AI & agentic commerce

Twelve practical AI use cases for online stores, ranked by effort and impact, with the data each one needs and the metric that shows whether it worked.

What an MCP server does for an online store, which capabilities it can expose, and the security controls each one needs.

How AI site search and Adobe Live Search work, and a search audit checklist that fixes zero results, synonyms, typos and merchandising on any store.

Which recommendation types to use, where each one belongs on your store, and how to measure their impact honestly.

How generative and AR try-on work, a category fit table showing which products benefit, and the questions to put to vendors.

A practical prompt-and-review workflow for AI product descriptions, with a reusable prompt template and a QA checklist for reviewers.

What a store chatbot can realistically handle, how to choose between a packaged tool and a custom build, and the guardrails to check before launch.

How ChatGPT finds and shows products, what changed with checkout in 2026, and a sourced checklist for feeds, structured data, reviews and crawler access.

UCP, ACP and MPP compared in one dated table: who maintains each protocol, what it does, its status as of November 2026 and what your store needs to take part.

AI agents now research and sometimes buy for shoppers. Here is how agentic commerce works, what is actually live as of November 2026, and a checklist to get your store ready.

Rule-based vs AI dynamic pricing, the guardrails that keep it safe (floors, MAP, fairness and disclosure rules), and what Adobe Commerce price rules can and cannot do.