Improve AI-readability score

AI-readability is a score we calculate for every audited page that estimates how easily LLMs can extract, understand and cite the page's content. It's a separate score from SEO because the factors are different. Some pages rank well in Google but score poorly here, and vice versa.

The 8 factors

1. Plain-text content density

LLMs read text, not images or JavaScript-rendered widgets. If 80% of your page is an interactive component, your readability score is low. Fix: surround interactive elements with plain-text summaries.

2. Heading hierarchy

A single H1, then H2s for major sections, then H3s under those. LLMs use this hierarchy to chunk your content for retrieval. Pages with no headings or with skipped levels (H1 → H4) score lower.

3. Answer-shaped paragraphs

When a paragraph starts by directly answering an implicit question ("X is the technique used to..."), LLMs are far more likely to quote it. Wandering intro paragraphs that build to a point get skipped.

4. Schema markup

FAQ schema, HowTo schema and Article schema all give LLMs structured hints. We check for valid JSON-LD on every page and score accordingly.

5. Canonical and metadata clarity

A clear title, description and canonical URL tell the LLM which version of the page to cite. Duplicate or missing canonicals create ambiguity and lower the score.

6. Internal linking

LLMs follow links to gather context. A page that links to 3-7 related internal pages with descriptive anchor text scores higher than an island page.

7. Reading level and sentence length

Sentences over 30 words and Flesch-Kincaid reading levels above 14 (postgraduate) get extracted less reliably. We don't penalize technical content, just unnecessarily complex prose.

8. Citation-worthy specifics

LLMs prefer to cite pages that contain specific numbers, dates, names and definitions. A blog post that says "studies show X is bad" scores lower than one that says "the 2024 Stanford study (n=1,247) found X reduced Y by 38%".

Where to see the score

Every audited page has an AI-readability tab next to its SEO score. The tab shows the overall score, a breakdown by the 8 factors, and the specific recommendations for that page.

Common quick wins

• Add an FAQ section at the bottom of pillar pages, marked up with FAQ schema.
• Rewrite the first sentence of every section so it directly answers what the section is about.
• Split paragraphs longer than 4 sentences.
• Add specific stats, dates and names to support claims.
• Audit headings: one H1, descriptive H2s, no skipped levels.

Expected timeline

Score changes are immediate (next audit). Actual LLM citation rates respond slowly – expect 4-8 weeks before you see new pages start being cited consistently. Track this in AI Visibility → Prompts to confirm.