A transparent look at every rule, weight, and data source behind the Auditly AI Readiness Score — so you can trust it and improve it.
Introduction
The AI Readiness Score is not a black box. Every point comes from a specific, measurable signal — checked consistently on every audit.
50+
Measurable signals checked per audit
6
Scoring categories with independent weights
SWOT
Prioritized recommendations and SWOT matrix
0–100
Composite AI Readiness Score range
How It Works
From URL submission to scored PDF report and prioritised recommendations.
User submits any publicly accessible URL. No account or authentication required.
URL is validated, normalized, and passed to the crawler pipeline via a secure server-side API route.
Auditly fetches the raw HTML of the submitted URL server-side, along with robots.txt and available sitemap.
Fetch is performed by a Node.js server action using the native fetch API. A custom User-Agent identifies Auditly. No JavaScript execution — raw HTML only.
The HTML is parsed for heading hierarchy, semantic elements, meta tags, canonical URLs, and Open Graph data.
Parsed with cheerio. Checks run against 15+ Technical SEO rules covering H1 count, H2–H3 nesting, canonical presence, meta description length, and Open Graph completeness.
All JSON-LD script tags are extracted and validated against expected Schema.org types.
Regex + JSON.parse extracts all `<script type="application/ld+json">` blocks. Each block is validated for @type, required properties, and cross-references.
Content is analyzed for AI answer-readiness: FAQ patterns, direct-answer formatting, and question headings.
Rule engine checks for FAQPage schema, question-format H2/H3 headings (regex), opening paragraph structure, and sentence length distribution.
Trust signals, entity consistency, and citation-readiness factors are evaluated.
Checks for About/Contact/Privacy links, author markup, Organization schema sameAs, llms.txt reachability, and content block count.
All raw signals are passed to our intelligence engine, which synthesizes them into a scored report with written recommendations.
Structured prompt with all extracted signals is sent to the AI analysis engine. Response is parsed into six category scores (0–100) and actionable recommendations per category.
The scored report is formatted and a branded PDF is generated for your organization.
PDF rendered with @react-pdf/renderer. Includes executive summary, six category breakdowns, top recommendations, and Auditly branding. Delivered via a signed blob URL.
Scoring System
Each category is scored 0–100. The composite AI Readiness Score is a weighted average of all six categories.
Evaluates the foundational technical health of the site — the prerequisites for any crawler, search engine, or AI system to access and understand the content.
Measures how well the content is structured to be selected as a direct answer by AI answer engines such as ChatGPT, Perplexity, and Google AI Overviews.
Assesses how well the site positions itself to be cited in AI-generated multi-source summaries and Retrieval-Augmented Generation (RAG) contexts.
Checks the technical AI-specific infrastructure — the files, permissions, and rendering approaches that determine whether AI systems can access and understand the site.
Evaluates the trust, authority, and credibility signals that AI systems and quality raters use to determine whether content is worth citing.
Measures how well the content is packaged for AI citation — entity recognition, quotable definitions, and structural clarity.
Weight note: The six weights sum to 100%. Weights are reviewed quarterly as the AI search landscape evolves. All changes are documented in the public roadmap.
Limitations
FAQ
Assess how effectively your content is prepared for answer engines and large language models. Get a comprehensive, prioritized assessment of your site's readiness.
Analyze WebsiteEnterprise Grade · Executive SWOT Reporting · Prioritized Action Plan
