What a standard SEO audit measures
A useful SEO audit covers the mechanics that let a search engine discover, understand and rank a page: crawl access, indexation, canonicalisation, metadata, internal linking, page speed, structured data, broken links, duplicate content and the keyword-to-page map. It may also cover backlinks and content gaps. These are necessary conditions for being found in a traditional result.
None of those checks reads the answer a buyer sees when they ask an assistant “what are the best tools for this?” That answer is assembled at request time. It may cite a page you have never optimised, or ignore the page that ranks first in Google.
What an AI visibility audit measures
The diagnostic samples a fixed set of buying questions in a logged-out AI session, three times each, and records the answer, the position of the brand, the competitors named and the sources cited. It then classifies those sources and compares the questions where competitors appear and you do not.
The unit is the answer, not the page. One question can produce three different answers; the report shows that spread rather than hiding it behind a single score. See the method for the exact protocol and the limits.
Where the two overlap
They are not opposites. Entity clarity, structured data, crawl access for AI crawlers, authoritativeness and content depth help both. A page that is inaccessible to a crawler will not be read by an answer engine either. A page with no clear claim is hard to cite in either surface.
The difference is the question each one answers. An SEO audit asks: can a search engine find, understand and rank this page? An AI visibility audit asks: when a buyer asks for a recommendation, does the answer include this brand, and which pages made that happen?
How to sequence them
- Fix the mechanical floor first. If the site is not crawlable, not indexed or structurally broken, an AI answer cannot reliably use it. A standard SEO audit is the right first pass; see technical SEO audit.
- Then sample the answers. Run the diagnostic on the ten questions that matter commercially. Do not start from a keyword list; start from the questions buyers ask when they are choosing.
- Work on the pages the answers actually cite. The source audit tells you whether the answer is built from directories, communities, editorial reviews, first-party pages or video. The action is usually on those pages, not in a new blog post.
- Re-run the same questions at day 60. A change in the answer without a change in the inputs is noise. A change that lines up with the work is evidence, not proof.
When an SEO audit is enough
If your buyers search Google and click a blue link, and your category has no meaningful AI answer surface, a standard SEO audit is the cheaper and more direct tool. If your category is a shortlist category — software, agencies, services, “best X for Y” — the AI answer is part of the buying path, and the SEO audit is only half the picture.
When you need the AI layer
You need it when a buyer can ask an assistant for a shortlist and get one without visiting a search result. That is common in B2B software, professional services and any category where the buyer is comparing named providers. The free check on the homepage is enough to see whether the question produces a list and whether you are in it.
Content audit vs AI source audit · Run the free three-question check · See the $850 diagnostic · Read the synthetic sample