Page Entity Density is the number of recognized entity mentions in a page’s cleaned text divided by that page’s Clean Word Count. It shows how densely a page names real things — people, organizations, places, products, and concepts — relative to its overall word volume.
| Term | Page Entity Density |
|---|---|
| Category | Entities, Links and Authority |
| Also known as | Page-Level Entity Density |
| Where it appears | Workbook Entity Density |
What it means in RankGear
In the Entity Density workbook, RankGear takes a page’s cleaned text — the main content after navigation, headers, footers, and other boilerplate are stripped out — and counts every recognized entity mention within it. That count is then divided by the same page’s Clean Word Count to produce a single ratio. Because both the numerator and the denominator come from the same cleaned page, the figure is self-contained: it describes how concentrated the entity references are inside the text a reader actually sees.
How to interpret it
Read the value as a proportion, not a raw tally. A higher density means the page mentions recognized entities more frequently for its length; a lower density means the same word budget is spent on non-entity language. Two things shape whether the number is comparable: the entity-recognition provider that was enabled for the run, and the run conditions themselves. Providers use different indexes and recognize entities differently, so only compare Page Entity Density across pages measured under the same provider and the same run. Before reading into a value, confirm the provider was actually available for that page — a low density can reflect a provider that returned little, not a genuinely sparse page.
| Component | What it contributes |
|---|---|
| Recognized entity mentions (numerator) | Total entity references the enabled provider identified in the cleaned text |
| Clean Word Count (denominator) | Word count of the same page after boilerplate is removed |
Example
A 1,200-word cleaned product page has 84 recognized entity mentions, giving a Page Entity Density of 0.07. A competing page in the same workbook run comes in at 0.03 — roughly half as entity-dense for its length. In the Entity Density sheet you line these up side by side to see which pages in the comparison set lean more heavily on named entities versus general prose, then decide whether that pattern fits the topic.
Important considerations
- Provider metrics use different indexes and scales; they are comparative indicators for ranking pages against each other, not Google scores.
- Density depends on the enabled entity-recognition provider — separate provider availability from the measured value before drawing conclusions.
- The figure is sensitive to how cleaned text is extracted; different boilerplate stripping changes both the entity count and the Clean Word Count.
- A higher density correlates with entity-rich content but does not by itself make a page rank — treat it as evidence to interpret alongside the rest of the workbook, not a lever to maximize.
Related terms
Part of the RankGear glossary · how RankGear measures · the 870 factors.