A prompt set is the collection of buyer questions that an AI visibility audit runs. It is fixed so that each run asks the same questions, and it is reviewable so the people relying on the results can see exactly what was asked.
A well-built prompt set covers different buyer intents, labels which questions name your brand, and uses the same wording on every AI platform. Changing questions between runs makes trends meaningless, because any change could come from the wording rather than from AI.
Visibility numbers are only as trustworthy as the questions behind them. If the prompt set is hidden, unbalanced or keeps changing, a rising score can simply reflect easier questions. A stable, visible prompt set makes results defensible to a team or a client.
VisibilityKit Intelligence uses a reviewable prompt set of buyer questions across six commercial intents. Branded prompts are labelled, and the same questions run on every platform it measures.
Yes. The prompt set is reviewable, so you can check exactly which buyer questions produced each result.
So that results are comparable between runs and between platforms. If the questions change, you cannot tell whether a change in visibility came from AI or from the new wording.
A question a real buyer would ask an AI assistant when researching a category, used to measure how AI answers treat your brand.
The commercial goal behind a buyer question, such as discovering options, comparing vendors or checking a specific brand.
A buyer question that names your brand, measured separately from open discovery questions so it cannot inflate your visibility.
The practice of understanding and optimizing for the types of prompts users use when querying AI platforms about your industry.
See how often ChatGPT, Perplexity, Gemini and Google AI Overviews mention, recommend and cite your brand, with the evidence behind every answer.
Explore Intelligence