Contextual Density Unique Terms is a topical-breadth rate that counts each distinct variant, entity, or related term only once, then divides that count by the number of meaningful words on the page. It measures how wide a page’s vocabulary is, not how often any single phrase repeats.
| Term | Contextual Density Unique Terms |
|---|---|
| Category | Content and Topical Relevance |
| Also known as | Contextual Density (Unique Terms), Unique-Term Contextual Density |
| Where it appears | BM25 Drafts, Density, and factor RG-KWD-100 |
What it means in RankGear
RankGear surfaces this metric in the BM25 Drafts and Density views and as factor RG-KWD-100. Unlike a raw density figure, which can climb simply because one keyword is repeated, the unique-term reading treats each distinct variant, named entity, or related concept as a single hit no matter how many times it recurs. That count becomes the numerator; the meaningful-word total for the page is the denominator. The result tells you how much of the topic’s vocabulary a page actually touches, giving you a breadth signal that sits alongside the repetition-based density numbers rather than duplicating them.
How to interpret it
Read this as a coverage indicator, not a target to chase. A low unique-term density usually means a page leans on the same few phrases while skipping the entities, questions, and supporting concepts that competitors for the same query tend to include. Compare the figure against the pages actually ranking for your term rather than against an absolute number, and close gaps by adding genuinely relevant information — the missing subtopic, the related entity, the question a reader would ask next — not by sprinkling synonyms to move the ratio.
| What is counted | How it is counted |
|---|---|
| Distinct variants, entities, and related terms | Once each, regardless of how often they repeat (the numerator) |
| Meaningful words on the page | The denominator the unique count is divided by |
Example
Two pages target “cold brew coffee.” Both mention the exact phrase a dozen times, so their raw density looks similar. But the first page also covers steeping time, coarse grind, dilution ratio, nitro, and caffeine content, while the second repeats “cold brew” and little else. The first page carries a higher Contextual Density Unique Terms reading because it touches more of the topic’s distinct vocabulary, which is the gap RG-KWD-100 is designed to expose.
Important considerations
- This is a comparative indicator of topical breadth, not a Google score. A wider vocabulary correlates with pages that cover a topic thoroughly; it does not by itself cause a page to rank.
- Added terms should preserve search intent. Padding a page with loosely related wording, keyword stuffing, or unsupported claims inflates the count without building real topical authority.
- Because each term is counted once, the metric rewards coverage over repetition — pair it with the repetition-based density readings rather than reading either in isolation.
- Provider and draft metrics are scaled for comparison within RankGear; judge a page against its actual competitors for the query, not against a fixed threshold.
Related terms
Part of the RankGear glossary · how RankGear measures · the 870 factors.