What should an AI visibility report be able to prove?

A visibility percentage is not very useful unless the report can explain where it came from.

Before trusting a dashboard, I think a team should be able to inspect:

• the exact prompt text and version

• the answer surface, locale, date, and sampling conditions

• the raw answer behind every derived mention

• every cited URL, including redirect and normalization decisions

• whether the prompt was discovery or branded verification

• the methodology version and any break in comparability

That evidence also changes what you do next. A missing official source suggests better product documentation. Repeated editorial citations may suggest a comparison or data gap. A methodology change should be annotated instead of celebrated as growth.

We are building this auditable evidence loop at https://corank.ai

Disclosure: I work on Corank.

For founders already tracking AI visibility: what is the one piece of evidence you wish your current report exposed?