Agentic commerce is the pattern where a person delegates a purchase to an AI agent, and the agent carries out discovery, comparison, and checkout. The buyer states intent — "reorder the coffee I bought in March, unless something better-rated is cheaper" — and the agent resolves it against merchant systems, often without any page being rendered for a human to look at.
For conversion work this inverts a core assumption. Nearly every optimization technique developed over the past two decades acts on human perception: headline framing, social proof, visual hierarchy, urgency, friction reduction in a checkout flow. When an agent transacts, none of that is in the loop. What the agent evaluates is structured data — price, availability, specifications, shipping terms, return policy — retrieved through an API or a feed.
The Protocol Landscape
Several competing-but-composable standards emerged across 2025 and 2026:
- ACP (Agentic Commerce Protocol) — from OpenAI with Stripe. Merchants publish a structured product feed and implement the Agentic Checkout Spec so ChatGPT can complete purchases directly.
- UCP (Universal Commerce Protocol) — released by Google as an open standard on 11 January 2026, developed with Shopify, Etsy, Wayfair, Target, and Walmart. It aims to remove the N×N integration problem by giving merchants one integration point for many agent surfaces.
- AP2 (Agent Payments Protocol) — Google's payments-authorization layer, announced September 2025 with 60+ partners including Mastercard, PayPal, and American Express. It carries three signed mandates — Intent, Cart, and Payment — as W3C Verifiable Credentials, establishing who authorized what.
- MCP (Model Context Protocol) — a general standard for exposing tools and data to models; UCP supports integration through it.
These stack rather than compete: ACP and UCP handle the commerce journey, AP2 handles proof of authorization.
Agentic vs. Traditional Conversion
| Traditional | Agentic | |
|---|---|---|
| Who evaluates | A person reading a page | A model reading structured data |
| What persuades | Copy, design, proof, urgency | Price, availability, specs, policy terms |
| Where friction lives | Form fields and checkout steps | Missing or ambiguous machine-readable data |
| Optimization lever | A/B testing the experience | Feed completeness and protocol support |
Where It Actually Stands
Adoption is real but early, and the honest position is that most merchants should not re-platform for it yet. The durable preparation — accurate structured product data, unambiguous machine-readable pricing, stock, shipping, and returns information — improves traditional search visibility and AEO regardless of whether agentic checkout becomes a significant channel.