How to Measure Your AI Visibility Score

To measure your AI visibility score, run a fixed set of buyer prompts across ChatGPT, Perplexity, Gemini, and Google AI Overviews, then score three things for each answer: whether your brand is mentioned, what position you hold versus competitors, and whether you get cited with a link. Turn those into a single 0-100 number and track it weekly. That number is your AI visibility score.

Traditional rank tracking checks where you sit on a page of blue links. AI visibility checks something different: when someone asks an AI assistant to recommend a tool like yours, does the answer name you or a competitor? If you never appear, you are invisible to a growing slice of buyers who never see a search results page at all.

What is an AI visibility score?

An AI visibility score is a single number, usually 0-100, that captures how often and how prominently AI answer engines surface your brand for the questions your customers actually ask. It rolls up three signals into one trackable metric. If you want the longer definition, read what is AI visibility.

• Mention rate: the share of test prompts where your brand appears at all.
• Position: where you land when several brands are named. First mention beats a footnote.
• Citation rate: how often the answer links to your site as a source, not just names you.

How do you measure your AI visibility score?

1. Build a prompt set. Write 15 to 30 prompts a real buyer would type, like "best free SEO tool for solo founders" or "how do I track brand mentions in ChatGPT". Cover your category, your use cases, and comparison queries.
2. Pick your engines. Test the ones your audience uses: ChatGPT, Perplexity, Gemini, and Google AI Overviews. Each behaves differently, so measure them separately before you average.
3. Run every prompt on every engine. Ask each question fresh, in a clean session with no memory of you, so the answer reflects what a stranger would get.
4. Score each answer. Mark mentioned yes or no, note your position among named brands, and record whether you were cited with a link.
5. Roll it into one number. Weight mention, position, and citation, then average across prompts and engines to land on a 0-100 score.
6. Repeat weekly. A single reading is a snapshot. The trend line is what tells you if your work is moving the needle.

Why measure it by hand at all?

Doing one round manually teaches you what the answers actually say. You see the exact competitor that keeps getting recommended instead of you, and the phrasing the AI uses to describe your category. That context shapes every fix you make later. But hand-running 30 prompts across four engines every week is tedious, and results drift between sessions, so most founders automate it after the first pass.

What counts as a good score?

There is no universal pass mark, because it depends on your category and how many rivals compete for the same answers. Treat your first measurement as a baseline, not a grade. A score of 20 that climbs to 45 over a month is a win. A steady 60 that never moves means you have plateaued and need new content or citations.

Watch the three signals independently too. A high mention rate with a low citation rate means AI knows you exist but does not trust you enough to link you. That gap points straight at your fix: earn more authoritative references and structure your pages so they are easy to quote.

Turn the score into action

Measuring is step one. Once you have a baseline, the job is to raise it. Start with the prompts where a competitor wins and yours never appears, then work through content and citations. When you are ready, read how to increase AI visibility and run a free SEO audit to find the pages holding you back.