Measuring
Engines
The assistants we ask, each one separately because they rarely agree: up to four of ChatGPT, Gemini, Copilot, Claude and Google AI Overviews.
All help
What to do
Read them side by side and don’t average them. Strong in one and absent in another is a normal result, and it usually points at thin corroboration.
How it works
- Each engine retrieves differently: different search backends, freshness, and weighting of your site against third-party sources.
- Choose up to four engines under AEO engines in Settings. New workspaces start with ChatGPT, Gemini, Copilot and Google AI Overviews. Claude is the one engine where you also choose the model. The others answer on whatever model their provider serves, so there is nothing to pick.
- ChatGPT, Gemini, Copilot and Google AI Overview are asked as the consumer products. Claude answers come from the Anthropic API with web search, not the Claude app.
- Google AI Overviews does not always produce an answer. An empty overview is recorded as "no result", and it never counts as you being absent.
Read next
Still stuck?
Signed in? Ask Viz inside your workspace. It can see your own results. Or get in touch and a person will answer.