Keyword density: what the percentage tells you
WordCounter.Cloud shows a list of the most frequent words in your text, each with a count and a percentage. This is called keyword density. It helps you see which words dominate a text. This article explains exactly how the list is created and what you can and cannot learn from it.
How the list is calculated
- The whole text is converted to lower case. "Coffee" and "coffee" are the same word.
- Certain punctuation marks are removed: period, comma, slash, hash, exclamation mark, dollar, percent, caret, ampersand, asterisk, semicolon, colon, curly braces, equals sign, hyphen, underscore, backtick, tilde and round brackets.
- The text is split into words at white space.
- Words with fewer than three characters are left out.
- Words from a short list of very common words are left out, for example "the", "and", "that" and "have", and the German "der", "die", "das", "und", "ist" and "mit". The list contains 40 entries for English, German, Spanish and French.
- For each remaining word the tool counts how often it occurs. The percentage is that count divided by the number of remaining words.
- The ten most frequent words are shown.
An example
Take this text: "Our coffee is roasted fresh. Coffee lovers like our coffee because the coffee is fresh. Is coffee good? Yes, coffee!"
"is" is too short, and "the" is on the list of common words. 16 words remain. "coffee" occurs 6 times, which gives 37.5 %. "our" and "fresh" each have 2 occurrences, 12.5 %.
Note that the percentage refers to the filtered words, not to the total word count shown at the top. The normal word count of this text is 20.
Details you should know
- Some punctuation stays. Question marks, quotation marks and apostrophes are not removed. "good?" and "good" are therefore two different entries.
- Hyphens disappear. "e-mail" becomes "email" and is counted together with "email".
- No word forms are merged. "roast", "roasted" and "roasting" are three separate words.
- Phrases are not counted. The list only contains single words, not combinations such as "fresh coffee".
- Numbers count too. A year such as "2026" has four characters and appears in the list if it occurs often enough.
- The list of common words is short. Words like "our", "with" or "for" are not on it and can appear in the top 10.
What the percentage is useful for
- Spotting repetitions. If a word appears far more often than you thought, you can vary your wording.
- Checking focus. If your main topic does not appear in the list at all, the text may not be as focused on it as you intended.
- Comparing versions. After an edit you can see whether a word has become more or less frequent.
What it is not useful for
There is no "correct" keyword density that we could recommend, and WordCounter.Cloud does not recommend one either. A text written to hit a certain percentage usually reads unnaturally. Write for your readers first and use the list as a check for repetitions.
Try it
Paste your text into the WordCounter.Cloud editor and open the keyword density section. The list updates as you type, and the analysis runs in your browser. If your text has to stay within a length limit, the guide on character limits may help.