Explainer
What is an AI visibility score?
An AI visibility score is a number from 0 to 100 that summarises how easily systems like ChatGPT, Perplexity, and Google AI can find, understand, trust, and cite your business when users ask for recommendations or facts.
Zero means the public signals are too thin or contradictory for a model to rely on you. 100 means your technical and editorial footprint is aligned with best practice across the signals we measure. Most real businesses sit in the middle — and move materially when schema, copy, and citations are fixed.
Why AI visibility scores matter
Traditional SEO answers: “Where do I rank in Google’s list?” AI visibility answers: “Do I appear inside the generated answer users read first?” Those are related but not identical. A household name can dominate blue links yet score poorly if its HTML is thin, schema is missing, or corroboration is fragmented.
Visus publishes illustrative benchmarks: large brands you would assume are “known everywhere” sometimes land in single digits on our rubric because their pages are not structured for extraction. Conversely, a focused local brand with disciplined markup and listings can punch far above its organic footprint.
What Visus measures
The Visus score aggregates roughly 29 signals across six categories. The exact weighting evolves as models and crawler policies change, but the buckets stay stable:
Entity clarity
Can an engine confidently answer “what is this organisation?” Name, description, specific business type, Open Graph tags, canonical URL stability.
Structural signals
Heading hierarchy, JSON-LD quality, question-shaped headings where appropriate, performance basics that affect crawl and render.
Content quality
Depth, factual grounding (numbers, dates, addresses), and whether paragraphs work as standalone citations — roughly 100–200 word blocks a model can quote without surrounding context.
Trust signals
Review markup with authentic aggregates, sameAs links, freshness indicators, and corroborating anchors.
Off-page presence
Mentions on video platforms, forums, Wikipedia/Wikidata where present, and authoritativeness proxies.
Technical access
robots.txt posture toward AI crawlers, presence of llms.txt, server-rendered critical content, clean HTTP responses.
UK benchmarks (illustrative)
Scores move as homepages change; treat these as snapshots for discussion, not live guarantees:
| Brand / example | Visus score (illustrative) |
|---|---|
| BBC | 78/100 |
| Fozias (after optimisation) | 80/100 |
| Nectar | 34/100 |
| Avios | 32/100 |
| Virgin Red | 22/100 |
| British Airways | 8/100 |
A high Google ranking does not imply a high AI visibility score. British Airways can own brand SERPs yet sit at 8/100 on Visus because the public HTML signals we test are weak for generative extraction.
What a good score looks like
- 60–100 (AI-ready): Clear entity, solid schema, accessible to crawlers, quotable copy.
- 40–59 (Developing): Partial implementation — quick wins still available.
- 0–39 (Low visibility): Major gaps; expect absence from most AI answers until fixed.
How to improve your score
Priority order for most SMEs:
- Deploy the correct specific schema type.
- Add AggregateRating that matches real review data.
- Publish llms.txt.
- Confirm AI crawlers are not blocked accidentally.
- Rewrite thin marketing copy into verifiable facts.
- Link a credible YouTube presence from the site.
- Complete directory and GBP citations with consistent NAP.
Check your AI visibility score
Free audit in about 60 seconds. See your score across dozens of signals and what to fix first.
Run free audit