Agentic commerce guides for online shops

These guides explain, from the merchant's side, how AI assistants find, compare and buy products, and what a shop changes so they can. Each one answers its question in the first lines and names its sources. Foundations covers the basics, Discover covers being named in AI answers, and Convert covers how agents order and pay.

Where should a merchant new to agentic commerce start?

Read the definition first, then what it means for a Swiss shop, then how assistants pick products. The technical guides on protocols, llms.txt, GTINs and variants come after, and measurement last.

  1. What is agentic commerce, and how is it different from ecommerce? The definition, a comparison with ecommerce and the five stages of an agent purchase.
  2. How does agentic commerce work for an online shop in Switzerland? What is live in Switzerland today, and how CHF, VAT, TWINT and delivery reach an agent.
  3. How do you get your products recommended by ChatGPT? What OpenAI says ChatGPT ranks products on, and the order of work for a shop.
  4. GEO vs SEO, what changes for an online shop? Which SEO work still counts for AI answers, and what assistants reward on top of it.
  5. UCP vs ACP vs AP2 vs MCP, which protocol does what in a purchase? Owner, launch date and status of each buying protocol, and which ones your channels need.
  6. What does llms.txt do for ecommerce, next to Markdown and JSON-LD? What Google says about llms.txt, who reads it, and why JSON-LD comes first.
  7. Do AI shopping agents need a GTIN to compare your product? What Merchant Center and the ChatGPT feed ask for, and what to do without a GTIN.
  8. How does ProductGroup schema declare sizes and variants for agents? A JSON-LD example for sizes and colours, with the properties Google supports.
  9. What is share of model, and how do you measure it? How to count how often assistants name you, and how many runs a reliable rate needs.

Where can you see a buying mission from start to finish?

The use cases trace one buying mission or one audit end to end, field by field, from the shopper's request to the checkout. Each one says whether its data was measured on a real shop or constructed to show the mechanics, so you know how far to trust every figure in it.

See all use cases, look up a term in the glossary or read every article by date on the agentic commerce blog.

Test what you just read on your own shop

The free readiness check applies the same rules as these guides to one of your product pages and lists what to fix, in under a minute.

What is stopping agents from buying your products?

Tell us your industry and describe the problem in a sentence or two, in your own words, for example that assistants recommend a competitor or that your prices never appear in AI answers. We reply with the stage we think is losing you sales and what we would check first.