Agentic commerce is infrastructure, not a chatbot label
An AI shopping agent can help discover products, clarify requirements, compare candidates, create a cart, and in supported cases move into checkout. Different platforms cover different parts of that journey.
The merchant should separate four layers:
| Layer | Function |
|---|---|
| evidence | product pages, policies, manuals, reviews |
| catalog | structured product identity, attributes, price, stock |
| conversation | questions, comparison, recommendation |
| transaction | cart, payment, order, post-purchase |
Shopify’s role
Shopify Catalog standardizes merchant product data for connected shopping experiences. Agentic Storefronts gives eligible merchants controls over connected AI channels. Shopify and Google co-developed the Universal Commerce Protocol, or UCP, for agent-to-merchant commerce flows.
Shopify says merchants control core product information and channel settings, while each agent decides what to show for a request. Catalog inclusion is not guaranteed placement.
Sources: Shopify Catalog, Shopify agentic storefronts, and Spring 2026 developer update.
OpenAI’s role
OpenAI Commerce documents structured product feeds and checkout integration. ChatGPT Shopping Research can clarify needs, compare products, and use current retail sources. OpenAI’s commerce path is therefore not limited to crawling a merchant page.
Sources: OpenAI Commerce and Shopping Research.
Google’s role
Google combines AI Mode and Gemini shopping experiences with Merchant Center and the Shopping Graph. Merchant Center now documents conversational attributes, including product Q&A, related products, variant options, and document links. UCP checkout is limited by participating merchants, regions, and rollout status.
Sources: Google conversational attributes and UCP checkout.
Claude’s role
Anthropic publicly documents web search with citations and separate web-retrieval crawlers. Its public merchant path is not equivalent to OpenAI product feeds or Shopify Catalog. For Claude, strong public product evidence and source-worthy pages remain central.
Sources: Claude web search and Anthropic crawler controls.
The merchant’s practical response
- Keep Shopify product records complete and current.
- Check Agentic Storefronts and connected channel eligibility.
- Maintain canonical public product evidence.
- Align page, Product
JSON-LD, variants, and feeds. - Publish limitations, compatibility, package contents, and policies.
- Measure real mentions, citations, and errors with fixed prompts.
The trend is clear: product discovery is moving into conversations and, in supported cases, into agent-assisted transactions. The correct response is better commerce data and evidence, not a guarantee that every agent will recommend the store.
Community discussion
Add to the article
Ask a technical question, share a storefront result, or challenge a conclusion with evidence.