How easy is your page to cite?
A search engine ranks pages. An assistant extracts excerpts and credits the source. They are different things, and the latter depends on what can be pulled from your HTML without running JavaScript. Seven criteria, score from 0 to 100.
- Module
- AEO/GEO
- Input
- A URL
- Runs on
- Rovemark server
- Cost
- Free, no signup
Being found is one thing. Being cited is another.
An assistant builds its answer by joining passages from sources and crediting each one. For your passage to make it in, it has to be liftable: present in the served HTML, inside a block with a clear heading, ideally answering a question and anchored to a verifiable fact. A page that only loads text after JavaScript hands the crawler an empty shell.
The score adds up seven criteria with different weights: structured data (22), heading hierarchy (16), content in the served HTML (16), headings phrased as questions (14), facts with numbers (12), authorship and date (10), and liftable blocks such as lists and tables (10). Each one shows what was measured, not generic advice.
- 01
You enter the URL of a page
A specific page, not the domain, citability belongs to the page, and it varies greatly within the same site.
- 02
The HTML is read the way a crawler reads it
Without running JavaScript, exactly as an AI crawler does. Anything that appears only after render does not count.
- 03
The seven criteria become a score
Each criterion shows what was found, how many H2s, how many questions, which schema types, so you know where to make changes.
- 01
Is citability the same as SEO?
- No. Classic SEO competes for position in a list of links. Citability determines whether your excerpt appears in the answer the assistant writes, and there is no second page there. Much of the work overlaps, but the weight is different: schema, hierarchy and served content matter more, link building does not count.
- 02
Why does structured data weigh so much?
- Because it is the only part of the page the model reads without having to interpret it. A JSON-LD block of type FAQPage, HowTo or Article hands over question, answer, author and date already labelled. Everything else in the HTML it has to infer.
- 03
My page has good content and scored low. How come?
- Almost always it is content that only exists after JavaScript, or a page too short in the served HTML. Open your URL with view-source: in front of it and see how much text is really there. That is what the crawler sees.
- 04
Why does a heading phrased as a question score points?
- Because people ask assistants in full sentences, and an H2 written as a question matches the shape of the prompt. A block that opens with the question and answers right below it is the easiest thing there is to lift.
- 05
Is a score of 100 possible?
- It is, but that is not the goal. A short landing page does not need a table or authorship, and not every page wants to be cited. The score is for comparing pages that should be cited but are not.
- 06
How often should I measure?
- When publishing content you want to see cited, and after any template change, changing the theme often breaks heading hierarchy and schema without anyone noticing.
The page is citable. Is it being cited?
The score measures what you control in the HTML. The assistant is the one answering for you out there, and seeing the real answer across several of them is what the platform does.