An AI visibility score is a 0–100 number that says how well answer engines know and recommend your brand. It is assembled from four observations across a set of buyer prompts: whether the brand is mentioned, where in the answer it appears, how it is described, and whether the engine cited the brand’s own pages. There is no industry standard, so every tool’s score is its own convention; what matters is the trend in one tool and the diagnosis underneath the number. You can get a first reading in a minute with the free AI visibility checker.
This explainer covers what goes into the score, how the engines differ, what the bands mean, why two tools disagree, and what actually moves the number. It is the methodology page behind the AI visibility pillar; the how-to-check guide covers running the measurement by hand.
The four components
| Component | What is recorded | Why it matters |
|---|---|---|
| Mention rate | Share of prompts where the brand appears at all | The floor. A brand the engine never names has no visibility to improve. |
| Position | First recommendation, one of a list, or a passing mention | Buyers act on the first one or two names; a footnote mention is worth a fraction of a lead. |
| Sentiment and accuracy | Whether the description is positive, neutral or wrong | A confident wrong description (wrong category, wrong price) costs more than absence. |
| Citation | Whether the brand’s own pages are among the sources | Cited pages get the click and shape how the engine describes you next time. |
Tools weight these differently. A share-of-voice tool leans on mention rate and position; a citation-intelligence tool leans on the sources; the checker on this site uses each engine’s own brand-awareness rating as a fast single-run proxy. All of them are trying to compress the same question into a number: when a buyer asks, are you in the answer, and are you in it well?
Why the score differs by engine
The engines retrieve differently, so a brand can score 60 on one and 10 on another. ChatGPT blends model knowledge with web search and leans on widely-cited sources, so brands with press and review coverage do well. Perplexity is retrieval-first and cites pages that answer the question directly, so a well-structured page can win a citation quickly even for a small brand. Gemini and Google AI Overviews draw on Google’s index and ranking signals, so organic SEO strength carries over. Reporting the score per engine is more useful than the blend, because the fix differs: entity coverage for one, page structure for another, plain SEO for the third.
What the bands mean
- Under 20: invisible. The engines have little or no third-party text about the brand. Buyers asking for recommendations in the category will not hear about you. This is where most early-stage and mid-market brands start, and it is not a verdict on the product.
- 20–49: emerging. Known, described thinly, rarely recommended. The engines can place you but do not put you on the shortlist.
- 50–74: established. Described accurately and named in category prompts. The work is now the comparison and “best of” prompts where specific competitors still outrank you.
- 75 and above: leading. Treated as a category reference. The risk is the follow-up turn: a buyer adds a constraint and a competitor with a page on that constraint enters the shortlist.
Why two tools give different numbers
Five reasons, all legitimate. The prompt sets differ (a tool sourcing prompts from search demand asks different questions than one using a curated list). The engines covered differ. The weighting differs. The run date differs, and engines update retrieval continuously. And the engines are non-deterministic: the same prompt minutes apart can return a different list. The best AI visibility tools guide compares the methodologies; the operational rule is simpler: pick one tool, keep the prompt set fixed, and read the trend.
What moves the score
In rough order of leverage for a brand under 50:
- Entity clarity. One canonical brand name, an About page that states what you are and for whom, Organization and Product schema, and consistent descriptions across your site, your app store listings and your social profiles. Engines that cannot resolve the entity cannot recommend it.
- Third-party corroboration. Mentions on the review, comparison and community sources the engines already cite for your category. Every check on this site lists those sources per engine; that list is the outreach plan.
- Answer-shaped pages. One page per buyer prompt, answer in the first two sentences, question-style headings, a comparison table where the prompt is comparative, FAQ markup that matches the visible questions. The AEO guide has the full pattern.
- Freshness. Engines prefer recently updated pages for anything with a price, a version or a date. Refresh the pages that carry your claims.
- Organic strength. Especially for Gemini and AI Overviews, ranking on Google for the category terms is most of the battle.
Above 50, the leverage shifts to tracking: which prompts you lost this week, which competitor entered, and which specific page would have held the slot. That is what AI visibility tracking does, and it is the point where a score stops being a report and becomes a work queue.
A score is a snapshot; tracking is the trend
One run tells you where you stand today. It cannot tell you whether you are improving, which prompts moved, or what changed after you published a page, because every run is a fresh non-deterministic answer. Repeat the same prompts weekly, record per engine, and the score becomes a trend line with a cause attached to each move. Start with the free checker for the snapshot, and read the AI visibility pillar for the rest of the cluster.
Frequently asked questions
What is an AI visibility score?
An AI visibility score is a 0–100 measure of how well answer engines such as ChatGPT, Perplexity and Gemini know and recommend a brand. It is built from how often the brand appears across a set of buyer prompts, where it appears in the answer, how it is described, and whether the brand’s own pages are cited. There is no industry standard; every tool defines its own version.
What is a good AI visibility score?
On the scale this site uses, under 20 means the engines effectively do not know the brand, 20–49 means they know it but rarely recommend it, 50–74 means it is established in category answers, and 75 and above means it is treated as a category reference. Most early-stage and mid-market brands start under 20.
How is an AI visibility score calculated?
Typically: run a fixed prompt set against each engine, record mention (yes or no), position (first, in a list, or a footnote), sentiment and citations per prompt, then weight those into a per-engine score and blend across engines. Some tools instead ask the engine to rate its own awareness of the brand, which is what the free checker on this site does for a single-run snapshot.
Why is my AI visibility score different in different tools?
Because the prompt set, the engines covered, the weighting and the run date all differ, and the engines themselves are non-deterministic. Compare a score with itself over time in one tool, not across tools. The trend is the signal; the absolute number is a convention.
How do I improve my AI visibility score?
Under 20: fix the entity, with a clear About page, Organization and Product schema, one consistent name, and mentions on the review and comparison sources the engines cite. 20–50: publish answer-shaped pages for the buyer prompts, with the answer in the first sentences, comparison tables and FAQ markup. Above 50: track the follow-up prompts where competitors enter, and write the specific page for that constraint.
How often should I check my AI visibility score?
Weekly if you are actively working on it, monthly otherwise. Daily checks mostly measure engine noise. A weekly cadence matches how quickly page changes show up in answer engines and is what AI visibility tracking runs by default.