What Is Keyword Density?
3 min read
Short answer
Check your own numbers
0 words · 0 characters
Keywords
0 words totalPaste some text above and the keyword table fills in.
The formula
density = (times the keyword appears ÷ total words) × 100A term used 15 times in a 1,500-word article has a density of 1%. That is it — there is no weighting, no adjustment for position, no accounting for synonyms.
For multi-word phrases, the denominator stays the total word count. So “word counter” appearing 10 times in 1,000 words is 1%, even though the phrase occupies 20 of those words. Every keyword tool works this way, which is worth knowing when a figure looks lower than expected.
What counts as a good keyword density?
There is no figure search engines reward. The honest ranges, from reading what actually ranks:
| Density | How it reads |
|---|---|
| 0% | The subject is never named — a real problem |
| 0.5–1% | Natural, typical of well-written prose |
| 1–2% | Focused, normal for a page about one thing |
| 2–3% | Noticeable repetition; worth reading aloud |
| Over 3% | Reads as stuffed |
The useful threshold is not a ranking penalty but a reading one. A term at 4% density is repeated once every 25 words, and readers notice long before an algorithm cares.
Why it stopped being a ranking factor
Early search engines counted terms, so people repeated terms. That arms race ended a long time ago.
Modern ranking uses semantic understanding: a page about “affordable laptops” is understood to be relevant to “cheap notebooks” without containing that phrase. Google’s own guidance has described keyword stuffing as a spam practice since 2011, and its documentation says plainly that repeating words to manipulate rankings harms the page.
There is no threshold at which stuffing is penalised, because it is not a threshold problem. Text written to hit a percentage reads badly, readers leave, and the behavioural signals do the rest.
What density is still good for
As an editing signal, it earns its place.
It tells you what your page is about. Paste a draft into the keyword density checker and read the top ten words. If they do not describe your subject, the page is unfocused — and that is a genuine ranking problem, just not the one people mean by density.
It finds the missing term. If you have written 1,500 words about speaking time and the phrase never appears in the top twenty, the article may be circling its subject without naming it. Readers searching for it will not find it either.
It finds the dominant term. One word appearing three times as often as anything else is almost always repetition you did not notice while writing.
Phrases reveal habits. The two- and three-word view is where the real findings are. “In order to”, “it is important to”, “one of the most” — these show up as your most common phrases and are almost always cuttable.
A workflow that works
- Write for the reader. Do not think about density at all.
- Paste the draft into the density checker.
- Read the one-word list. Does it describe the subject?
- Switch to phrases. Cut the repeated filler.
- Check the readability afterwards, since changing words changes how it reads.
Writing to a percentage produces worse text than writing well and then checking the percentage.
Frequently asked questions
Should my keyword be in the first 100 words? Usually it will be, if the page is genuinely about that subject and the opening states the subject. Doing it deliberately is harmless; contorting a sentence to manage it is not worth it.
Does the title tag count towards density? Not in this tool, which measures body text. Titles and headings matter to search engines, but as structure rather than as frequency.
What about LSI keywords? “LSI keywords” is not a technique search engines use — latent semantic indexing is a specific 1980s method that Google has said it does not use. Writing about related concepts because they belong in a complete answer is good practice; the label is marketing.