AI Visibility: The shortlist your customers see before they search

Table of Contents

For most of the last twenty years, being found online meant ranking in search engine pages. A customer typed a few words into a search bar, scanned a page of links, and then clicked. Marketers learned to measure that world in positions, impressions and clicks, and to buy their way onto the page when organic rank wasn’t enough.

That world hasn’t gone away. But a growing share of the research that used to happen there now happens in a conversation with an AI assistant. Someone asks ChatGPT (or Grok, Claude or Gemini) which robo-advisor makes sense for a first-time investor, or what a good alternative to their current provider would be. The assistant answers with a handful of names and a sentence about each. There is no second page to scroll. The customer leaves with a shortlist, and the brands on it got there without anyone seeing an ad or a ranking.

AI visibility is the measure of how often your brand makes that list. I don’t mean how the assistants respond when someone asks about you by name, which tells you very little. I mean how often they name you when people ask the questions that come before a decision: which option is best, what the alternatives are, which one they should choose.

This matters for three reasons:

The first is that losses here are invisible. If an assistant leaves you off the shortlist, nothing in your search or site analytics records it, and the customer simply never arrives.

The second is that it reflects something real. The assistants learn from what the world has written about you, from press coverage and reviews to forum threads and your own pages. So how often they name you is a useful read on how well known you are in your category.

The third is that, like search rankings, it can be moved. What the web says about a brand changes with media, PR and content, and the assistants follow.

That’s why we built the AI Visibility Snapshot at AdNode. We start by agreeing with you on who your real competitors are, and on every name the assistants might use for those competitors. We write eight to twelve questions in your customers’ own language, none of which mention your brand. Then we put each one to ChatGPT, Claude, Gemini and Grok twice, once with web search on and once with it off. The search-off answers show what a model learned in training, and the search-on answers show what today’s web is telling it. Every brand is counted by the same rule, so what comes back is a fair picture: your share of voice against your competitors, the questions where they appear and you don’t, and how each assistant describes you.

The part I find most useful is what happens next. A visibility score on its own is a curiosity. We read each gap as a media decision. Missing from a category shortlist points to paid search on category terms, and to CTV or programmatic when most of the assistants leave you out. A competitor dominating the “alternatives to” questions points to competitor terms and the communities where those comparisons happen. Gaps on educational questions usually call for thought leadership, promoted on LinkedIn to the people you want to reach. Reddit threads show up in these answers constantly, so community presence matters too. For our clients, we run the same questions twice a month, so we can see whether what we bought on your behalf, and your overall marketing efforts, actually moved the number.

If you’ve wondered whether the assistants recommend you, or who they recommend instead, it’s worth finding out. The snapshot is free. You can request one at adnode.net/ai-visibility-snapshot, and one of our team members will walk you through the results on a short call. Even if the answer is uncomfortable, it’s much better to know where you’re starting from.