AEO Score Methodology: How AEOprobe Grades Your Website

Most AEO tools give you a score but will not tell you how they calculated it. You get a number, a letter, maybe a color — and you are expected to trust it. AEOprobe takes a different approach: we publish our full scoring methodology. Every category, every weight, every penalty is documented here so you can verify your score, understand what drives it, and make informed decisions about what to fix.

This is not just a philosophical choice. Transparent methodology is a practical requirement. If you cannot see how a score is calculated, you cannot reproduce it, audit it for errors, or confidently prioritize your remediation work. You are operating on faith rather than data.

Why Transparent Methodology Matters

The AEO tool market in 2026 is growing fast, and most platforms keep their scoring proprietary. Some claim 120-point audits without publishing what those points measure or how they are weighted. Others generate impressive-looking dashboards with scores that cannot be independently verified. This opacity creates two problems.

First, you cannot prioritize effectively. If a tool tells you your score is 62 but does not explain which factors dragged it down or how much each factor contributes, you are guessing at where to invest your time. Is your low score because of blocked AI crawlers (critical, fixable in minutes) or because of slow page load times (important, but may require infrastructure changes)? Without methodology transparency, you cannot tell.

Second, you cannot evaluate the tool itself. A scoring system that hides its methodology could be assigning arbitrary weights, double-counting factors, or missing critical checks entirely. The only way to know whether an AEO score is meaningful is to see how it is built.

AEOprobe publishes everything. Here is exactly how your score is calculated.

The 9 Audit Categories

Every AEOprobe audit evaluates your website across 9 distinct categories. Each category targets a specific dimension of AI search readiness — from whether AI crawlers can physically access your site to whether your content is structured in ways that AI engines prefer to cite.

1. AI Bot Access

This is the gatekeeper category. AEOprobe checks your live robots.txt against 14 AI crawlers: GPTBot, ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot, Google-Extended, Bytespider, CCBot, Amazonbot, AppleBot-Extended, Meta-ExternalAgent, Cohere-ai, and AI2Bot.

For each crawler, the audit determines whether your robots.txt explicitly allows access, explicitly blocks access, or has no specific rule (defaulting to the site-wide policy). The category score reflects the percentage of AI crawlers that can access your content, with additional penalties for blocking high-impact crawlers like GPTBot and ClaudeBot.

Why it matters: If your robots.txt blocks GPTBot, your site will not appear in ChatGPT search results regardless of how good your content is. This single configuration file controls whether AI engines can read you at all. We have seen sites with perfect content scores receive an overall F grade because they blocked every AI crawler.

2. Structured Data

AEOprobe validates your JSON-LD, Microdata, and RDFa structured data with field-level error reporting. The audit checks for the presence and correctness of schemas that AI engines use to understand your content — including Article, FAQ, HowTo, Organization, Product, BreadcrumbList, and WebPage schemas.

The scoring evaluates three dimensions: schema presence (do you have structured data at all?), schema validity (does it parse correctly and follow the specification?), and AEO relevance (are you using the schemas that specifically help AI engines extract and cite your content?).

Why it matters: Structured data gives AI engines explicit context about your content. A page with proper FAQ schema tells ChatGPT "here are the questions this page answers" — making it dramatically easier to cite. Missing or malformed schemas force AI engines to infer context from raw HTML, which is less reliable.

3. Content Quality

This category evaluates how well your content is structured for AI extraction. AEOprobe analyzes answer-first paragraph patterns (does your content lead with the answer?), heading hierarchy (is the H1-H6 structure logical and semantic?), content length and depth, and the use of semantic HTML elements.

The audit also checks for AI-extractability patterns — whether your content contains clear, concise statements that AI engines can lift into generated responses without losing meaning. Content that buries answers in long paragraphs or relies heavily on context that would be lost in extraction scores lower.

Why it matters: AI engines prefer content that is structured to be cited. Answer-first paragraphs, clear headings, and concise factual statements are more likely to appear in AI-generated responses than dense, meandering prose.

4. Meta Tags

The meta tags audit checks title tags, meta descriptions, canonical URLs, language declarations, and viewport configuration. For AEO specifically, the audit evaluates whether your title and description provide clear, concise signals that help AI engines understand what your page is about and whether it is relevant to a user query.

Why it matters: Meta tags are among the first signals AI crawlers process when evaluating a page. A missing or misleading title tag can cause AI engines to mischaracterize your content or skip it entirely.

5. Open Graph

AEOprobe validates your Open Graph (og:) and Twitter Card meta tags. While these are primarily used for social sharing previews, they also influence how AI engines preview and characterize your content. The audit checks for og:title, og:description, og:image, og:type, and og:url, plus their Twitter Card equivalents.

Why it matters: AI engines use Open Graph data as supplementary context for understanding your content. Pages with complete OG tags provide richer signals than those without. This category carries less weight than bot access or structured data, but it contributes to the overall picture of how well your site communicates with automated systems.

6. Sitemaps

The sitemap audit checks whether your XML sitemap exists, is properly formatted, is referenced in robots.txt, and contains valid, accessible URLs with accurate lastmod dates. AEOprobe also checks for sitemap index files and validates that URLs in the sitemap return 200 status codes.

Why it matters: AI crawlers use sitemaps to discover content efficiently. A missing, outdated, or malformed sitemap means AI engines may miss pages entirely or waste crawl budget on broken URLs. Accurate lastmod dates signal freshness, which influences how AI engines prioritize your content.

7. Performance

AEOprobe evaluates page load performance factors that affect both user experience and crawler efficiency — including response time, page size, render-blocking resources, and image optimization. The audit does not replicate a full Lighthouse run but focuses on the performance factors most relevant to AI crawler behavior.

Why it matters: Slow pages can time out during AI crawler fetches, resulting in incomplete or failed crawls. AI engines have finite crawl budgets and deprioritize slow sites. Performance is not the most heavily weighted AEO category, but poor performance can undermine every other optimization.

8. Security

The security audit checks HTTPS implementation, mixed content issues, security headers (HSTS, X-Content-Type-Options, X-Frame-Options, Content-Security-Policy), and certificate validity. AEOprobe flags any configuration that could cause AI crawlers to reject or deprioritize your content.

Why it matters: AI engines penalize insecure sites. A page served over HTTP, with mixed content warnings, or with an expired certificate is less likely to be cited as a trusted source. Security is a trust signal — and AI-generated answers prioritize trustworthy sources.

9. Accessibility

AEOprobe evaluates accessibility factors that overlap with AI readability — including image alt text, ARIA landmarks, semantic HTML element usage, heading hierarchy completeness, and form labels. These factors affect both human accessibility and the ability of AI engines to parse your content correctly.

Why it matters: Accessibility and AI readability share the same foundation: clear, semantic, well-structured HTML. Sites that are accessible to screen readers tend to be more accessible to AI engines. Missing alt text, for example, means AI cannot understand your images. Missing ARIA landmarks mean AI cannot identify page regions.

How Grades Are Calculated

Each of the 9 categories produces a score from 0 to 100. These category scores are combined into an overall score using a weighted average that reflects the relative importance of each category to AI search readiness.

Category Weighting

Not all categories contribute equally to your overall AEO score. The weighting reflects a practical reality: some factors are binary gatekeepers (if AI cannot access your site, nothing else matters) while others are incremental improvements.

Category	Weight	Rationale
AI Bot Access	Highest	Gatekeeper — if AI crawlers are blocked, all other optimizations are invisible
Structured Data	High	Directly enables AI engines to understand and cite content accurately
Content Quality	High	Determines whether content is structured for AI extraction and citation
Meta Tags	Medium	Core signals for content characterization by AI crawlers
Sitemaps	Medium	Enables efficient content discovery by AI crawlers
Open Graph	Lower	Supplementary context — helpful but not critical for AI citation
Performance	Lower	Affects crawl success but rarely the primary blocker
Security	Lower	Trust signal — important but typically binary (HTTPS or not)
Accessibility	Lower	Overlaps with AI readability but indirectly rather than directly

Severity Penalties

Within each category, individual checks carry different severity levels. A missing robots.txt file (which blocks nothing but provides no explicit signals) is less severe than a robots.txt that explicitly blocks GPTBot. An incomplete FAQ schema is less severe than completely missing structured data.

Severity levels are classified as:

• Critical — Blocks AI access or fundamentally prevents citation. These issues pull category scores down sharply. Examples: blocking GPTBot in robots.txt, serving pages over HTTP only, returning 5xx errors to crawlers.
• Major — Significantly reduces AI readability or citability. Examples: missing JSON-LD structured data entirely, no answer-first content patterns, missing title tags.
• Moderate — Reduces effectiveness but does not block AI access. Examples: incomplete Open Graph tags, missing alt text on images, no sitemap lastmod dates.
• Minor — Best practice improvements with incremental impact. Examples: suboptimal heading hierarchy, missing Twitter Card tags, large uncompressed images.

Critical issues receive the largest score penalties within their category. A single critical issue in AI Bot Access (like blocking GPTBot) will pull that category score below 50 regardless of how many other bots are allowed.

Letter Grade Mapping

The weighted average of all 9 category scores produces an overall score from 0 to 100. This maps to a letter grade:

Grade	Score Range	What It Means
A+	95–100	Exceptional AI search readiness. All categories scoring high. Your site is technically optimized for AI citation across all dimensions.
A	90–94	Excellent. Strong across all categories with only minor improvements possible. AI engines can access, parse, and cite your content effectively.
B	80–89	Good. Solid foundation with room for improvement in some categories. Most AI engines can work with your content, but you may be missing citations due to specific gaps.
C	70–79	Adequate. Your site has the basics in place but has meaningful gaps — likely in structured data, content quality, or both. Targeted fixes can move you to B or A quickly.
D	50–69	Below average. Multiple categories have significant issues. AI engines can probably access your site but may struggle to extract and cite your content reliably.
F	Below 50	Critical issues. Your site likely blocks AI crawlers, lacks structured data, or has fundamental technical problems that prevent AI engines from working with your content.

How to Use Your Score to Prioritize Fixes

Your AEO score is not just a number — it is a diagnostic tool. Here is how to translate it into action:

Start with Red Categories

AEOprobe color-codes each category: red (critical issues), yellow (needs improvement), green (passing). Always fix red categories first. A red AI Bot Access score means AI engines are blocked — fixing this single issue often produces the largest score jump.

Follow the Severity Hierarchy

Within each category, fix critical issues before major ones, and major before moderate. This is not just about score optimization — it is about impact. A critical robots.txt fix that unblocks GPTBot has more real-world impact than a minor Open Graph improvement, even if both categories show similar percentage scores.

Use the Fix Recommendations

Every issue AEOprobe identifies comes with a specific fix recommendation. These are not generic suggestions like "improve your structured data." They are actionable instructions like "add FAQ schema with the following structure" or "remove the Disallow: / rule for GPTBot in your robots.txt." Follow them in priority order.

Re-audit After Changes

After implementing fixes, run the audit again to verify the improvements registered. AEOprobe audits are free and take 60 seconds — there is no cost to re-checking. This also helps you catch implementation errors (like a JSON-LD syntax error in newly added structured data) before they affect your AI visibility.

Diminishing Returns at the Top

Moving from F to C is usually fast and high-impact. Moving from C to B requires more targeted work. Moving from B to A+ involves incremental improvements across multiple categories. Prioritize getting the foundations right (bot access, structured data, content quality) before chasing perfect scores in supplementary categories like Open Graph or Accessibility.

What Makes This Different

Publishing our full methodology is a deliberate choice that most competitors avoid. When you see an AEO score from AEOprobe, you know exactly what it measures, how it is weighted, and why. You can reproduce it, challenge it, and use it to make informed decisions.

This is also why we publish individual category scores — not just an overall grade. An overall B could mean "strong across the board with minor gaps" or "excellent content quality dragged down by blocked AI crawlers." The category breakdown tells you which interpretation is correct and where to focus your effort.

Run a free AEO audit now to see your score across all 9 categories — and use this methodology guide to understand exactly what it means and what to fix first.