AI VISIBILITY SCORE
When buyers ask AI about your category, does your brand come up?
Measure your presence. Understand what deserves attention.
Buyers ask AI to recommend products, compare providers and find companies that can solve their problems—often without naming a brand.
Kojable's AI Visibility Score measures how often your brand appears in answers to a defined set of those questions. Our in-house scoring method gives you a transparent number from 0 to 100, backed by the individual questions, platforms and repeated checks behind it.
The score establishes where you appear and where you are absent. Answer Intelligence takes you further: examining what the answers say, how they frame your business and which information gaps deserve investigation.
The score describes a defined sample of AI answers, not your visibility across every AI system or every customer conversation. Measurement scope is agreed separately from the free AI brand audit.
- 0
- No brand exposure
- 100
- Brand exposure on every assessable check
Unbranded buyer questions · Agreed platforms · Repeated checks
A useful number should lead to a better question.
Your company may be absent when buyers ask AI for providers in your category. It may appear for one type of need but disappear when the question becomes more specific.
Or AI may mention your brand frequently while describing the wrong services, overlooking a key product feature or comparing you with the wrong alternatives.
These are different problems. They need different responses.
AI Visibility Score helps you see where your brand enters the answer.
Answer Intelligence helps you investigate what buyers are being told and whether the evidence supports it.
Together, they help your team move from a measurement to a justified next step—not another dashboard number to watch in isolation.
See where your brand enters the conversation.
The AI Visibility Score is a 0–100 measure of brand exposure across a defined set of unbranded buyer questions and AI platforms.
An unbranded question does not contain your company name. It might ask which providers serve a particular industry, which products meet a specific need or which companies are worth comparing.
We track questions that name your brand separately. Those questions help assess how AI represents a company that is already known; they do not measure whether AI brings that company into an unprompted category answer.
The headline score is supported by the underlying question-level results, so you can see differences by platform, topic and buyer need.
- What it tells you
- How often your brand was named in assessable answers within the agreed measurement sample.
- What it does not tell you
- Whether the description was accurate, whether the brand was an appropriate recommendation, why an AI system selected it, or how many buyers saw the answer.
A clear method behind the number.
-
Choose relevant buyer questions
We use a consistent set of questions tied to your category, products, services and buyer needs. Questions that explicitly name your brand are measured separately from the Visibility Score.
-
Repeat each check
In our standard U.S. English setup, we run each question three times on each included platform or AI-search surface. Repeated checks help show variation that a single answer can hide.
-
Record assessable brand exposure
We look for your brand name and approved aliases in the answer text. Each completed, assessable answer that names your brand counts as one exposure, even if it mentions the brand more than once.
A completed answer without your brand counts as zero. A valid Google search that displays no AI Overview also counts as zero for Google AI Overview exposure. Failed requests, incomplete answers, refusals and unresolved brand matches are excluded rather than treated as negative outcomes.
-
Give each question and platform pair equal weight
For each question–platform pair with assessable checks, we divide brand exposures by the number of checks we could assess. We then average those pair-level rates and multiply by 100.
This produces the AI Visibility Score. Results retain the question panel, platforms, market, language, collection period, eligible checks and exclusions needed to interpret it.
Pairs with no assessable checks are excluded from the average and disclosed in the measurement coverage.
A score is comparable over time only when the underlying questions, platforms, eligibility rules and testing conditions are sufficiently consistent. Three repetitions are Kojable's standard U.S. English setup, not a universal statistical guarantee.
One question. Two useful percentages.
“Which consulting firms specialise in data projects for regulated healthcare organisations?”
Imagine that Google is checked three times for AI Overviews. All three searches complete successfully, but an AI Overview appears only once. That one Overview names your brand.
- 33%Brand exposure across valid Google AI Overview checks
-
Your brand appeared in one out of three valid searches.
- 100%Brand mention rate when an AI Overview appeared
-
The one AI Overview that appeared mentioned your brand.
Both figures are correct. They answer different questions. The first shows exposure across the tested searches; the second shows mention frequency within the Overviews that appeared.
The next question is whether the answer described your expertise accurately and presented relevant evidence. That is where Answer Intelligence comes in.
Illustrative calculation only. Not a recorded customer result.
See the pattern behind the number.
Brand mentions by question and platform
See which buyer needs and AI surfaces include your brand, where appearances vary and where your company is consistently absent.
Citations to your website
See when an AI answer cites your pages—even when it does not name your brand. Website citations are a separate signal; they do not automatically count as brand exposure in the score.
Competitor mentions and citations
Track the selected competitors that appear for the same questions, along with citations to their websites. Use Competitor Intelligence to examine how AI compares the alternatives, not just how often each appears.
Change from your baseline
Repeat comparable questions after an agreed change. Review the overall score and the underlying question-level results to see what moved, what stayed stable and where further investigation is justified.
Being mentioned is not the same as being understood.
Consider three possible findings:
- Your brand is absent. The score identifies the questions and platforms where it does not appear. Answer Intelligence and supporting research can help examine the buyer need, available information and possible gaps before recommending work.
- Your brand is mentioned but misrepresented. The appearance counts toward visibility, but the answer may describe your business inaccurately or omit important expertise. Answer Intelligence examines the claim against reliable evidence.
- Your website is cited without your brand being named. The source is present, but the buyer may not connect it with your company. Citation analysis and Answer Intelligence investigate the distinction without assuming what caused the answer.
Kojable connects these findings through its See → Align → Prove → Compound approach.
See — Establish the baseline.
Use the AI Visibility Score and underlying answers to identify where your brand appears and where to look more closely.
Align — Investigate and prepare a justified improvement.
Use Answer Intelligence to examine descriptions, omissions, competitor framing and supporting evidence. Prioritise a scoped change for your team's review and approval.
Prove — Compare what changed.
After an approved change is published, retest comparable buyer questions and report the observed differences. A change in score or answer content does not, by itself, establish causation.
Compound — Build on what you learn.
Use the findings to choose the next worthwhile question, source or page to investigate. Ongoing support is optional and follows an agreed scope.
Compare the AI surfaces your buyers may encounter.
Kojable's current scoring coverage includes ChatGPT, Gemini, Google AI Overviews, Google AI Mode, Meta Muse Spark and Claude. We keep distinct Google experiences separate and record the model or surface, market and testing conditions used.
Meta Muse Spark is measured through the Meta Model API with web-search grounding. This does not mean Kojable measures private or personalised conversations in Instagram, WhatsApp or Facebook. API-based checks of other providers likewise do not cover every consumer application mode.
The exact platforms, markets, languages, question panel and measurement period are confirmed in the agreed scope. The score is not a universal ranking of your business across all AI services.
Better visibility questions for different businesses.
B2B and professional services
See whether your firm enters answers when buyers research specialist expertise, industries, capabilities and potential providers. Then investigate whether the description supports a fair comparison.
Consumer and ecommerce brands
Measure whether your products are mentioned for relevant needs, use cases and category questions. Connect missing appearances to product information and recommendation-fit analysis.
Marketing, growth and agency teams
Use a repeatable baseline to prioritise questions, content reviews and competitor research. Give stakeholders the score and the evidence behind it, then compare subsequent measurements on equivalent terms.
Practical questions about the AI Visibility Score.
What is Kojable's AI Visibility Score?
It is a 0–100 measure of how often a brand appears in assessable AI answers to a defined set of unbranded buyer questions. Each question–platform pair contributes equally to the headline score. The score is interpreted alongside the questions, platforms, market, language and testing period behind it.
Is a higher score always better?
A higher score means your brand appeared more often in the measured sample. It does not necessarily mean the answers were accurate, the recommendations were appropriate or your business gained customers. The value of a change depends on the questions and what the answers actually communicated.
How is the AI Visibility Score different from Answer Intelligence?
The Score measures whether your brand is mentioned. Answer Intelligence investigates what an answer says about your products, services, claims and competitors, and examines the available evidence. It helps turn a visibility finding into a practical question or proposed improvement.
Do website citations count toward the score?
No. The score is based on mentions of your brand name or approved aliases in assessable answer text. Citations to your website are tracked separately, including cases where an answer cites your site without naming your brand.
Why does a Google search without an AI Overview count as zero?
For the Google AI Overview measurement, a valid search with no Overview provides no AI Overview exposure. It therefore counts as zero, not as a failed request. We report Overview appearance and brand mention rates separately so the distinction is clear.
Is this included in the free AI brand audit?
No. Kojable's free initial audit provides a Brand Summary, Associated Topics and Competitor Snapshot. The AI Visibility Score uses a separately scoped question panel, repeated platform checks and defined measurement rules.
Can Kojable guarantee that my score will improve?
No. Third-party AI systems generate answers independently. Kojable measures observed appearances, helps investigate information gaps and compares subsequent results. We do not guarantee mentions, citations, recommendations, ranking changes or revenue outcomes.
Know where you appear. Understand what deserves to change.
Start with a defensible view of how often your brand enters relevant AI answers. Examine the questions behind the score, connect important gaps to Answer Intelligence and decide what work is justified.
New to Kojable? The free AI brand audit is a separate starting point for an initial Brand Summary, Associated Topics and Competitor Snapshot.