GEO Visibility Score
How citable is your site in ChatGPT, Perplexity, Claude and Gemini answers? Your GEO score rates that AI search visibility across the five dimensions that move the new ranking surface. In other words, generative engine optimisation rather than SEO.
Waiting for input
Enter a URL above to start the scan. We store nothing. Refresh and it's gone.
- Fetch homepage HTML
- Check robots.txt
- Read sitemap.xml
- Read llms.txt
- Score across 5 tiers
Live scan · off-site
The live crawler runs on studionine.agency and isn't reachable from this preview. Run the full scan there. It fetches your homepage, robots.txt, sitemap.xml and llms.txt, then scores across all five tiers.
Can AI engines reach your pages at all: robots.txt, sitemap, llms.txt.
Once they're in, do they understand who you are: title, schema, OG.
Standalone, factual sentences an LLM can quote without you in the room.
Recent dates, lastmod on sitemap, dateModified in schema.
Author bylines, Article schema, sameAs links to verified profiles.
Prioritised fixes · ordered by impact
Raw signals (what we found)
Under the hood…
The score is a weighted blend of five tiers. Each tier scores 0–100; the overall score is a weighted average.
- Crawlability · 25%: homepage HTTP 200, robots.txt present, sitemap.xml present,
llms.txtpresent (the emerging convention for an LLM-readable summary of a site), and no AI bots blocked. - Comprehension · 30%: clear title (50–60 chars), 120–160 char meta description, canonical URL, complete OpenGraph, Organization JSON-LD (the single biggest signal), WebSite JSON-LD, and enough content to summarise (~400+ words).
- Citability · 25%: sentences an LLM can quote standalone (8–32 words, third person, with a proper noun or number signal), one clear H1, multiple H2s, headings that lead with claims (not “How…” or “What…” questions), a named author, and BreadcrumbList schema.
- Freshness · 10%: recent year visible in content,
<lastmod>on sitemap entries,dateModifiedin JSON-LD. - Authority · 10%: Article/BlogPosting schema with author + publisher + dates,
sameAslinks to verified profiles (LinkedIn, X, Crunchbase, Wikidata), meta description quality.
What it doesn't do: it doesn't query LLMs to see if you come up for category questions (yet), and it doesn't render JavaScript. Instead, it scores what an AI crawler sees on first fetch. We store nothing. Refresh and it's gone.
A score of 70+ means an AI engine asked a relevant question about your category will likely find, understand and quote you. 40–70 is where most sites sit, fixable in a week. By contrast, below 40 means you're effectively invisible to the new ranking surface.
Start a
project
The best work starts with a conversation, and a decision worth making.
Get in touch
