AI Visibility describes whether and how visible a brand, website or person is in AI-generated answers. It measures mention, position, description and source linking across multiple platforms and prompts.Open glossary entry → shows whether AI assistants name your brand. RankScan sends your prompts to ChatGPT, Perplexity, Gemini, Claude and Microsoft Copilot. How many of these AI assistants run depends on your plan. RankScan then evaluates every answer: is your brand named, named early, linked?
How to read the overview #
Open “AI Visibility” in the menu and the “Overview” tab. The page has these sections, from top to bottom:
- Score The sentence next to the score says in how many AI answers your brand is named, for example “in 12 of 40”. Next to it is the leader, the brand with the most mentions. The score measures you half against the leader and half against a fixed target.
- Brand awareness Prompts with your brand name show whether AI assistants know you. They do not count towards the score because the brand almost always appears there.
- History The toggle switches between “Named” and “Score”. “Named” shows the share of answers that name your brand. The second line shows the leader.
- Per AI assistant For each assistant you see how often your brand is named and who leads there. The “AI Overviews are AI-generated answer boxes from Google that appear above the classic search results. They summarize information from the web and can change click distribution.Open glossary entry →” row shows how often your website is linked there as a source. It does not count towards the score.
- Comparison with your competitors For each brand you see “named”, “named early”, “Link to website” and “Points”. Your brand is marked “you”.
- Topics For each topic you see the score, mentions in the last three scans and the change. A click opens the topic.
What RankScan counts #
- Named The answer names your brand.
- Named early The mention is in the first three sentences of the answer.
- Link to website The answer links the brand's website.
Only answers to prompts without the brand name count. Follow-up questions and refusals by the AI do not count.
Important #
RankScan measures and compares. The app does not control what AI assistants say about your brand. It shows where you stand, who is named instead and which sources the answers cite.
Tip #
Always read your mentions next to the leader. 12 of 40 answers is a lot if the leader gets 14. If the leader gets 35, that is your biggest lever.
Next step #
The analysis is only as good as your prompts. Write Your Own Prompt Sets shows how to write good ones.