Count the words you repeat
Paste your text, upload a Word, PDF or text file, or load the example. Single words and two to five-word phrases, with counts, density and the sentences each one appears in.
Paste or type and every word is counted and ranked, along with every two, three, four and five-word phrase. You get the count, the density, the whole stop word list where you can change it, what share of the text your top ten words cover — and, when a word looks overused, the sentences it appears in so you can decide for yourself. Free, no sign-up, nothing leaves your browser.
Paste, read, drill in.
Type, paste, or drop a .docx, .pdf or .txt file onto the editor. Files are read in your browser; nothing is uploaded.
Word or phrase length, common words hidden or not, case sensitivity, numbers in or out, and a minimum count. Your settings are remembered.
Every sentence containing it is collected on the In context tab. A list tells you a word is frequent; the sentences tell you whether that is a problem.
One page instead of a count and a shrug.
Single words and two, three, four and five-word phrases, each with its count and its share of the text. Phrases that start or end on a common word are dropped, which is what makes the list readable.
All 172 words, shown in full. Click any one to put it back in the count, or add your own — your product name, a client’s name, a word you know you overuse.
Click any entry and every sentence containing it is collected in one place, with the match highlighted. Click a sentence to jump to it in your text.
Unique words, type-token ratio, how many words you used exactly once, and what share of the whole text your top ten words account for.
The distribution of word lengths across the text. English prose peaks at three letters; a curve peaking at six usually means a lot of abstract nouns.
Every count is arithmetic done in your browser. No server call carries your text, which is also why there is no limit and no charge.
A real run, recorded on 19 September 2026, on the text behind the “Try an example” button — 170 words about keyword density, which repeats itself on purpose.
density — 4 times, 4.60%. advice — 3 times, 3.45%. keyword — 3 times, 3.45%. Then appears, came, cent, count and engines at two each.
Two-word phrases: keyword density and search engines, twice each. Three-word: advice about keyword, twice.
Every frequency tool hides common words. Almost none will tell you which ones, and that single omission makes every percentage they print unverifiable.
The number is easy to compute and easy to misuse, so here is the part that matters.
Raw term frequency mattered to search engines around 2003. Writing to a density target has been useless for roughly twenty years, and the one-to-two-per-cent figure still being quoted has no source anyone can name.
Thirty-one uses of one phrase in a two-thousand-word page is worth knowing — not because of the percentage, but because a reader will notice and will find the page exhausting.
Density is a symptom; repetition is the disease. Click the word, read the ten sentences it lives in, and it becomes obvious whether it is your subject or your tic.
Four jobs, one list.
Finding the word you have leaned on twelve times without noticing. The In context tab is the part that actually changes the draft: seeing the sentences side by side makes the fix obvious.
Checking a page is not repeating itself, and that the phrases it does repeat are the ones it is about. Export the CSV, keep it with the brief, compare against the next draft.
Basic text analysis for an essay or a corpus exercise — frequency, type-token ratio, hapax counts — without installing anything or learning a package.
The word counter has a keywords tab. This is what it is not.
The complete list rather than the top 25. Four and five-word phrases. Sorting, filtering, a minimum count, case sensitivity. The stop word list itself, editable. Vocabulary statistics. And every sentence a word appears in.
Words, characters, reading time, six readability formulas and 35 platform limits, with a quick top-25 keyword view alongside. It is the right page when frequency is one thing you want to glance at rather than the thing you came for.
What people ask about counting word frequency online.
A tool that counts how many times each word appears in a text and ranks them. This one also counts phrases of two, three, four and five words, gives each entry its share of the text, and — the part most tools stop short of — shows you every sentence a given word appears in, so you can judge whether the repetition is a problem or the subject.
Free, with no sign-up and nothing metered. The counting happens in your browser, so the only limit is what your device can hold; documents of a hundred thousand words are comfortable.
Stop words are the common words — the, and, of, is — that dominate any frequency list and tell you nothing. Nearly every tool hides some set of them and does not say which, which makes their percentages impossible to check. Ours is the 172 words on the Stop words tab, in full, and you can switch any of them back on or add your own.
Each entry as a share of the counted items — the words or phrases that survived your current settings. That means the figure changes when you hide common words, which is correct: the denominator changed. The download gives you the same number to four decimal places for the whole list.
No, and it has not been one for roughly twenty years. Treat a frequency list as a repetition check on your own writing, not as a target. If a phrase appears thirty times in two thousand words, the problem is that a reader will notice, not that the percentage is above some number.
There is no good number, and anyone quoting one — usually one to two per cent — cannot tell you where it came from. The useful question is whether a reader would notice the repetition. Click the word in the list, read the sentences it appears in, and you will know.
Unique words divided by total words: how much of your vocabulary is new. It falls as a text gets longer, because every writer runs out of new words eventually, so it is only meaningful when comparing texts of similar length. Use it to compare two drafts of the same piece, not two different pieces.
Words that appear exactly once. In ordinary English prose they are usually around 40 to 60 per cent of the distinct words in a text. A sharp drop below that range generally means the writing is circling a small vocabulary.
Two, three, four and five. A phrase is dropped when it starts or ends on a stop word, because "of the" is not a phrase anyone is looking for while "rate of return" is. That single rule is what makes a phrase list readable.
Yes. The CSV button downloads the complete list — every phrase, its count and its density — at the current settings, not just the rows visible on screen. It opens directly in Excel, Numbers or Google Sheets.
Not by default: Apple and apple are counted together. Turn on "Case sensitive" and they separate, which is what you want when the text is about a company rather than a fruit.
No. Every count runs in your browser and there is no server call carrying your text. Files you drop in are parsed locally. Your draft and your stop word changes are kept in your own browser storage.
The word counter shows the top 25 entries for one, two and three-word phrases with the stop word list hidden behind a checkbox. This page gives you the whole list with four and five-word phrases, a minimum-count filter, sorting, case sensitivity, the stop word list itself, vocabulary statistics, and every sentence a word appears in.
Counting shows the repetition. These help you fix it.
This page is free and unmetered because it runs in your browser. Plan allowances apply only to the AI tools, and they are on the pricing page.
Upgrade
Drop your file in