How we measure

We answer one question repeatably: when buyers ask AI assistants about a category, does your client get mentioned — and how do they compare to competitors?

The method

• Fixed question set. For each client we track a defined list of buyer-style questions. The brand name is never put into the question — we look for it in the answer.
• Official APIs only. We query each assistant through its official API. We don’t scrape gated products or bypass any access controls.
• Repeat sampling. AI answers vary between runs, so we ask each question multiple times and across scheduled refreshes, then report trends rather than a single snapshot.
• Mention & citation detection. We detect where the brand and named competitors appear, and capture the surrounding snippet and any cited links.
• Share-of-voice. We compare how often the client is named versus its competitors for the same questions.

What we don’t claim

• We don’t guarantee a specific mention, citation, or ranking.
• We don’t present a single answer as definitive — visibility is a trend, not a fixed number.
• We don’t cover sources we can’t access through permitted means; coverage is stated honestly and expands as official access allows.

Want to see it on a real brand? Run a free check →