Surfaced

AI visibility check

Will AI assistants find, trust and quote your website?

Paste any public address. Surfaced fetches it the way a crawler does, scores what it finds, and tells you what to fix first.

Free · no signup · results in about 10 seconds

Scoring is weighted by business type. A dentist is judged on Google reviews; a SaaS product isn't.

The crawlers that decide whether you exist robots.txt
  • OAI-SearchBotBuilds the ChatGPT search indexCites you
  • Claude-SearchBotBuilds the Claude search indexCites you
  • PerplexityBotBuilds the Perplexity indexCites you
  • Google-ExtendedGemini training and groundingTrains only
  • GPTBotOpenAI model trainingTrains only

Blocking the first three removes you from AI answers. Most sites block them by accident, in one line of robots.txt.

What does Surfaced actually check?

Every scan fetches your page without running JavaScript, reads your robots.txt bot by bot, and parses the structured data an engine would rely on. It then scores four dimensions and ranks the gaps by how much each one costs you in AI answers.

  • GEO

    Can crawlers reach the page, read real text in the raw HTML, and see signals of trust like bylines, dates and cited sources?

  • AEO

    Is the content shaped like an answer? Question headings, self-contained sections, schema markup and paragraphs short enough to lift verbatim.

  • Reviews

    Do engines find proof? Review schema, visible testimonials and a linked Google Business Profile they can tie back to your brand.

  • Social

    Is your footprint declared? sameAs entity links and presence on the platforms AI answers cite most: YouTube, LinkedIn and Reddit.

Why doesn't traditional SEO cover this?

SEO earns a ranked link that a person chooses to click. An AI assistant never shows that list. It reads a handful of sources, synthesises one answer, and names whoever it trusted. Ranking well helps, but it no longer decides whether you appear.

What changesTraditional SEOAI visibility
The goalRank above competitorsBe the source that gets quoted
Who reads the pageA browser that runs JavaScriptA crawler that usually does not
What gets extractedThe whole pageOne self-contained section
Access controlGooglebot in robots.txtSix separate citation bots
Proof of qualityBacklinksSchema, reviews, entity links

The practical consequence is that a site can rank on page one and still be invisible to an assistant, usually because its content never reaches the crawler in readable form.

How does the scan work?

A scan takes about ten seconds and touches your site twice: once for the page itself and once for robots.txt. Nothing is stored beyond an anonymous score used to build peer benchmarks, and no crawling continues after the report is generated.

  1. Fetch the raw HTML Surfaced requests your page and reads exactly what comes back, with no JavaScript executed and no rendering. This is the view most citation crawlers get.
  2. Read robots.txt bot by bot Each of the eleven AI crawlers is checked separately, because a rule that allows Googlebot can still block Perplexity.
  3. Parse the signals Headings, schema.org JSON-LD, Open Graph tags, review markup, outbound references, alt text and paragraph shape.
  4. Score and rank the fixes You get four scores out of 100 and a list of recommendations ordered by impact, each one naming what was observed and what to change.

Common questions

What is generative engine optimization?
Generative engine optimization is the practice of structuring a website so AI assistants can reach it, read it, and quote it. Where SEO competes for a ranked link, GEO competes to be the source an assistant cites inside its answer.
Does Surfaced work if my site is built with React or another JavaScript framework?
Yes, and that is often where it finds the biggest problem. Surfaced reads the raw HTML with no JavaScript executed, the same way most AI citation crawlers do. If your content only appears after JavaScript runs, the scan shows you how little a crawler actually sees.
Do I need an account to run a check?
No. Paste a public URL and you get the full report immediately, including every score, the crawler access table and the prioritised fix list. There is no signup and no email required.
Where do the review and social scores come from?
From what your own site declares: schema.org review markup, a linked Google Business Profile, Open Graph tags and sameAs entity links. Those are the signals AI engines actually read. Surfaced does not scrape Google or Meta.
Can I use the results in my own dashboard?
Yes. Every check is also available as JSON at /api/scan with an open CORS policy, so you can call it from a static site, a spreadsheet script or your own reporting stack.