What is AEO / GEO, and is it different from SEO?

Answer Engine Optimization (also called Generative Engine Optimization) is optimizing to be cited inside an AI-generated answer rather than to rank as a blue link. The foundations overlap heavily with SEO — crawlable, structured, trustworthy content — but the emphasis shifts toward machine-readable attribution and passage-level clarity.

Can I make ChatGPT recommend my site?

Not directly, and anyone selling that is overselling. What you can do is make your site easy to fetch, unambiguous about who you are, and quotable at the passage level — which is what determines eligibility when a model goes looking for sources on your topic.

Should I block AI crawlers?

It depends on your business. Publishers who sell content access often should; a service business that wants to be recommended almost never should. The important thing is that it is a decision, not a default someone set in a plugin.

How is this score calculated?

From signals observable on a single fetched page: structured data types, entity and authorship markers, content depth, heading structure and outbound citations. It is a structural readiness estimate, not a measurement of how often you are actually cited — see the methodology page for the full rubric.

AI search visibility checker

Answer engines don't rank ten links — they pick a handful of sources and cite them. That selection rewards different things than classic SEO: unambiguous entity data, content that survives without JavaScript, and claims a machine can attribute to someone. Enter a domain and we'll check the signals that decide whether your page is usable as a source.

61%score under 50 on AI visibility

Across 959 websites analyzed

Measured across every site in our public database.

Why this matters

Most AI crawlers do not run JavaScript

This is the single biggest and least-known gap. GPTBot, ClaudeBot, PerplexityBot and CCBot fetch HTML and read what is in it. A site that renders its content client-side is, to them, a nearly empty page — no matter how good it looks in a browser. Server-side rendering is not a performance nicety here; it is the difference between existing and not existing as a source.

Being cited requires being attributable

An answer engine needs to say where a claim came from. A page with a named publisher, a date, and an identifiable organisation behind it is a safe citation; an undated wall of anonymous marketing copy is not. This is why Organization schema, author bylines and publish dates carry weight far out of proportion to their effort.

Answer the question in the first two sentences

Models extract passages, not pages. A section that states the answer plainly and then explains it gets quoted; one that builds up to the answer after four paragraphs of context gets skipped. Writing question-shaped headings with the answer immediately underneath is the highest-leverage editorial change you can make.

How to fix it

1. Serve your content in the HTML
View source (not the inspector — the actual source) and search for a sentence from your page. If it isn't there, no AI crawler can read it. Fix with server-side rendering, static generation, or at minimum prerendering for crawler user-agents.
2. Publish an entity record
Organization schema on the homepage with `name`, `url`, `logo`, `sameAs` and address, plus a real About page. This is how a model resolves 'who is this' into a specific company rather than a string it has never seen.
3. Put a date and a name on everything substantive
Published and updated dates, an author with a bio, and — where the claim is factual — a link to the source you got it from. Outbound citations are not a leak of authority; they are evidence that the page is sourced, which is exactly what a citation-selecting model is looking for.
4. Structure for extraction
One H1. Descriptive H2s phrased as the questions people actually ask. Short answer paragraphs directly beneath them. Tables for comparisons and specifications — they survive extraction far better than the same information spread across prose.
5. Decide your robots.txt policy deliberately
GPTBot, ClaudeBot, PerplexityBot, CCBot and Google-Extended each respect robots.txt. Blocking them protects content from training use but also removes you from the answers. Both are defensible; drifting into either by accident is not.

Want this fixed rather than explained?

We do this as fixed-price work — no hourly billing, no retainer. You get a quote before anything starts.

See fixed prices →

Frequently asked questions

What is AEO / GEO, and is it different from SEO?: Answer Engine Optimization (also called Generative Engine Optimization) is optimizing to be cited inside an AI-generated answer rather than to rank as a blue link. The foundations overlap heavily with SEO — crawlable, structured, trustworthy content — but the emphasis shifts toward machine-readable attribution and passage-level clarity.
Can I make ChatGPT recommend my site?: Not directly, and anyone selling that is overselling. What you can do is make your site easy to fetch, unambiguous about who you are, and quotable at the passage level — which is what determines eligibility when a model goes looking for sources on your topic.
Should I block AI crawlers?: It depends on your business. Publishers who sell content access often should; a service business that wants to be recommended almost never should. The important thing is that it is a decision, not a default someone set in a plugin.
How is this score calculated?: From signals observable on a single fetched page: structured data types, entity and authorship markers, content depth, heading structure and outbound citations. It is a structural readiness estimate, not a measurement of how often you are actually cited — see the methodology page for the full rubric.