AI visibility score

An AI visibility score is a metric that summarizes how visible a brand is in the answers of AI assistants such as ChatGPT, Gemini, Microsoft Copilot and Google AI Mode. It usually combines how often the brand is mentioned with how prominently it appears, averaged over many prompts and repeated runs.

Why the AI visibility score matters

An AI visibility score tells you whether buyers meet your brand inside the answer itself, which is often the only place they look. In a Pew Research Center analysis of 900 U.S. adults' browsing in March 2025, Google users who saw an AI summary clicked a traditional result in 8% of visits, against 15% for users who did not see one. They clicked a link inside the summary in just 1% of visits.

When fewer people click, rankings and traffic stop telling the whole story. Microsoft makes the same point about Copilot and Bing: in AI search, "visibility is no longer defined by ranking position alone." A brand that is named in the answer is seen even when nobody clicks. A brand that is left out is not considered at all.

How an AI visibility score works

An AI visibility score turns hundreds of AI answers into one number you can follow over time. Most tools build it in four steps:

Choose the prompts. A set of questions your buyers ask, such as "best accounting software for freelancers" or "alternatives to [competitor]".
Run them on several engines, repeatedly. The same prompt goes to each AI assistant on a schedule, because each run can return a different answer.
Detect the brands in each answer. The tool records whether your brand appears and, if it does, where it sits compared with the other brands named.
Average the results. Each answer gets a score, and the score of the period is the average over all answers.

Repetition is not optional. SparkToro had 600 volunteers run 12 prompts through ChatGPT, Claude and Google's AI a combined 2,961 times. It found "a <1 in 100 chance" that ChatGPT or Google's AI would give the same list of brands in any two of 100 responses, and the order of the list repeated even less often. A single answer is a sample, not a measurement.

There is no industry standard. Each tool weighs presence and prominence its own way, so a score from one tool cannot be compared with a score from another.

Formula

A common way to write the score combines a mention weight and a position weight:

Score of one answer = mention weight + position weight ÷ rank, if the brand is mentioned; 0 if it is not.
AI visibility score = sum of the answer scores ÷ number of answers.

The rank is the brand's place among the brands named in the answer: 1 for the first brand named, 2 for the second, and so on.

Worked example (fictional values). With a mention weight of 0.7 and a position weight of 0.3, a brand is checked in 20 answers over a week. It is named first in 4 answers, second in 5 and fourth in 3, and missing from the other 8.

First: 0.7 + 0.3 ÷ 1 = 1.0, so 4 answers give 4.0
Second: 0.7 + 0.3 ÷ 2 = 0.85, so 5 answers give 4.25
Fourth: 0.7 + 0.3 ÷ 4 = 0.775, so 3 answers give 2.325
Missing: 8 answers give 0

The total is 10.575. Divided by 20 answers, the score is 0.53, shown as 53 out of 100. The same brand has a mention rate of 60% (12 answers out of 20) and an average position of 2.2.

How Hawkscan calculates it

Hawkscan uses the formula above with a mention weight of 0.7 and a position weight of 0.3.

Answers counted. Every completed run of a tracked prompt on ChatGPT, Gemini, Microsoft Copilot or Google AI Mode. A failed run is left out instead of counting as a miss.
Rank. The order of first mention among the brands you track in the project: your brand and your competitors. A brand you do not track never pushes yours down.
Period. The score is the average over the completed runs of the selected period, shown from 0 to 100.

Two properties follow from the weights. Being mentioned matters most: any mention earns at least 0.7, and a better rank adds up to 0.3 more. And the score always sits between 70% and 100% of the mention rate, so a gap between the two tells you how prominent your mentions are.

AI visibility score vs mention rate vs average position

The three metrics come from the same answers but answer different questions.

Metric	Question it answers	Range
AI visibility score	How visible is the brand overall?	0 to 100
Mention rate	How often is the brand named?	0% to 100%
Average position	Where does the brand sit when it is named?	1 = first

How to improve your AI visibility score

Start with the prompts where your score is lowest and read the answers that leave you out. They show which brands the engines prefer and which pages they cite.

Rank for the searches behind the answer. Engines retrieve pages through query fan-out, so the pages that rank for each sub-question are the pages they can use.
Get named where engines read. Comparison articles, review sites and forums that already rank shape the shortlist. Google warns that seeking inauthentic "mentions" isn't as helpful as it might seem.
Keep the basics solid. Crawlable pages, clear headings and answers stated in plain text help every engine use your content.

To follow the result, Hawkscan tracks your AI visibility score every day on each engine, with your mention rate and average position next to it.

Common questions

What is a good AI visibility score?

A good AI visibility score depends on how crowded your category is. In the analysis Gumshoe ran for SparkToro's study, the top brands for headphone prompts often reached 90 to 100% visibility, while the best scores for a brand design agencies prompt sat in the 30s and 40s, a category with far more possible answers. Compare yourself with your competitors on the same prompts, not with an absolute target.

Why does my AI visibility score change when nothing changed on my site?

AI answers vary from one run to the next, even for the same prompt. The engines may also retrieve different pages, update their models or cite new sources about your competitors. Watch the trend over several weeks rather than the move of a single day.