Answer engines
How each engine decides what to recommend
ChatGPT, Perplexity, Gemini, Claude, and Copilot read the web differently. Here is what each one looks for, and how to check whether it can read your pages.
- ChatGPT Shopping: how it reads product dataWhere ChatGPT gets the price, availability, and product details it shows shoppers, what OpenAI documents about feeds and merchant ranking, and how to check what your product pages send it.
- How ChatGPT decides what to recommendChatGPT does not browse your site the way a person does. Here is how it reads the web, what a page needs to be readable to it, and how to check yours.
- How Claude reads and recommends a siteClaude can fetch and read live pages to answer. Here is how it reads the web, what a page needs to be legible to it, and how to check yours.
- How Google AI Mode picks the pages behind its answersAI Mode answers in a conversation and links out to supporting sites. Here is what Google publishes about eligibility, what a page needs to be usable, and how to check yours.
- How Google Gemini and AI Overviews choose what to showGemini and AI Overviews answer above the blue links. Here is how they read the web, what a page needs to be readable to them, and how to check yours.
- How Grok answers and when it links a sourceGrok answers from live web pages and live X posts, and its citations are optional by design. Here is how it reads, what a page needs to be usable, and how to check yours.
- How Microsoft Copilot picks what to surfaceCopilot answers from live web results and reads the pages behind them. Here is how it reads the web, what a page needs to be readable to it, and how to check yours.
- How Perplexity decides what to citePerplexity retrieves live pages and footnotes its sources. Here is how it reads the web, what a page needs to be one of those sources, and how to check yours.
See how many of the four engines can read your site.
Run a free scan. No signup. You get a readiness score and the gaps to fix, in about a minute.