Agentic Commerce

What is Agentic Commerce?

Agentic commerce describes a fundamental shift in how businesses buy and sell. Instead of procurement teams manually browsing supplier websites, requesting quotes, and comparing spreadsheets, AI purchasing agents handle these tasks autonomously. These agents operate within parameters set by the organisation (budgets, quality standards, compliance requirements, preferred supplier lists) and execute purchasing workflows from discovery through to order placement without requiring human intervention at each step.

The term "agentic" distinguishes this model from earlier forms of AI-assisted commerce. Previous generations of tools might recommend products or flag pricing anomalies, but they still required a human to make the final decision and complete the transaction. Agentic commerce goes further: the AI agent has the authority and capability to act independently, making decisions and executing transactions within its defined boundaries.

This shift is enabled by new open protocols, including Google's Universal Commerce Protocol (UCP), Anthropic's Model Context Protocol (MCP), and OpenAI's Agent Commerce Protocol (ACP), that give AI agents standardised ways to discover products, query pricing, and complete purchases across any participating seller.

Why Agentic Commerce Matters for B2B Businesses

For manufacturers and B2B suppliers, agentic commerce changes the rules of visibility. When AI agents handle procurement, they do not browse websites the way humans do. They query structured data feeds, protocol registries, and machine-readable product catalogues. A supplier that lacks structured product data or has not registered with the relevant commerce protocols simply will not appear in agent-driven purchasing flows. Being invisible to AI agents means being invisible to a growing share of B2B spend.

Gartner forecasts that by 2028, $15 trillion in B2B transactions will flow through agent-mediated exchanges. This is not a distant future scenario; it is a transition already underway. Enterprises are deploying AI procurement tools that autonomously manage tail spend, reorder consumables, and identify new suppliers for specialised components. Manufacturers that prepare now, by structuring their product data, registering with commerce protocols, and optimising for agent discovery, will capture this demand. Those that wait risk being excluded entirely.

The commercial implications extend beyond simple discoverability. Agentic commerce compresses sales cycles, reduces the cost of customer acquisition, and opens direct channels between manufacturers and end buyers. For businesses accustomed to selling through distributors and intermediaries, this represents both an opportunity and a disruption: the chance to build direct relationships with buyers whose AI agents can now find them without a middleman.

How Agentic Commerce Works

At its core, agentic commerce relies on three components: the AI agent (acting on behalf of the buyer), the seller's machine-readable product and pricing data, and the protocols that connect them. When a procurement team configures an AI agent with a purchasing requirement (say, 500 units of a specific industrial fastener meeting ISO standards), the agent queries available commerce protocols to identify qualifying suppliers.

The agent evaluates results against its criteria: price, lead time, minimum order quantities, certifications, supplier reputation, and any other parameters the buying organisation has defined. It may request quotes from multiple suppliers simultaneously, compare responses, and select the optimal option. In many configurations, the agent can complete checkout and place the order autonomously, with the human buyer only reviewing exceptions or high-value purchases that exceed predefined thresholds.

For sellers, participation requires publishing product data in formats that AI agents can consume. This includes structured data markup on websites, registration with protocols like UCP, and machine-readable pricing and availability information. The more complete and accurate a seller's data, the more likely their products will surface in agent-driven procurement workflows, and the more likely they are to win the transaction.

Key Takeaways

Related Terms