Keyword Density Checker
| Density | Reading |
|---|---|
| 0% | Keyword absent - page cannot rank for it |
| 0.5-2.5% | Natural zone - the term appears where it belongs |
| 2.5-4% | Watch zone - fine once, suspicious when systematic |
| 4%+ | Stuffing territory - modern engines demote, readers bounce |
Keyword density is the percentage of a text made up of one keyword or phrase: hits × words-per-keyword ÷ total words × 100. The checker above counts phrase occurrences exactly - case-insensitive, multi-word aware - and reads the result against the honest bands: 0.5 to 2.5 percent is where natural writing lands, and 4 percent-plus is stuffing territory.
Bottom line: density is a diagnostic, not a target. The classic advice to hit 2-3 percent dates from era-of-dumb-counter search engines; modern ranking uses embeddings and meaning, so a page cannot rank purely by repetition - but it can absolutely fail by it. The one failing score that matters is 0%: a page that never names its subject cannot be matched to it, by any engine.
The honest part: the metric's surviving uses are editorial, not manipulative. It catches accidental repetition (the same phrase thirty times reads badly to humans too), verifies a draft actually mentions its topic before publication, and serves the newest reader of all - citation engines, which extract answers from text that names its subject clearly and answers questions directly. Write for those, and density takes care of itself.
How to use
- Paste the text and type the keyword or phrase - matching is case-insensitive and multi-word aware.
- Read the density with the bands: 0% means the page cannot rank for the term; 4%+ means rewrite for humans.
- Fix by editing, not by inserting: where the term belongs naturally (the intro, a heading, one body mention), it usually already earns its percentage.
Frequently asked questions
What is a good keyword density percentage?
There is no target number worth optimizing for - but there are useful bands. Zero percent means the page never names its subject and cannot be matched to it. Half a percent to about two and a half percent is where natural writing lands when the topic is genuinely covered. Above four percent reads as stuffing to both readers and ranking systems. Treat the number the way a thermometer works: it tells you something is off, not what to write.
How does the phrase matching count multi-word keywords?
The checker lowercases everything, splits the text into words, and slides a phrase-sized window across them - so "rem conversion" matches only when those two words appear adjacent and in order. Partial overlaps inside longer words do not count, and punctuation between words breaks a phrase match. That makes the count slightly conservative: headings, lists and punctuation split what a human would read as one mention.
Does keyword density still affect SEO?
Not the way it did. Matching a query still requires the words to appear - zero mentions cannot rank - but repetition beyond natural usage adds nothing and can trigger quality demotion. Modern systems understand synonyms, morphology and topical coverage, so a page about rem conversion ranks for "convert pixels to rem" without ever containing that exact string. The metric survives as an editorial check: it catches accidental repetition and missing subject naming, which are real problems.
Why does my count differ from another tool's?
Definitions differ on three axes: what counts as a word (hyphenated compounds, numbers, apostrophes), whether the keyword counts by occurrence or by total keyword words, and case handling. This checker lowercases, splits on non-alphanumerics, counts each occurrence as one hit times the phrase's word count, and divides by total words. Any two tools will agree within rounding once you match those definitions - a mismatch bigger than rounding means the definitions differ, not that one is broken.