Keyword Density Checker
Analyse term frequency and spot over-optimisation.
| Phrase | Count | Density |
|---|---|---|
| running | 7 | 7.53% |
| shoes | 7 | 7.53% |
| cause | 2 | 2.15% |
| matters | 2 | 2.15% |
| most | 2 | 2.15% |
| too | 2 | 2.15% |
| add | 1 | 1.08% |
| black | 1 | 1.08% |
| blisters | 1 | 1.08% |
| built | 1 | 1.08% |
| buy | 1 | 1.08% |
| choosing | 1 | 1.08% |
| consider | 1 | 1.08% |
| cushions | 1 | 1.08% |
| day | 1 | 1.08% |
| distances | 1 | 1.08% |
| end | 1 | 1.08% |
| feet | 1 | 1.08% |
| fit | 1 | 1.08% |
| gait | 1 | 1.08% |
| Phrase | Count | Density |
|---|---|---|
| running shoes | 6 | 6.52% |
| shoes that | 2 | 2.17% |
| Phrase | Count | Density |
|---|---|---|
| shoes that are | 2 | 2.20% |
“running”, “shoes” appear unusually often. Read the copy aloud — if it sounds repetitive, it will read that way to visitors too.
About this tool
Paste your copy to see which terms actually dominate it, including two and three word phrases. Common stop words are filtered so the list reflects real subject matter, and anything repeated to an unnatural degree is flagged.
How to use it
- Paste the page copy you want to analyse.
- Review single words alongside two and three word phrases.
- Check nothing is repeated so often that it reads as stuffing.
There is no correct density
Keyword density was a real ranking factor around 2005, when search engines largely counted term frequency. It has not been one for a long time, and chasing a target percentage now produces worse copy and no benefit.
Modern ranking uses semantic analysis. Engines recognise that "running shoes", "trainers", "footwear for runners" and "sneakers" are related concepts, and they evaluate whether a page comprehensively covers a topic — not whether it repeats a phrase a set number of times.
So what is this tool for? Diagnosis, not optimisation.
It is genuinely useful for:
- Catching accidental repetition. Writers fall into phrases without noticing. Seeing a term at 6% is a prompt to reread.
- Checking you are on topic at all. If the top terms are not what the page is about, that is worth knowing.
- Spotting AI-generated repetition. Generated copy often loops the same phrase far more than a human would.
- Auditing inherited content. Pages written to a 2012 SEO brief are often visibly stuffed.
Read the list, then read the copy aloud. Your ear is the better instrument.
What phrases tell you that single words do not
Single-word counts are noisy. "Running" appearing 15 times could be a page about running shoes, running a business, or running a marathon.
Two- and three-word phrases carry the actual topic, which is why they are shown separately. They also map better to how people search — the majority of queries are multi-word, and long-tail phrases convert far better than head terms.
Stop words are filtered by default so the list reflects subject matter rather than grammar. The filter is applied differently by phrase length: single stop words are dropped entirely, but a phrase is only dropped if every word in it is a stop word — otherwise "state of the art" and "out of stock" would disappear, and those are exactly the phrases worth seeing.
What to do with the output. If your target phrase appears once in 800 words, the page probably is not about what you think it is. If it appears thirty times, it reads like spam. Somewhere in between, stop measuring and go improve the substance.
Frequently asked questions
- What density should I target?
- There is no magic number, and chasing one is a mistake. Modern search engines use semantic analysis, not term counting. Write naturally; use this to catch accidental over-repetition, not to hit a quota.
- What counts as keyword stuffing?
- Repetition that stops the copy reading naturally — the same phrase forced in several times per paragraph. If it sounds awkward when read aloud, it is stuffing, whatever the percentage says.
- Is my unpublished copy safe here?
- Yes. The analysis runs in your browser, so draft articles and client copy are never uploaded or added to any index.