Count repeated words and their share of a passage. Spot overused terms and compare vocabulary before editing a draft.
How to use
Paste the passage and read the ranked rows as word, count, and percentage. Letter case is ignored, while each word occurrence contributes to the total.
Example
“Blue sky, blue sea” counts blue twice among four words, giving it a 50% share.
FAQ
Does it group forms of a word?
No stemming is applied. Write and writing remain separate. Treat the report as an editing aid, not a ranking score.
Related tools
A practical task
Read repeated words in their sentence context. A necessary technical term can repeat without making the article poor, while repeated vague phrases can make it harder to read. Use the ranked list to find candidates for review, not automatic deletion.
Another useful question
Can the report identify a repeated phrase?
It reports individual word frequency. Use the keyword density checker for a chosen phrase, or inspect the passage when repeated wording spans several words.
Details for interpreting the result
| Detail | Explanation |
|---|---|
| Words and tokens | A word count depends on token rules. Punctuation, apostrophes, digits, and languages without spaces create edge cases. Use the receiving editor's definition when a formal limit must match its own counter exactly. |
| Counts and percentages | A frequency count tells how often an item occurs. Its percentage depends on the chosen denominator: all words, all characters, or another set. Read the unit before comparing two frequency reports. |
| Case sensitivity | Case sensitive matching distinguishes capital and lowercase letters. Ignoring case merges some of those distinctions, but it does not turn different spellings or word forms into the same term. |
| Code points | A Unicode code point identifies a character or a component of one. A letter followed by a combining accent contains two code points even when the screen displays one accented letter. |
| Visible symbols | What looks like one symbol can contain several Unicode components. Emoji sequences can include a base symbol, a skin tone, and joining characters. Counting or splitting them by bytes changes their structure. |
| Similar looking text | Two strings can look identical but use different Unicode sequences. An accented letter can be stored directly or as a letter plus an accent. Visual similarity alone does not establish byte equality. |
| Words and meaning | Sharing words does not prove that two passages make the same claim. Negation, order, context, and different subjects can change meaning while vocabulary stays similar. Read the passages before making an editorial judgment. |