The hard part of reporting on AI visibility is not gathering the data. It is explaining a number that moves on its own to someone who is paying you to move it.
Start the section with the thing itself: here is what a model says when someone asks about you. Paste the answer. Clients understand a paragraph of text about their own business immediately, and they understand a visibility score not at all until it has moved twice.
The score is how you track it. The answer is what it means.
Set the expectation early that this moves without anyone touching it. Models get updated, sources get reindexed, phrasing shifts. If the first time a client hears this is the month the line goes down, it sounds like an excuse rather than a property of the channel.
Said in advance, it is context. Said afterwards, it is a defence.
Month to month, show what changed in the answers — a new competitor appearing, a citation moving from a support page to a product page, a qualification that has stopped showing up. Those are concrete and attributable.
Save the score trend for the quarterly view, where the noise has had a chance to average out into something that means anything.
You will not be able to prove that a mention in an AI answer produced a sale. Traffic correlation gets you closer, and it is worth showing alongside, but the honest framing is that this is a channel where the brand is either present or absent in the moment of decision.
Clients are generally more comfortable with a clearly stated limit than with a confident claim that does not survive a follow-up question.
Contents
Before you track anything, sit down and read one answer properly. Here's what to look at.
Read →Everyone starts by tracking the prompts they wish customers asked. Here's how to find the ones they actually do.
Read →Not your homepage. A working theory of the sources that shape what AI answers say about a brand, and what that means for where you spend effort.
Read →Tell us your domain and we'll send you a visibility report — free, no card.