The first time I ran one of my own blog posts through a word frequency counter, I felt a little exposed. I had used the word "basically" fourteen times in a 1,200 word article. Not "occasionally" fourteen times. Fourteen. My eyes had glossed right over every one of them while writing, because that is exactly what crutch words do: they hide in plain sight. A frequency count does not care about your intentions. It just counts, and the number sits there on the screen daring you to argue with it.
I build tools at toolz.dev, and I spend most of my week writing: documentation for WordPress plugins, marketing copy, Laravel tutorials, release notes. Word frequency analysis has quietly become one of the checks I run before anything ships. It is a thirty-second pass that catches repetition my brain refuses to see, confirms that a page actually mentions the topic it claims to be about, and occasionally saves me from publishing something that reads like it was assembled from three different drafts. This guide is everything I have learned about counting words well.
TL;DR: A word frequency counter splits your text into words and reports how many times each one appears, ranked from most to least common. Use it to catch overused crutch words, check keyword density for SEO, and understand what a document actually emphasizes. Fold case so "The" and "the" count together, turn on the stop-word filter to hide function words like "the" and "and," and read the percentage column to see each word's share of the text. The Word Frequency Counter runs entirely in your browser, so nothing you paste is uploaded.
What is a word frequency counter?
A word frequency counter reads a block of text, breaks it into individual words - a step Unicode Standard Annex #29 specifies far more carefully than "split on spaces" suggests - and tells you how many times each distinct word appears. The output is a ranked table: every unique word, its count, and usually its percentage of the whole document. The word that shows up most often sits at the top.
This is a different question from the one a plain word counter answers. A word counter tells you that a document has 847 words. A frequency counter tells you that 31 of those words are "the," 12 are "data," and 9 are "however." The total is one number about length. The frequency table is a distribution about emphasis. Length tells you whether you hit an assignment target. Distribution tells you what the writing is actually about and where it is repeating itself.
The concept is old and well studied. Linguists have measured word frequencies for more than a century, and the pattern that emerges is remarkably consistent across languages: a small number of words account for a huge share of any text, and frequency drops off sharply after that. This is Zipf's law, the observation that the most common word tends to appear about twice as often as the second most common, three times as often as the third, and so on. You will see this shape in your own writing the moment you sort a document by frequency. The top of the list is dominated by short function words, and the interesting content words trail behind them.
Why would I count word frequency?
I reach for frequency analysis for four distinct reasons, and it helps to know which one you are chasing before you read the table.
The first is catching repetition. Every writer has crutch words, the ones they lean on without noticing. Mine are "basically," "actually," and "just." Yours might be "really," "very," or a particular technical term you have fallen in love with. Repetition is invisible from the inside because you wrote each instance in a separate moment and never saw them stacked up. A frequency count stacks them up.
The second is keyword density for search. When I write a page targeting a specific phrase, I want to confirm the phrase and its close variants appear often enough that a search engine understands the topic, without appearing so often that the page reads as stuffed. The percentage column turns this from a vague worry into a number I can check.
The third is understanding an unfamiliar document. Paste in a long report, a transcript, or a competitor's article, sort by frequency, filter out the function words, and the top of the list is a fair summary of the subject in about ten seconds. It is a crude summary, but it is fast and honest.
The fourth is editing for variety. If "important" appears eleven times, some of those are candidates for a synonym or for deletion. The count does not tell you which ones to change, but it tells you to look.
How do I use the Word Frequency Counter?
The workflow is short. Paste your text into the input box on the Word Frequency Counter, and the ranked table appears immediately. There is no upload step and no button to wait on for the basic count.
From there, the options shape what you see. Decide whether case matters. For most writing you want case folding on, the default, so that "Data," "data," and "DATA" count as one word rather than three. Turn on the stop-word filter when you want the ranking to surface content words instead of grammar. Set a minimum word length to skip very short tokens. And if the tail of the list is long and uninteresting, set a top-N limit to trim it down to the words worth reading.
When you have the view you want, copy the table. It copies as tab-separated rows, which means it pastes cleanly into a spreadsheet column or a document table without any reformatting. I keep a small spreadsheet of crutch-word counts across articles so I can see whether I am improving over time, and the tab-separated export drops straight into it.
What are stop words, and should I remove them?
Stop words are the high-frequency function words that hold sentences together but carry almost no topical meaning on their own: "the," "a," "and," "of," "to," "is," "in," and their many companions. In almost every English document, these words dominate the top of a raw frequency list. If you count a thousand-word article without filtering, the first six or seven rows will be stop words, and the words the article is actually about will be buried below them.
Whether to remove them depends entirely on your goal. Turn the filter on when you are trying to see what a document is about, because stripping the grammar leaves the content words standing. Leave it off when you need an exact, unfiltered count of every token, for example when you are studying sentence structure, measuring true length, or checking how often you overuse a specific connective like "however" or "moreover."
There is no single official list of English stop words. Different tools and libraries ship slightly different sets, and the choice is a judgment call rather than a standard. This tool uses a compact, widely used list of common English function words. If a word you consider meaningful is being filtered, that is a sign your topic overlaps with the function-word set, and you should turn the filter off for that particular count.
How are contractions, hyphens, and other languages handled?
Word splitting sounds trivial until you look closely, and then it becomes a nest of edge cases. What counts as one word?
This tool splits on Unicode letters and digits, and it keeps an apostrophe or hyphen when it sits between two letter or number runs. That single rule handles the cases that matter most. The contraction "don't" stays a single word instead of splitting into "don" and "t." The compound "state-of-the-art" stays intact instead of shattering into four fragments at every hyphen. But a stray hyphen used as a dash, or a trailing apostrophe used as a quotation mark, does not get glued onto a neighbouring word, because there is no letter on the other side of it.
Because the matching is Unicode-aware, accented and non-Latin scripts work too. "Café" counts as one word with its accent intact, and text in languages that use non-Latin letters is split on the same letter-and-digit rule rather than being dropped. Numbers are treated as words, so "2026" and "42" appear in the table. If you are counting prose where numerals are noise, the minimum-length option and a quick scan will let you ignore them.
What is keyword density, and how does this tool show it?
Keyword density is a word's share of the total word count, written as a percentage. If a 500-word page uses your target keyword 10 times, its density is 2 percent. The frequency counter shows this in the percentage column next to each count, so you never have to do the division yourself.
Here is the honest version of the SEO advice: there is no magic density number that ranks a page, and there never was. Search engines stopped rewarding raw keyword repetition many years ago, and Google's own guidance warns specifically against keyword stuffing, the practice of cramming a term onto a page to manipulate rankings. What density is genuinely useful for is the opposite check. It confirms that a page mentions its topic at all, and it flags the pages where you have accidentally repeated a phrase so many times that a human reader would notice and a search engine might treat it as spam. I use it as a smoke detector, not as a target.
The table below shows how I read the different signals a frequency count gives, depending on what I am trying to learn.
| Goal | What to look at | Setting that helps | What a bad result looks like |
|---|---|---|---|
| Catch crutch words | Content words with surprisingly high counts | Case folding on, stop words off | One informal word appears 10+ times in a short piece |
| Check keyword density | Percentage column for your target phrase | Case folding on | Target under 0.5 percent, or a single word over 4 to 5 percent |
| Summarize a document | Top 10 to 15 rows | Stop words on, min length 3 | Nothing, this is what the filter is for |
| Edit for variety | Any content word with a high count | Stop words on | The same adjective repeated across every paragraph |
| Measure true length | Total words figure | All filters off | A count far from your target |
Word frequency counter versus a plain word counter
These two tools get confused constantly because their names overlap, so it is worth stating the difference plainly. A word counter answers "how much?" It returns totals: words, characters, sentences, sometimes reading time. It is the right tool when you have a length target, a tweet limit, or an essay minimum. A word frequency counter answers "which words, and how often?" It returns a distribution, and it is the right tool when you care about emphasis, repetition, or density rather than length.
They pair well. I usually run a draft through the word counter to confirm length, then through the frequency counter to check that the length is not padded with the same three words. If you also work with lists rather than prose, the line sorter will deduplicate and order lines, and the text diff checker will show you exactly what changed between two versions of a document. And if the goal of your edit is to vary word choice, the case converter handles the mechanical part of reshaping headings and identifiers once you have decided what to change.
Is my text private?
Yes. The entire analysis happens in your browser using JavaScript. The text you paste is never sent to a server, never logged, and never stored. This matters more than it sounds, because the documents most worth analyzing are often the ones you least want to upload: an unpublished manuscript, a client's confidential report, an internal memo, a draft under embargo.
You do not have to take my word for it. Open your browser's network tab and watch it while you count, and you will see that no request goes out. Or disconnect from the internet entirely and keep working, because once the page has loaded, the tool needs no connection at all. This browser-first approach is the same principle behind everything I build, and if the topic interests you, I wrote more about it in the guide to data privacy with online tools. For a broader picture of where a frequency counter fits into a writing and editing workflow, the developer productivity tools guide covers the tools I reach for most.
Frequently asked questions
What is word frequency?
Word frequency is the number of times each distinct word appears in a piece of text. A word frequency counter reports it as a ranked table, listing every unique word with its count and often its percentage of the whole, so the most-used words appear at the top.
How do I count how often a word appears?
Paste your text into the tool and read the ranked table, which lists each word with its count. To focus on one word, sort alphabetically or scan the table for it. Every occurrence is counted, and case folding is on by default so capitalised and lowercase forms count together.
What are stop words and should I remove them?
Stop words are high-frequency function words such as the, and, of, to, and is that carry little topical meaning. Removing them is optional. Turn the filter on when you want the ranking to surface the content words a document is about; leave it off when you need an exact, unfiltered count of every word.
Does the counter treat uppercase and lowercase as the same word?
By default, yes. Case folding is on, so The, the, and THE are counted as one word. Turn on case-sensitive mode when capitalisation matters, for example to count a proper noun separately from the same letters used as an ordinary word.
How are contractions and hyphenated words handled?
Words are split on Unicode letters and digits, but an apostrophe or hyphen between two letter or number runs is kept. So don't counts as one word rather than two, and state-of-the-art stays a single token instead of being broken at every hyphen.
What is keyword density and how does this tool show it?
Keyword density is a word's share of the total word count, expressed as a percentage. The tool shows it in the percentage column next to each count, so you can check whether a target phrase appears often enough to be relevant without being repeated to the point of keyword stuffing.
Is there a limit on how much text I can analyse?
There is no fixed size limit. Because the analysis runs in your browser rather than on a server, you are bound only by your device memory, so full articles and book-length manuscripts are handled without an upload cap.
Is my text uploaded anywhere?
No. The text is analysed by JavaScript running in your browser and is never transmitted, logged, or stored. You can confirm this by watching the network tab while you count, or by disconnecting from the internet, the tool keeps working offline.
Written by Liton, builder of toolz.dev. I make browser-based developer tools and write about the craft of building them.
