ai visibility is a real, measurable metric now, and you don't need a $500-a-month tool to track yours.
the common advice is "just check if ChatGPT mentions your brand." that's not measurement, that's a vibe check. AI visibility is now a defined, four-part metric with real enterprise tooling built around it, and the fact that the tooling costs hundreds a month doesn't mean the measurement itself is out of reach. it means nobody's shown solo operators the manual version yet.
AI visibility is defined as how often and how accurately a brand is mentioned, cited, or recommended in AI-generated answers. Unlike a search ranking it is closer to binary per answer: you appear or you do not. It has four measurable dimensions: platform breadth, query depth, citation quality, and sentiment.
the four things that actually make up "ai visibility"
Enterprise AI-visibility platforms track four separate dimensions, not one binary yes-or-no:
- Platform breadth, whether you show up on ChatGPT, Perplexity, Google AI, and Gemini, or only one of them.
- Query depth, how many of the actual questions your buyers ask trigger your name, not just the one obvious headline query.
- Citation quality, whether the AI links to your content as a source, or just mentions your name with nothing to click.
- Sentiment, the actual tone you're described in when you do show up. Mentioned favorably and mentioned as "reportedly struggles with X" are both technically a mention.
Here's the one line worth remembering: a high mention rate with weak citations and flat sentiment isn't visibility, it's a name-drop. You need all four to know whether the work is actually paying off.
the manual version, built to run without a subscription
You don't need a $500-a-month platform to get a real read on this. You need a list and thirty minutes a week.
Build a 15 to 20 question prompt list. Split it three ways: evaluation questions ("what's the best [category] for [use case]"), comparison questions ("how does [you] compare to [generic competitor]"), and definitional questions ("what is [your category]"). The evaluation group is where buying decisions actually form. The definitional group is where people who've never heard of you get their first impression.
Run each question through ChatGPT, Perplexity, and Google's AI Mode. Not once, several times over a few days, since the same question can return different answers on different runs. Log what you find:
- Did your brand appear at all? (platform breadth, query depth)
- Was it a linked citation, or just your name with no link? (citation quality)
- What was the actual tone, positive, neutral, or hedged? (sentiment)
Track it over time, not once. A single check tells you where you stand today. A monthly log tells you whether the work you're doing is actually moving anything.
why this is worth the thirty minutes
Roughly 85% of brand mentions in AI answers come from third-party pages, not the brand's own website, according to Similarweb's 2026 research. That means most of the signal deciding whether you show up isn't something you can directly edit, it's something you have to earn and then verify is actually landing. Checking manually is the only way to know if that work is compounding or going nowhere.
And the stakes of getting this wrong compound. Once an AI system starts citing a given source for a given question, it tends to keep citing that same source going forward, citation history becomes its own signal. Waiting to check where you stand isn't neutral. Every month you don't know is a month a competitor's citation history gets a little harder to catch up to.
This is the same shape as the adoption gap that shows up everywhere else in AI right now: most people using AI tools never check whether the setup is actually working, they just assume it is or isn't. Checking is the whole difference.
what to do once you've measured something
If you find weak platform breadth, you likely have a technical access problem, worth checking against the crawler-access basics covered elsewhere in this series. If you find decent mentions but no citations, your content probably isn't structured for a model to extract a specific answer from. If sentiment is off, something you don't control is framing you badly, and that's a different fix than either of the above. The manual tracking sheet doesn't just tell you if you're visible. It tells you which of these three problems you actually have, so you're not guessing at the fix.
If thirty minutes a week isn't the bottleneck and you'd rather have the full teardown done for you, the AI Visibility Audit runs the same four dimensions across every buying query that matters for your category, plus the fix plan in priority order.
Want the condensed version to keep? The 5-Minute AI Visibility Test (PDF), free, no email wall.
want your read on all four dimensions today?
the free 60-second ai visibility scan runs the first pass against your site, results on the page, no email needed.
run the free scanfrequently asked questions
What is AI visibility?
AI visibility is how often and how accurately a brand is mentioned, cited, or recommended in AI-generated answers. It has four measurable dimensions: platform breadth, query depth, citation quality, and sentiment.
How do I track AI visibility without paying for a tool?
Build a list of 15 to 20 real buyer questions, run each through ChatGPT, Perplexity, and Google's AI Mode several times a week, and log four things per result: whether you appeared, whether the mention was linked, the tone, and who else was named.
How often should I check my AI visibility?
Weekly for the log, monthly for the review. A single check tells you where you stand today. A trend over two months tells you whether your work is moving anything, which is the only question that matters.
Why does the same question return different answers?
Because AI answers vary between sessions and shift as models update their search behavior. This is why any single check is closer to a coin flip than a measurement, and why sampling the same question several times matters.
What if I appear often but get no citations?
That points at your own pages rather than your reach. Mentions can come from a model's broad sense of your brand, while a citation requires it to retrieve one of your specific pages and judge it worth linking to.
written by elisabeth hitz, certified in anthropic's ai fluency program (framework & foundations, and ai capabilities & limitations), plus claude 101 and claude cowork. source: similarweb, "what is ai visibility? the complete guide" (shai belinsky, apr 2026), including similarweb's 2026 generative ai brand visibility index. drafted 25 july 2026.