What generative engine optimization actually is.

Generative engine optimization is the practice of making a page easy for an AI answer engine to quote. It is a different job from ranking, and it is measured differently. This page explains the difference and shows the weights our own grader uses.

The one-sentence version

Classic search ranks documents so a person can choose one to click. An answer engine does something else entirely. It retrieves passages, assembles an answer, and names a few sources.

Retrieval is the step where a model gathers candidate passages before it writes anything. A citation is the moment your page is named as the source of a claim in that answer. Citability is the property of a page that makes those two steps likely rather than accidental.

The practical consequence is blunt. A page can rank well and never be quoted, because ranking rewards the document and quoting rewards the passage.

What gets measured

Our grader splits a site's score into two faculties. SEO modules carry 55% of the total and AI-search modules carry the remaining 45%. These are the five that decide the AI-search half.

Module	Weight	What it asks
Entity Clarity	12%	Can a model tell who you are?
Citability	12%	Is there anything quotable here?
Topical Authority	8%	Do you cover the subject in depth?
AI Accessibility	8%	Can the page be read without running JavaScript?
Trust Signals	5%	Is there an author, a date, a source?

What makes a passage quotable

Our Citability module scores each page out of 70 points, split across five things. The thresholds below are the ones the grader actually applies, not general advice.

Specific data. Five or more concrete figures earns full marks. Vague claims earn none.
Structure. At least two of a list, a table, or a definition list.
Questions and answers. Three or more question headings, plus FAQPage markup.
Definitions. Three or more sentences that plainly define a term.
Self-contained sentences. At least 60% of sentences between eight and thirty words.

That last one surprises people. A sentence that needs the previous paragraph to make sense cannot be lifted into an answer, so it is worth less than a shorter one that stands alone.

Why this is a fair fight for a small site

Ranking in classic search leans hard on inbound links, and links take months or years to accumulate. Whether a passage is quotable does not depend on who links to you. It depends on how the page is written, which you control today.

That is the honest reason a small site should care about this now. It is the one part of search where being new is not automatically a disadvantage.

Common questions.

Is GEO just SEO with a new name?

No. They overlap but reward different things. Classic search ranks documents so a person can click one, while an answer engine extracts a passage and quotes it. Our own grader keeps them separate for that reason: SEO modules carry 55% of the grade and AI-search modules carry 45%.

Do I need to remove my SEO work to do GEO?

No, and you should not. Almost every GEO signal is a content or markup change that leaves ranking factors untouched. Adding a definition, a comparison table or an FAQ block does not cost you anything in classic search.

How would I know if an AI engine can already cite me?

Ask one. Put a question your page answers to ChatGPT or Perplexity and see whether your site is named. For a structural read rather than a spot check, run the free scan on this site and look at the Citability and Entity Clarity scores.

Does GEO work on a brand new domain?

It is the one place a new domain can compete. Ranking in classic search depends heavily on inbound links, which take months to earn. Whether a passage is quotable is a property of the page itself, so it does not wait on a link graph.

Want to know where your own site stands on the five modules above? The free scan reads one page and scores all ten areas, with no account and no card.

Run a free scan →