VISIBILITY · VISIBILITY SCORE
A visibility score you can take apart.
Visibility Score is built from five sourced inputs - Citation Rank, Share of Voice, sentiment, source authority, and structured-data completeness - and always reported with its parts, its spread, and its method. A number you cannot interrogate is not a measurement.
What is an AI visibility score?
An AI visibility score measures how often - and how prominently - AI assistants like ChatGPT, Gemini, and Perplexity surface your brand when shoppers ask what to buy. Tru Commerce's Visibility Score is built from five weighted, sourced signals and benchmarked against your category. We report it with its sub-scores and its run-to-run spread, never as a bare number, because assistant answers are non-deterministic and a single reading hides that.
Visibility Score combines presence, share, sentiment, and AI-readability into a category benchmark you can decompose - every reading ships with its five inputs, the prompt set behind it, and the variance across runs.
- →Composite of 5 sourced inputs, weighted for agent-recommendation impact
- →Benchmarked against your category cohort, refreshed weekly
- →Sub-score breakdown so you fix the right signal
- →Mention rate and recommendation rate reported separately, never blended
- →Published prompt set, run count, and per-engine spread with every reading
How it works
Four capabilities, one signal
Composite of five signals
Citation Rank · Share of Voice · sentiment trend · authoritative source coverage · structured-data completeness. Each weighted by what actually moves agent recommendations.
Category percentile
You're at 62 - top quartile? Bottom half? We benchmark against your category cohort so the number means something.
Sub-score breakdown
When the composite moves, you see which of the five inputs drove it. So you can fix the right thing, not the cheapest thing.
Mentioned and recommended, counted separately
Being named in a comparison answer is not the same as being the brand the assistant puts forward. We never blend the two - the gap between them is usually where the deal is lost.
Variance, not a single reading
Assistant answers are non-deterministic: the same prompt returns different brand sets on different runs. Every score carries its run count and spread, so you can tell a real move from noise.
What a Visibility Score measures
A Visibility Score answers one question a board actually asks: how findable is our brand when an AI agent gets to decide what to recommend? Classic rank tracking tells you where a page sits on a results screen. It says nothing about whether ChatGPT, Gemini, Perplexity, Rufus, Copilot, or Claude reaches for you when a shopper asks what to buy. The Visibility Score is built for that new surface - a single, comparable number for AI findability across your category.
It is deliberately a composite rather than a single reading. Any one signal can be gamed or spiked. A number built from five weighted inputs moves only when your actual standing moves, which is what makes it safe to put in a quarterly deck.
How the score is calculated
Citation Rank (35%) is where your brand lands when agents answer category queries - the largest weight because it is the most direct read on presence. Share of Voice (25%) captures how much of the category conversation you own, not just whether you appear at all. Sentiment (15%) accounts for the fact that being mentioned negatively is not the same as being recommended. Authoritative source coverage (15%) reflects how often the sources agents trust - reviews, editorial, structured feeds - carry your brand. Structured-data completeness (10%) measures how machine-readable your catalog is, because an agent cannot recommend what it cannot parse.
Weights are tuned quarterly against observed model behavior and published when they change, so the number stays honest as the surfaces themselves shift.
How to move it
Because the score decomposes, you fix the right thing instead of the cheapest thing. When it dips, the sub-score breakdown shows whether you lost citation share to a named competitor, whether sentiment softened, or whether a catalog change broke your structured data. Each cause has a different fix - a content and PR push versus a review-response program versus a feed cleanup - and the score tells you which one earns the lift.
Track it weekly, report it monthly. Most brands see the composite respond within a few weeks of the first structured-data and citation-anchor work, well before classic SEO would register the same effort.
The playbook compounds. House of Zelena, a founder-led D2C maternity brand, ran this loop for 6 months: ChatGPT mention share rose 22% → 56%, and the brand moved from #6 to #1 across three major LLMs in its category - against incumbents with 50× the marketing spend.
Case study
From #6 to #1 across 3 LLMs
Go deeper
Related products and proof
Agentic commerce statistics 2026
The first-party corpus behind every score here: 1,795 buyer questions, five engines, three regions, re-run weekly.
Why one visibility number is a vanity metric
Mentioned is not recommended, and one run is not a measurement. The five questions to ask any vendor - including us - before a score goes in a board deck.
Citation Rank
The largest input to your Visibility Score - where your brand lands when agents answer category queries. Free scan in 24 hours.
Share of Voice
The second-largest input - the percentage of category-relevant AI answers your brand appears inside, per platform.
House of Zelena case study
ChatGPT mention share 22% → 56% and category rank #6 → #1 across three LLMs in 6 months - the visibility playbook in practice.
Free Citation Rank scan
The fastest way to see your starting point: drop your URL, get your rank across six AI surfaces in 24 hours.
FAQ
Common questions
A visibility score is a single number that measures how findable your brand is to AI assistants - how often ChatGPT, Gemini, Perplexity, Rufus, Copilot, and Claude surface you when shoppers ask what to buy. Tru Commerce's Visibility Score is a composite of five weighted signals: Citation Rank, Share of Voice, sentiment, authoritative source coverage, and structured-data completeness.
Start with the free Citation Rank scan: drop your URL and get your rank across six AI surfaces in 24 hours, with three prescriptive recommendations. Connecting your catalog then unlocks the full Visibility Score with weekly refresh and sub-score breakdowns.
A good score is relative to your category, not absolute - which is why we report it as a category-cohort percentile. Top quartile in your cohort means agents reach for you more often than most direct competitors; below the median means competitors are taking citations that should be yours.
Fix the sub-score that's dragging the composite. Structured-data and catalog cleanup improves what agents can parse; citation-anchor content and earned media improve Citation Rank and source coverage; review-response programs move sentiment. The sub-score breakdown tells you which lever earns the lift.
Citation Rank 35%, Share of Voice 25%, sentiment 15%, authoritative source coverage 15%, structured-data completeness 10%. Weights are tuned quarterly based on observed LLM behavior; we publish methodology updates.
Yes - Growth and Enterprise include category-cohort percentiles plus head-to-head views against 8 (Growth) or unlimited (Enterprise) competitors.