The New Standard for Brand Visibility in the Age of AI

We have developed proprietary sampling methodology techniques that allow us to poll Large Language Models (LLMs) for their brand preferences. We do this by asking the models thousand of questions about a product category and aggregating an overall score.

Evertune taps into both the developer API and the public interface of AI models (e.g. ChatGPT or Claude) to provide a complete view of how AI perceives a brand across different access points. This dual approach ensures that the full consumer experience is captured alongside any bias included within an AI model’s initial training data.

The AI Brand Score is calculated as a blended metric across all models, prompt types, and thousands of responses. 


The key distinction of AI Brand Score compared to other AI visibility metrics is that it weights frequency (how often the brand appears in responses) with rank (where the brand appears in the response), with top-ranked mentions getting more weight than lower-ranked ones.

For example, a brand mentioned only 40% of the time in responses but which consistently appears at the top will have a higher AI Brand Score than a brand mentioned 70% of the time in responses, but which is always at the bottom. Being mentioned matters, but being mentioned first can have an outsized impact on consumer behavior.

The holy grail AI Brand Score of 100 means that an AI model included a brand in a response in the first ranked position 100% of the time. Here is the scale we use to understand what the score means:

AI Brand Score scale:
90-100 = Excellent visibility
80-70 = Fine
60-45 = Needs improvement
45-25 = Critical
25 and below = Virtually invisible to AI - act now!

No. Traditional SEO focuses on getting pages to rank prominently in search engine results pages (SERPs) and driving clicks. GEO, also called AI Search Optimization or Answer Engine Optimization, is about ensuring a brand is included, summarized, or cited in AI answers instead of, or in addition to, earning that click. The AI Brand Score measures how a brand appears in those answers.

Absolutely. Evertune can generate AI Brand Index scores for any product category. This allows you to see AI preferences for different products, audiences and price points. For example you could see what brands AI recommends for "SUVs" or get more specific with "luxury SUVs" or even call out product features and price points like "3 row luxury SUVs under $100k".

For the AI Brand Index, the questions we ask the models deliberately do not include a brand name. This allows us to see what brands the models are most likely to recommend. We have additional analytics tools that deal with aided awareness and questions directly about a brand.

Absolutely! Sign up today and see how Evertune's AI Brand Index scores are different on each AI model (like ChatGPT or Gemini) and how they change from model version to model version. Our platform also analyzes how brands perform for different product features and consumer preferences, and what words are most AI most associates with a brand.

The scores represent what AI has learned from consuming vast amounts of internet content. Evertune helps brands create content targeted to the topics they are underperforming in and understand which sources are the most influential to AI models. These features result in content roadmaps that are focused on increasing visibility within AI search. We measure the direct impact of these efforts within our platform.