Why do the four engines give me such different results?
Because they work differently. Perplexity is retrieval-heavy and cites the live web aggressively. ChatGPT blends training knowledge with browsing, so an established brand can surface without being cited at all. Gemini leans on Google's index and its own entity understanding. Claude is more conservative about citing.
Divergence is the normal state, not an error. It's also actionable: strong Perplexity numbers with weak ChatGPT numbers point to a site that's technically readable but not yet an established entity — a different problem from being invisible everywhere.
This is why the report breaks results down per engine instead of showing one blended figure.
Related FAQs
Can I see the actual AI answers behind my score?
Yes — the citation map shows which sources were cited for which queries, per engine, so you can see what the engines drew on rather than take it on faith.
What is authority score, and how is it different from Domain Authority?
Authority score rates the sources citing you, in the eyes of AI engines. Domain Authority rates your own domain, based on backlinks, per an SEO tool.
What is prompt coverage, and why is mine low when my score is decent?
Prompt coverage measures how much of your category's question space you appear in. A high score with low coverage means you're known for one thing only.
How long until my score improves after I make changes?
Weeks, not days — and it varies by fix type. Structural changes can register in one to three weeks; third-party mentions take one to three months.
The score is an estimate — so how much should I trust it?
Trust the direction more than the value. It's reliable for tracking your own movement and comparing competitors on identical queries, not as an absolute.
Is my score comparable to another company's?
Only within the same query set. Two companies in different categories are measured against different questions, so comparing raw numbers is meaningless.
What counts as a good score?
Bands rather than a target number, because what's achievable depends heavily on your category. 0-25 is effectively invisible; 51-75 is reliably visible.
How is the visibility score calculated?
Four inputs, combined into a single 0-100 figure: mention rate, citation rate, prompt coverage and source authority. Citation and coverage dominate.
Can't find what you're looking for?
Related terms
Answer Engine
AEO & GEO FundamentalsaAn answer engine is any platform that responds to a query with a direct, synthesized answer instead of a list of links to click through.
ChatGPT
AI Platforms & EnginescChatGPT is a generative AI chatbot developed by OpenAI, capable of answering questions, browsing the web, and citing sources in its responses.
Claude
AI Platforms & EnginescClaude is a generative AI model developed by Anthropic, used as a chat assistant and integrated into various products and workflows.
Gemini
AI Platforms & EnginesgGemini is Google's family of generative AI models, integrated into Google Search, Workspace, and a standalone chat assistant.
Perplexity
AI Platforms & EnginespPerplexity is an AI-native answer engine built around live web search, known for showing inline citations for every claim in its responses.
