Keyword Density Checker
Count 1-3 word phrase frequencies and density percentages in pasted text, with stopword filtering.
Top keywords
Density = count divided by total tokens (latin words plus CJK characters). The table reports numbers, not targets.
How to use it
- Paste the text you want to inspect into the input box. Latin letters and digits form words; CJK characters and kana are counted one character at a time, because those languages have no spaces between words.
- Choose a phrase length: 1 word for single keywords, 2 or 3 words for multi-word phrases such as balcony garden or balcony garden plan.
- Leave the stopword box checked to hide common English words such as the, and or from the table. Uncheck it to see every key, including stopwords.
- Read the top keyword table: each row shows the count and the density percentage, where density equals the count divided by the total number of tokens, times 100.
- The copy button puts the top 50 keys with counts and densities into a plain-text report.
Common use cases
- Checking how often a target phrase appears before or after an edit, without guessing.
- Comparing two drafts of the same article by running the checker on each and reading the tables side by side.
- Spotting accidental repetition of a filler phrase that crept in during editing.
- Teaching keyword metrics with numbers students can recompute by hand.
Why it stays private
Tokenizing and counting happen in your browser with plain JavaScript. No text is uploaded, no analytics script watches the input, and clearing the box removes everything.
FAQ
Q: What density should I aim for? A: This tool reports counts and percentages only. Many guides cite a broad reference band around 0.5% to 2.5% for a primary term in natural writing, but there is no official target, and repeating a phrase to hit a number usually reads badly. Treat the table as a measurement, not a goal.
Q: How are CJK texts handled? A: Chinese, Japanese and Korean characters have no word boundaries, so each character is one token and phrases of length 2 or 3 are built from adjacent characters. That is the standard approach for these languages, though it is not a linguistic word segmentation.
Q: Does the stopword filter change the density percentage? A: No. The denominator always stays the total token count. The filter only hides single-word stopword rows from the table, so the remaining percentages still refer to the same whole text.
Q: Are numbers counted as words? A: Yes. Runs of digits count as tokens, so version numbers and years show up in the table like any other word.
Your data stays on this device
Keyword Density Checker runs 100% locally. Everything you paste or drop is processed by your own browser and never transmitted. Nothing is stored on our servers — close the tab and it is gone.
Nothing you type is sent to a server, logged, or used for analytics. The page works even with your network disconnected after first load.