AI visibility audit
SEO Back Office is a GEO agency. The AI visibility audit is the product every engagement starts with, and this page says what's in one.
What is an AI visibility audit?
An AI visibility audit measures how often a brand gets named and cited in AI-generated answers, which pages those answers pull from, how the brand is described when it does appear, and which competitors get named instead. It’s a log rather than an opinion. You end up with the queries, the platforms, the dates, the cited URLs and a ranked list of the third-party pages that decide your category.
Every engagement at a GEO agency should start here, ours included, because the alternative is buying outreach and content against a guess.
What’s in the query set
Forty to eighty questions, built from what your buyers type into an AI rather than from a keyword tool. The best-X-for-Y questions, the versus questions, the alternatives-to questions, the is-it-worth-it questions, and the ones about the problem rather than the product, where a category gets recommended before any brand does.
You approve the set before anything runs. If your sales team hears a question five times a week and it isn’t in there, it goes in, because those are the ones worth a baseline.
Competitors go in as well. We run the same set against the three or four brands you name, because your mention rate means very little without theirs sitting next to it.
Which platforms and how long it takes
Every query runs across ChatGPT, Perplexity, Gemini, Claude and Google’s AI Overviews and AI Mode. The runs repeat over two to three weeks.
The repetition is the point. Ask the same question on Monday and again on Thursday and you’ll see different brands named, so one pass gives you a screenshot and three weeks give you a baseline. The outreach and the content that follow get measured against it, month after month.
Three weeks is the floor rather than padding. The same brand can be named in most runs one week and in a handful the next with nothing changed on either side, which is why a single check gets treated as an anecdote.
The four fields we log
Each run records whether you were named, where in the answer you appeared, which pages the answer cited, and how you were described. That fourth field catches what a rank tracker never will, like a model quoting pricing you retired two launches ago, or filing you in the wrong category because a directory did it first.
Out of the log come the lists. The citation list ranks every third-party page by how often it shows up across your queries, and that’s your outreach plan. The correction list names every source carrying a wrong description, and that’s your cleanup plan. The gap list shows the answers where a competitor is named and you aren’t.
None of the log is a score. Every row carries the query, the platform, the date, the answer text and the URLs it cited, so anything in the summary can be checked against the run that produced it.
What you get and what happens next
The deliverable is the full log, the citation list, the correction list, the gap list and a written plan for the next ninety days. The plan names the publishers to contact first and the pages on your site to build or fix, with what to expect from each.
Who it’s for: a brand with a product people already compare, and a team that can act on a list of publishers. If nothing on your site ranks yet, the audit will tell you that too, and the honest recommendation then is to fix the ranking first.
It stays useful whether or not you hire us afterwards. The citation list reads the same no matter who works it, and plenty of teams take it in-house. If you’d rather we ran the ninety days, that’s the sprint on the GEO services page, and the methodology page writes out the whole method behind this one, including the parts an audit can’t measure.
See where you stand.
Send us your category and your three closest competitors. We’ll run the first round of queries before the call, so the conversation starts from the answers rather than from a pitch deck.
Request an AI visibility audit No inflated promises. A clear baseline and a practical roadmap.