, or facilities goods, the question is no longer whether Amazon Business is a viable channel. It's whether your product data is structured well enough that an AI agent will ever find you.

The answer, for most sellers right now, is probably not. That's the opportunity.