How to Count Word Frequency (Overused Words and Keyword Density)
Editing an English draft often comes with a suspicion that some word is doing too much work, without being able to name it. Paste the passage, sort by count, and the repeated verbs and adjectives surface at once.
Open the tool：Word Frequency Counter
When word frequency tells you something
A paragraph that reads flat usually has one verb or connective repeating. Editing by feel misses occurrences; a frequency list puts the worst offenders at the top, which is where the rewriting should start.
Checking keyword density works the same way: paste the body copy of a landing page and see how often the target term appears. High-frequency words in subtitles, papers or support transcripts also reveal habits and recurring questions.
- Finding overused words in an English draft
- Estimating keyword density in landing copy
- Reviewing vocabulary in subtitles or papers
- Surfacing recurring terms in support transcripts
How counting works here
Text is split on whitespace and compared exactly: case matters and punctuation is not stripped. A capitalised word and its lowercase form get separate rows, and so does a word carrying a comma or quote. Mixed capitalisation therefore scatters what is conceptually one word across several entries.
Output is 'word: count', sorted from most to least frequent, one per line. When counts tie, order follows where the words first appeared. The result pastes straight into a spreadsheet for further filtering.
- Case-sensitive: capitals and lowercase count separately
- Punctuation is kept: dog. and dog are separate
- Sorted by count, formatted as word: count
- CJK text is treated as one word per sentence
What it does not compute
There is no lemmatisation, so run, runs and running occupy three rows and have to be merged by hand or with a stemming tool. There is no stop-word filtering either, so the, a and of sit at the top; delete those before counting if they get in the way.
Percentages are not included — density is the count for a term divided by total words, which you calculate yourself. There is also no universal target density, since it varies by content type, so treat it as a writing signal rather than a ranking switch.
- No lemmatisation: run, runs and running stay apart
- No stop-word filtering
- No percentages — compute density yourself
- No universal keyword density threshold
FAQ
Why are The and the counted separately?
Because counting is case-sensitive, so each gets its own row and its own total. Lowercase the whole text before counting if you want them combined; mixed capitalisation also hides how often a word really appears.
Does punctuation affect the count?
Yes. Punctuation is not stripped, so dog and dog. are different entries, and so is dog,. Remove punctuation or replace it with spaces before counting, otherwise the same word is split across several rows.
Can it count Chinese or Japanese words?
Not accurately. Splitting happens on spaces, and those languages do not use them, so a whole sentence counts as one word and you get an entry like the entire sentence with a count of 1. They need a tokeniser instead.
Related tool：Word Frequency Counter