OĞUZ EROLADS & AI

AI Visibility Audit

4 min read12 August 2026

Your customer no longer types a question into Google and picks from ten results. They ask ChatGPT and get a single answer. If that answer mentions three brands and you’re not one of them, that customer never saw you at all. And you can’t see this in Analytics — because the visit never happened.

An AI visibility audit turns this invisibility into something measurable: I ask AI tools the questions people in your industry actually ask, record the answers that come back, and show in writing who gets mentioned and who doesn’t.

What I do in the audit

First we build a question list together — the kind of questions your customers would genuinely ask. “What’s the best X,” “how do you choose Y,” “who do you go to for Z.” Usually 20-40 questions.

Then I ask these questions across different AI tools: ChatGPT, Perplexity, Google AI Overviews, and Gemini. For every answer I record which brands get mentioned, which sources are cited, and where you stand.

Then I move to the why. If you’re not mentioned, the reason is usually one of three: your site is closed off to AI crawlers, your content isn’t written in a citable format, or your brand simply doesn’t appear across enough distinct sources online. Telling these three apart is the most valuable part of the report.

What you get in the report

Not a slide deck — a usable document: a table showing who’s mentioned in which question, your current visibility status, a list of technical blockers, and a prioritized action list.

Every item states what to do and why. You don’t have to hire me to implement it — your own team can do it, or you can hand it to someone else. The list is yours.

Who this makes sense for, and who it doesn’t

It matters for businesses where the customer researches before deciding: consulting, healthcare, education, B2B, travel, legal, software. In these fields people ask AI and the answer directly influences the buying decision.

There are places it doesn’t matter, and I say so upfront: impulsive, on-the-spot purchases, or businesses that are purely local and win business through foot traffic and signage. If that’s you, I wouldn’t recommend the audit — that budget is better spent on ads.

What I did on my own site

The site you’re reading this on right now has a structure open to AI crawlers, an llms.txt file, a connected entity schema, and structured Q&A markup across 285 pages. In other words, I applied this to myself first — you can see live examples of most of what the audit would flag, right here.

If you’re curious about the technical side, see preparing your site for AI search. If you want ongoing tracking after the audit, take a look at AI brand tracking.

How do we start? We have a short call first to talk about your industry and which questions matter. If it makes sense, I send you the scope and timeline in writing. Reach me via the contact page.

Frequently Asked Questions

AI answers change every time — is this measurement reliable?

You’re right, they do change. That’s why I don’t rely on a single query result: I repeat the same question in different phrasings across different tools, then look at the pattern. A one-off answer isn’t proof; your competitor showing up in eight out of ten answers is. I give you that as a ratio in the report.

Do you guarantee I’ll show up in ChatGPT after the audit?

No, and be wary of anyone who does. Nobody can guarantee which source a model will choose. What I do is measure and make fixable the factors that determine your odds — technical accessibility, content format, how widely your brand sources are spread.

I already do SEO — is this different?

Different but connected. SEO moves you up the search results page; AI visibility targets being mentioned inside the answer itself. A solid SEO foundation helps AI visibility but isn’t enough on its own — models read content differently and favor different formats.

How long does the audit take?

It depends on the number of questions and how broad your industry is; a typical audit takes a few business days. I firm up the timeline once we’ve built the question list together in the initial call.