When you start tracking your brand’s performance in AI search, you’ll encounter a set of metrics that are new to most marketers. Unlike SEO metrics (rankings, organic traffic, domain authority), AI visibility metrics measure different phenomena — presence in AI-generated text, not positions in a ranked list. This guide defines every metric you’ll encounter and explains what it actually measures.
Core Visibility Metrics
AI Visibility Score
A composite index measuring how prominently a brand appears across AI search engines, typically expressed as a number from 0–100. Visibility score aggregates multiple signals — mention rate, citation rate, sentiment, and competitive position — into a single headline number.
What changes it: New citations from authoritative sources, more brand mentions in AI responses, improved sentiment scores, competitor declines.
What to watch out for: Different platforms calculate this differently. A score from one tool is not directly comparable to a score from another tool.
Citation Rate
The percentage of relevant queries where your brand receives a citation (a linked or attributed source) in AI responses. Distinct from mention rate — a citation is a formal attribution, while a mention may be informal.
Formula: (Queries with brand citations) ÷ (Total monitored queries) × 100
Why it matters: Citations drive direct referral traffic. Mentions drive brand awareness. Both matter, but for different goals.
Mention Rate (also: Brand Mention Rate)
The percentage of relevant queries where your brand appears anywhere in the AI response — including informal mentions, comparisons, and recommended lists. Broader than citation rate.
Formula: (Queries where brand is mentioned) ÷ (Total monitored queries) × 100
Benchmarks: Varies widely by industry and query type. For branded queries, expect 80–95%. For category queries (“best project management tools”), 10–40% is competitive depending on category size.
Prompt Coverage
The percentage of your tracked prompt set where your brand appears in at least one response. Measures breadth of coverage across your prompt library rather than density within individual responses.
Use case: Prompt coverage reveals which topic clusters your brand is present in vs. absent from — a useful complement to mention rate for gap analysis.
Share of Voice (AI)
Your brand’s portion of all AI mentions across a category, relative to competitors. If your brand is mentioned 200 times and the total mentions for all tracked brands in your category is 1,000, your AI share of voice is 20%.
Formula: (Brand mentions) ÷ (Total category mentions across all tracked brands) × 100
Important: Share of voice is zero-sum within a defined competitor set. Gaining share means someone else is losing it.
Visibility Index
A normalized rank-weighted score that accounts not just for whether your brand is mentioned, but where in the response it appears. Being named first in a “top tools” list is more valuable than being named last.
Position weighting: First mention typically receives 2–3× the weight of a later mention. Specific weighting varies by platform.
Trend and Change Metrics
Position Drift
A change in where your brand appears within AI responses over time — moving from first mention to third mention, or from prominent recommendation to brief footnote. Position drift can occur without any change in mention rate.
Why it matters: Even if your brand is still being mentioned, declining position in AI responses can signal a shift in how the AI model weights your authority relative to competitors.
Citation Velocity
The rate at which new citations are being acquired over time. A high citation velocity suggests your content is being actively discovered and incorporated into AI model knowledge.
Calculation: New citations per week or month, tracked over a rolling period.
Signals: Spikes in citation velocity often correlate with new authoritative articles, press coverage, or high-authority backlinks being acquired.
Sentiment Score
A measure of whether AI mentions of your brand are positive, neutral, or negative. Calculated by analyzing the tone and framing of AI-generated text about your brand.
Scale: Usually −1 (fully negative) to +1 (fully positive), or expressed as percentages (% positive, % neutral, % negative).
Nuance: AI sentiment is different from social sentiment. AI models reflect the aggregate sentiment of their training data and retrieved sources — shifting it requires changing what authoritative sources say, not just social conversation volume.
Coverage Gap Score
A composite measure of how much relevant AI response surface area your brand is missing from. Combines prompt coverage rate with share of voice to show both breadth and depth of visibility gaps.
Use case: Identifies priority areas for content creation — the topic clusters where the gap between your coverage and competitors’ is largest.
Diagnostic Metrics
Hallucination Rate
The percentage of AI mentions that contain factually inaccurate information about your brand. Measured by auditing AI responses against your known brand facts.
Types: Wrong product features, wrong pricing, wrong founding date, wrong description of services, confusing your brand with a competitor.
Action: Hallucinations are reduced by publishing clear, authoritative, unambiguous information in formats AI models can reliably parse — schema markup, structured FAQs, authoritative third-party coverage.
Source Authority Score
A measure of the quality and authority of sources that are citing your brand or mentioning your brand. High source authority → higher probability your brand’s association carries weight with AI systems.
What it includes: Domain authority of citing sources, editorial standards of publications, relevance of citing content to your category.
Geographic Distribution
A breakdown of AI mention rates and citation rates by region, language, or market. AI models can behave differently across languages and regional versions, producing different brand visibility for the same brand.
Why it matters: A brand with strong US AI visibility may be nearly invisible in European AI responses. Geographic distribution reveals these asymmetries.
Engine Distribution
A breakdown of visibility metrics by AI engine — ChatGPT, Perplexity, Gemini, Claude, etc. Each engine has different training data, retrieval mechanisms, and citation behaviors, producing different brand visibility profiles.
Strategy implication: Optimizing for engine distribution requires understanding which engines your target audience uses and prioritizing accordingly.
Interpreting Your Dashboard
When reading AI visibility metrics, keep these principles in mind:
Absolute vs. relative: Your absolute mention rate matters less than your share of voice relative to competitors. A 15% mention rate in a category where the leader has 10% is a strong position; the same 15% where the leader has 60% is a weak one.
Trend over snapshot: Any single metric reading is less meaningful than its trend over 30–90 days. AI model behavior shifts with model updates, retrieval changes, and shifts in your content and authority signals.
Engine weighting: Weight engine-specific metrics by your audience’s actual engine usage. Gemini visibility matters more if your audience is heavy Google users; ChatGPT visibility matters more for tech-forward audiences.
Lagging vs. leading: Citation velocity and source authority score are leading indicators — they predict future visibility changes. Mention rate and share of voice are lagging indicators — they measure the cumulative result of past optimization work.