Getting Started

Understanding Your Scores

Every AEO analysis scores your page across 7 dimensions. This guide explains what each dimension measures, why it matters for AI answer engines, and exactly how to improve your score.

80–100
Well optimized
Strong AI citation potential
50–79
Needs work
Significant gaps to address
0–49
Major opportunity
High-impact improvements available
Note
The overall score is a weighted average of all 7 dimensions. AEO Readiness has the highest weight (25%) because it most directly affects whether AI engines extract and cite your content.
1

AEO Readiness

Weight: 25% of overall score

Does the page have the structural signals AI answer engines look for when deciding whether to cite content?

What it measures

  • Speakable schema markup (tells voice assistants which text to read aloud)
  • FAQ schema — structured question-and-answer pairs that AI engines can extract directly
  • HowTo schema for instructional content
  • Answer-first content structure (does the page answer the question in the first 100 words?)
  • Direct answer density — ratio of direct, declarative statements to hedged or vague language

How to improve

  • Add FAQ schema to pages answering common questions. JSON-LD format is preferred.
  • Move your primary answer to the very first paragraph — don't bury it after preamble.
  • Use HowTo schema for any numbered instruction content.
  • Add Speakable schema to your most important answer paragraphs.
  • Write in direct declarative sentences: "X is Y" rather than "X may be considered Y in certain contexts".
Example

A page scoring 35 on AEO Readiness likely has the information buried in the middle of a long article with no schema markup. The AI has to work hard to find and extract the answer — so it may cite a competitor whose FAQ schema makes the answer instantly accessible.

2

Information Retrieval Cost

Weight: 20% of overall score

How much cognitive and computational effort does an AI system expend to extract a useful answer from your page?

What it measures

  • Answer placement — how many words before the first direct answer appears
  • Content-to-noise ratio — proportion of useful content vs. filler, disclaimers, navigation noise
  • Paragraph complexity — average sentence length, clause nesting, readability score
  • Page architecture — logical flow from question → answer → supporting evidence
  • Chunking — whether content is broken into discrete, independently useful sections

How to improve

  • Use the "inverted pyramid" structure: conclusion first, then supporting detail.
  • Keep paragraphs to 3–5 sentences. Long paragraphs increase extraction cost.
  • Use headers (H2, H3) to label each section with the question it answers.
  • Remove boilerplate: excessive disclaimers, cookie notices embedded in content, promotional padding.
  • Break complex explanations into numbered lists or definition pairs.
Example

A score of 40 here often means the key answer is at paragraph 8 of 12, wrapped in subordinate clauses and hedge words. An AI reading your page has to process 600 words to extract what could be said in 50.

3

HTML5 Semantics

Weight: 15% of overall score

Does the page use correct semantic HTML5 elements that give structure and meaning to content?

What it measures

  • Proper use of <article>, <section>, <header>, <nav>, <main>, <aside>
  • Meaningful heading hierarchy (H1 → H2 → H3, no skipped levels)
  • Use of <figure> and <figcaption> for images and charts
  • Definition lists (<dl>, <dt>, <dd>) for terminology
  • Structured lists (<ul>/<ol>) vs. faux-lists made with line breaks or dashes

How to improve

  • Wrap main page content in a single <main> element.
  • Ensure each page has exactly one <h1> that describes the primary topic.
  • Don't skip heading levels — never jump from H1 to H3.
  • Replace div-soup layouts with semantic elements. A blog post should be an <article>.
  • Use <dl> for FAQ-style content: term + definition pairs.
Example

Many CMS-generated pages score poorly here because page builders output everything as nested <div> elements. AI parsers use semantic structure as a signal of content quality — a well-structured semantic page is easier to parse and more likely to be cited.

4

Accessibility

Weight: 10% of overall score

Is the page accessible to all users and technologies — including the AI systems that crawl and parse it?

What it measures

  • ARIA labels on interactive elements (buttons, links, form fields)
  • Alt text on all meaningful images
  • Heading hierarchy (also affects HTML5 semantics dimension)
  • Link text quality — descriptive text vs. 'click here' or 'read more'
  • Form labels properly associated with inputs
  • Color contrast (evaluated via metadata and inline styles when available)

How to improve

  • Add descriptive alt text to every image. Describe what the image shows, not just its appearance.
  • Replace 'click here' links with descriptive text: "Download the AEO checklist" vs. "click here".
  • Ensure every interactive element has an accessible name (aria-label or visible text).
  • Associate all form inputs with <label> elements using the 'for' attribute.
  • Use tools like Axe or WAVE to audit accessibility issues.
Example

Accessibility improvements directly improve AEO scores because the same semantic signals that help screen readers help AI parsers. A page with good alt text gives AI systems more context about images in the content.

5

E-E-A-T Signals

Weight: 15% of overall score

Does the page demonstrate Experience, Expertise, Authoritativeness, and Trustworthiness — the signals AI engines use to evaluate source credibility?

What it measures

  • Author byline with linked author bio page
  • Author credentials or expertise indicators (job title, qualifications, experience)
  • Publication date and last-updated date (structured data preferred)
  • Citations and links to authoritative external sources
  • Trust signals: company information, contact details, privacy policy links
  • Review or testimonial schema (for product/service pages)

How to improve

  • Add a visible author byline to every article with a link to their bio page.
  • Include a 'Last Updated' date alongside the original publication date.
  • Cite your sources — link to the studies, data, or primary sources you reference.
  • Add Organization schema or Person schema with structured credentials.
  • Include your company's EIN, address, or regulatory body membership where relevant.
Example

AI answer engines, particularly Google's AI Overviews, heavily weight E-E-A-T for YMYL (Your Money, Your Life) content — health, finance, legal topics. A medical article without author credentials and citations scores low even if the content itself is accurate.

6

Content Patterns

Weight: 10% of overall score

Does the content use the structural patterns that AI language models recognise as high-quality, reference-worthy content?

What it measures

  • Definition-first paragraphs: does each concept get defined before it's used?
  • Numbered lists for processes and steps
  • Cause-effect language: explicit causal relationships ('because', 'therefore', 'which results in')
  • Comparison structures: tables, pros/cons, side-by-sides
  • Summary sections: TL;DR, key takeaways, or conclusion blocks
  • Question-answer pairs throughout the content (not just in FAQ schema)

How to improve

  • Start explanatory sections with a one-sentence definition: "X is Y. Here's why it matters..."
  • Add a 'Key Takeaways' or 'Summary' section at the top or bottom of long articles.
  • Convert narrative descriptions of processes into numbered steps.
  • Add comparison tables for any content comparing two or more options.
  • Use explicit transition phrases that signal reasoning: "This means...", "As a result...", "The reason is..."
Example

LLMs are trained on patterns. Content that uses the same structural patterns as Wikipedia, academic papers, and reference documentation — definition first, evidence second, conclusion explicit — scores significantly higher because the model already knows how to extract meaning from these patterns.

7

CRO

Weight: 5% of overall score

Does the page have clear calls-to-action, social proof, and conversion elements that signal a legitimate, active business?

What it measures

  • Presence and prominence of CTAs (call-to-action buttons/links)
  • Social proof elements: testimonials, review counts, trust logos
  • Trust signals: money-back guarantee, security badges, certifications
  • Conversion-supporting copy: benefit statements, value propositions
  • Exit intent or engagement elements (newsletter signup, related content)

How to improve

  • Add at least one clear CTA per page section — not just at the bottom.
  • Include testimonial quotes or aggregate review scores near CTAs.
  • Use trust logos (payment processors, certifications, press mentions) near conversion points.
  • Write benefit-first CTAs: 'Get your free AEO report' beats 'Submit'.
Example

CRO signals matter for AEO because AI engines factor in whether a page appears to be from a legitimate, active business. A page with no CTAs, no social proof, and no clear business intent scores lower on trust signals.