Heading Hierarchy Quality
stableCategory: content-structure · Methodology v4.5
It measures the structural quality of a page's heading outline (H1–H6) inside the main content.
Signal Source
- Source
https://{domain}- Kind
- html_dom
Score Bands
| Verdict | Condition |
|---|---|
| Pass | exactly one H1, no skipped heading levels, and at least two distinct heading levels used within the main content (nav, header, footer, and noscript are excluded) |
| Partial | an H1 is present but the hierarchy has issues — multiple H1s, a skipped level (e.g. H1 to H3), or only a single heading level used |
| Fail | no H1 is found in the main content, or a subordinate heading appears before the first H1 |
Description
The Heading Hierarchy Quality parameter checks whether a page uses one clean, logical heading outline. friendly4AI reads the <h1> through <h6> tags in document order and confirms three things: exactly one H1, no skipped heading levels (an H1 jumping straight to H3, say), and at least two distinct heading levels in use. Only main-content headings count. Anything inside <nav>, <header>, <footer>, and <noscript> is stripped before analysis, so boilerplate can't distort the result.
What does this parameter measure?
It measures the structural quality of a page's heading outline (H1–H6) inside the main content. The scan pulls the headings in document order, then checks them against the three conditions above: single H1, no skipped levels, at least two levels.
Why does heading hierarchy matter for AI-readiness?
Headings are the outline AI systems and answer engines follow to work out how a page is organized and what it's actually about. A single, descriptive H1 names the subject. Ordered H2/H3 nesting then lets engines map subtopics and pull the right passage for a given question. Multiple H1s, headings that appear before the main title, or skipped levels blur that outline, and passage extraction gets less reliable as a result. Those are precisely the failure modes this parameter catches.
How is heading hierarchy scored?
Under the v4.5 methodology, this Content Structure parameter scores in three tiers against the main-content headings:
- Pass (100): exactly one H1, no skipped levels, and at least two heading levels in use.
- Partial (50): an H1 is present but the hierarchy has issues — multiple H1s, a skipped level (such as H1 to H3), or only one heading level used.
- Fail (0): no H1 in the main content, or a subordinate heading appears before the first H1.
The processor checks two scopes: the main-content headings, which drive the score, and the full DOM. It reports both in evidence, so a multiple-H1 or out-of-order problem stays visible even when it sits in an excluded region. These bands match the published rubric.
How do I fix heading hierarchy issues?
- Use exactly one H1 for the page title, and place it before any H2 or H3 in the main content.
- Don't skip levels. Follow H1 with H2, and H2 with H3, instead of jumping to a deeper level.
- Use at least two heading levels so the page has real structure rather than a single flat list.
- Got a stray H1 in your header or hero? Confirm which one is the page title. Duplicate H1s drop the score to partial.
- Re-scan after restructuring and check that the
issuesandskipped_levelsevidence fields come back empty.
Heading hierarchy works alongside related Content Structure signals: semantic HTML, no-JS content, and page metadata.
Version History
- Introduced
- v4.2
- Last changed
- v4.5
Key takeaways
- Signal: https://{domain}
- Category: Content Structure
- Passes when: exactly one H1, no skipped heading levels, and at least two distinct heading …