LSI Words in H4 Tags counts how many times SERP-derived related terms appear inside a page’s H4 headings, returning a single occurrence count across all H4 text. A page with no H4s simply scores zero — and because the term list is drawn from the live SERP, this is an observation of how deep-level subheadings echo a topic’s supporting vocabulary, not a target to chase.
| Factor ID | RG-LSI-005 |
|---|---|
| Family | Related Terms (LSI) |
| Measurement | Related-term occurrence count |
| Measured zone | H4 text |
What it measures
This factor counts LSI term occurrences inside H4 headings. It looks specifically at the fourth-level heading text on the page — the deepest structural subheadings most templates use — and tallies how often the topic’s related terms surface there. The result is a raw count of matches, reflecting the degree to which lower-level subtopics are labeled with the supporting vocabulary that pages ranking for the query tend to share.
How RankGear measures it
RankGear parses the H4 text on the page and sums case-insensitive, word-aware matches for each run-level LSI item. Matching is word-aware rather than substring-based, so a related term is credited only when it appears as a whole word, and every occurrence is counted — the same term found in three separate H4s contributes three to the total.
How to read it
| If the page… | What the count reflects |
|---|---|
| Has no H4 headings | Scores zero — there is no H4 text to search, which is expected and not a penalty to design around. |
| Uses H4s that name real sub-subtopics | Related terms tend to appear naturally, producing a modest, honest count. |
| Repeats the same related term across many H4s | The count rises, but occurrences are tallied, not distinct terms — repetition inflates the number without adding coverage. |
How to optimize it
Use related terms in H4s only when they precisely label legitimate lower-level subtopics. This is an observation of your existing structure rather than a target to hit: if your content is deep enough to warrant H4-level headings, name those sections accurately and the relevant vocabulary will follow. Do not add H4s, or reword honest ones, purely to raise the count — a heading that misrepresents its section serves neither readers nor the measurement.
Important considerations
- A page without H4s appropriately scores zero — the absence of fourth-level headings is a legitimate structural choice, not a deficit.
- It counts occurrences rather than distinct terms, so repeating one related term across several H4s raises the number without broadening topical coverage.
- The source LSI list is SERP-dependent: the related terms RankGear looks for are derived from the live result set, so the target vocabulary shifts as the SERP shifts.
- Do not deepen the heading hierarchy for term placement — inventing H4s to hold related terms distorts your document structure for a number that only ever tracks it.
- This factor correlates with position within a measured result set; it is a prioritization signal, not a cause of ranking, and does not make a page rank on its own.
Related factors
Part of the Factors reference · how RankGear measures · glossary.