What is AI visibility, and how do you measure it?

AI visibility is how often, and how accurately, your company shows up in the answers AI assistants give when a buyer asks about your category. It is the AI-era version of a ranking. Where a ranking told you where you sat in a list of links, AI visibility tells you whether an assistant names you at all when it writes the answer that replaces that list.

Most companies have never measured it. They know their Google position and their traffic, and assume the two still describe how buyers find them.

Increasingly they do not. A growing share of buyers now get their shortlist from ChatGPT, Google AI Overviews, or Perplexity, and never scroll the list of links at all. This is a plain guide to what AI visibility is, how to measure it, and why a strong search presence does not guarantee it.

What AI visibility means

AI visibility is your presence inside AI-generated answers: whether an assistant cites your page, mentions your brand, or leaves you out when it responds to a question in your category.

It has two parts, and both matter. The first is presence: are you there at all. The second is accuracy: when you are mentioned, does the assistant place you in the right category with the right strengths. A confident but wrong description can cost you more than an absence.

The surfaces that make up your AI visibility are the ones your buyers actually use. Today that means ChatGPT, Google AI Overviews and AI Mode, Perplexity, and the assistants built into the tools your buyers already work in. Each builds its answer differently, so your visibility can be strong on one and weak on another.

AI visibility vs search rankings

A ranking is a position in a list. AI visibility is whether you are named in the answer that sits above the list, or replaces it.

The two used to move together. If you ranked first you got the click, and that was the whole game. In an AI answer there is often no list to rank in: the assistant reads several sources, synthesises one response, and cites a few of them.

You can hold your page-one ranking and still be absent from that response, because the model chose its sources on different criteria. That gap is the reason AI visibility needs measuring on its own. Your ranking report looks healthy while your presence in the answer your buyer actually reads quietly erodes, a pattern I covered in organic traffic falling while rankings hold.

How to measure AI visibility

You measure AI visibility the way you would run any honest test: fix the inputs, repeat the runs, and record what comes back.

Start with a fixed set of prompts. Write the questions your buyers actually ask an assistant: how they describe your category, the problems you solve, and the names of your competitors. Freeze the wording so future runs are comparable.

Run each prompt in a clean session, with memory and personalisation switched off, several times, across the assistants your buyers use. A single run of a non-deterministic system tells you almost nothing, so repetition is not optional.

Record three things for every result: whether you are cited, mentioned, or absent; whether the description of you is accurate; and which competitors appear in your place. That record, repeated over time, is your AI visibility. The full method, including how to capture a baseline before you change anything, is in measuring brand visibility.

What drives your AI visibility

Four things decide whether an assistant can find you, understand you, and trust you enough to cite you.

Retrieval comes first. If an AI crawler cannot reach and render your pages, you never enter the set of sources the model draws from, and nothing else can help. It is worth checking whether your site is blocking AI crawlers before anything else.

Entity clarity comes next. A model has to work out who you are with enough confidence to stake a recommendation on you, and it does that from consistent signals across the web, not from one page.

Then extraction: your key answers have to be written in a form that lifts cleanly into a response. And finally corroboration: independent sources describing you the way you would describe yourself. The mechanism behind all four is in how AI assistants choose sources.

Why AI visibility can be low when SEO is strong

The most common surprise in a diagnostic is a company with strong rankings and weak AI visibility. It almost always comes down to one of three things.

The first is retrieval. Your pages rank because Google indexed them long ago, but an AI crawler is blocked, or your key content only appears after JavaScript runs and the crawler never sees it.

The second is a fuzzy entity. You call yourself a "growth partner" on one page, an "agency" on another, and something else on your profiles, so a model cannot confidently tell who you are and reaches for a clearer competitor.

The third is extractability. Your content ranks, but it never states its answer plainly, so there is no clean passage for an assistant to quote. Strong rankings do not fix any of the three, which is why AI visibility is a separate thing to measure and a separate thing to fix.

The takeaway

AI visibility is the AI-era equivalent of a ranking: whether, and how accurately, an assistant names you when it answers a buyer's question. It is measurable, it is often lower than a company's search presence suggests, and most companies have never checked theirs.

If you want a measured picture of where you stand and which layer is costing you, that is what an AI visibility audit provides.

This article is part of the SEO in the AI Era: The Complete Guide guide.