Definition
A system that answers a question with a written answer rather than a list of links: ChatGPT with web search, Perplexity, Google's AI Overviews and AI Mode, Gemini with search grounding.
In the report
The report scores four: ChatGPT with web search, Perplexity, Gemini with Google Search grounding, and Google AI Overviews. Each is asked every question with the buyer's country set and scored separately, so the split between engines is visible.
Worked example
The same question gets a different answer on each engine. In eight runs from September 2026, ChatGPT with web search and Perplexity returned an AI answer to nearly every question; Google AI Overviews returned one for 5 of the 13 to 15 questions on average, and on a UAE gym run for only 2. A stainless steel fabricator was named in 6 of 14 Perplexity answers and 0 of 14 Gemini answers to the same questions. The report scores each engine separately, because an average would hide exactly that.
How to move it
Each engine reads differently. The engine pages say what each one fetches, how it cites and what to change for it.
Where it shows up
What is AI visibility? How often AI answer engines name a company when a buyer asks who to consider in its category, and whether they cite its own pages when they do. It is measured as a share of recorded answers, weighted by position in the list and by citation, across ChatGPT, Perplexity, Gemini and Google AI Overviews.
Why is my competitor in ChatGPT's answer and I'm not? Because a page ChatGPT fetched for that question names them and not you: a review platform where they have more reviews, a roundup they are in, a comparison article, a thread. The engine did not choose them; the page did. The report's citations show which page it was, and that page is the fix.
See answer engine measured on your site
Your website, region and work email. The score appears in about three minutes, and the report is a link you keep.