Word frequency counter
See which words a text actually leans on, and how often.
Articles, pronouns, prepositions and other words that top every count are left out, so what is left describes this text. “The” and “the” count as one word here. A hyphenated word counts once, an apostrophe stays inside its word, and the result is word and count separated by a tab, ready to paste into a spreadsheet. Up to 200,000 characters.
Result
Start typing and the result appears here. No button needed.
Word frequency counter, without the extra steps.
Count how often each word appears in pasted text, ordered from most to least common. The words that top every count — the, of, and — are left out by default so what remains describes this text rather than English in general. The result is word and count separated by a tab, ready to paste into a spreadsheet.
How to use this tool
- 1Paste the text you want to analyse.
- 2Choose whether common words are skipped and whether capitals matter. The counts update as you type.
- 3Read the summary, then select Copy result to take the table with you.
When Word frequency counter is the right tool
- A draft feels repetitive and you want to see which word you have actually used eleven times.
- A page is being checked for how prominent its main term is, without guessing from memory.
- A pile of survey answers or support messages needs its recurring words surfaced before anyone reads all of them.
- A list of identifiers or tags needs a count per distinct value, where capitals matter.
Skipping the obvious words is what makes the list useful
In any English text the top ten words are the, of, and, to, a — in some order, in every document ever written. A frequency list that starts with them tells you about English rather than about your text. Those function words are left out by default and counted separately, so the first line of the result is a word you actually chose. The list is kept short on purpose: aggressive stop-word lists remove words like can or will that are doing real work in a specification or a contract.
What counts as one word
Words are found with the browser's Unicode segmentation rather than by splitting on spaces, so don't stays one word instead of becoming don and t, and text in a language written without spaces is still segmented into words. A hyphen between two words joins them, because well-known is one word to the person reading it. Numbers count as words. Capitals are folded together by default, and can be kept apart when the text is code, tags, or names where Bug and bug are not the same thing.
Counting is exact, not clever
Each distinct written form is its own entry: run, runs, and running are three words, as are a singular and its plural. Collapsing them needs a dictionary per language and silently merges words that differ in meaning — which makes a number look more authoritative while being less true. Ties are broken alphabetically so the order is stable, and the output is tab-separated, which is what makes it paste straight into two spreadsheet columns.
A few quick answers.
A short list of English function words — articles, pronouns, prepositions, common auxiliaries — that otherwise take every top position and say nothing about the text. The count of what was removed is shown, and Count every word puts them back. The list is deliberately short: longer stop-word lists start removing words that carry meaning in the document you are analysing.
No. Words are counted exactly as written, so plurals, tenses, and different forms of the same root are separate entries. Reducing words to their root needs a language-specific dictionary and quietly merges words that are genuinely different, which would make the numbers less trustworthy rather than more.
Copy result and paste: each line is the word, a tab, then the count, which spreadsheets split into two columns automatically. The order is most frequent first, with ties broken alphabetically, so the table is already sorted when it arrives.
The counting does: segmentation comes from the browser's Unicode rules, which handle scripts without spaces between words. The skipped-word list is English only, so for other languages choose Count every word — the function words will top the list, which is at least honest about what the text contains.
Because the word above it is more frequent. If the order looks wrong, check whether capitals are being counted separately: with that on, The and the are two entries and each is lower than their combined total. The summary states how many words were counted in total and how many distinct ones there are, which is the quickest way to spot it.
It gives you the raw counts that density is calculated from, and the total, so a percentage is one division away. It deliberately does not present a density figure or a target, because the right number depends on the text and no honest tool can tell you what it should be.