AI Visibility Score

The AI Visibility Score condenses presence across many questions and several platforms into a single number between 0 and 100. Its purpose is comparison over time, not precision in any single moment.

It is built from a fixed set of questions spanning four intents — brand, category, problem and comparison. Keeping that set unchanged between runs is what makes the number meaningful. If the questions change, the score stops being a trend and becomes a coincidence.

The calculation weighs two things: how often a brand is named, and how prominently. Being the first recommendation in an answer counts for more than being mentioned in a closing aside, because readers rarely get to the aside.

It is worth being clear about what the score is not. It is not a ranking, it is not a guarantee, and it is not comparable between two companies measured on different question sets. Read as an absolute grade it is misleading; read as your own line moving up or down, it is the clearest signal you have.

Most AEO work pays off slowly and invisibly, which makes it difficult to justify continuing. A single number tracked against a fixed question set turns a vague sense of progress into something you can defend to a client or a board — and, just as usefully, shows plainly when three months of effort moved nothing.

A brand appears in 17 of 80 analyzed answers — roughly a 21% presence rate. Two months after publishing comparison pages and getting listed on two review sites, the same 80 questions return 31 mentions. The score moves from 21 to 38. Because the questions never changed, that difference is signal rather than noise.