Content Depth

What this parameter measures

This parameter checks whether your page has enough substantive content for what it is. friendly4AI scopes the analysis to the main content area (preferring a <main> block and stripping <nav>, <footer>, and <noscript> noise), then counts the substantive word count and the number of H2/H3 sections. It also detects a site category (blog_docs, ecommerce, landing, or default) from explicit category artifacts, the URL path, and HTML cues, because a documentation page and a landing page have very different reasonable depth. A low-coverage fallback widens the scope to the full page when the main content area accounts for under 65% of the text, so boilerplate-heavy layouts are not unfairly penalised.

Why it matters for AI-readiness

Thin content gives AI systems little to extract, summarise, or cite. A page that covers a topic thoroughly, organised into clear sections, gives a model multiple passages it can pull for different questions, which raises the chance your content is surfaced. Depth is judged relative to category, not in absolute terms: a 400-word product page can be appropriately complete while a 400-word documentation page is plainly thin. Matching real depth to the page's purpose, with proper sectioning, makes your content far more useful to answer engines than a shallow wall of marketing copy.

How we score it

Category-aware thresholds drive this Authority and Trust parameter in the v4.4 methodology. After detecting the category, the processor compares the word count and H2/H3 section count to that category's targets: default requires 1500 words and 3 sections (partial at 500 words); blog_docs requires 2000 words and 4 sections (partial 800); ecommerce 800 words and 3 sections (partial 400); landing 500 words and 2 sections (partial 250). A page passes (100) when it meets both the full word and full section thresholds. It earns a partial (50) when it reaches the partial word threshold OR has at least one fewer than the full section target (minimum 2). Otherwise it fails (0). The applied threshold name is reported in evidence so you can see which category was detected.

How to fix common issues

• Confirm the detected site_category in evidence; depth is judged against that category's thresholds, not a single global number.
• Expand substantive explanatory content to clear the word target for your category, focusing on real information rather than boilerplate.
• Break content into clear H2/H3 sections so the section count meets the category target; flat pages with no subheadings score poorly.
• Reduce navigation- and footer-heavy layouts; if your main content sits below 65% of page text, the low-coverage fallback widens scope but real depth still wins.
• Re-scan and check the word_count, section_count, site_category, and threshold_applied evidence fields.