Word Frequency Counter

About this tool

A word frequency counter lists every word in a piece of text along with how many times it appears, sorted from most to least common. This one also counts two- and three-word phrases, can leave out common filler words, and exports the table as CSV. The counting happens in your browser, so the text you paste is never uploaded.

How to use the word frequency counter

1. Paste your text into the box. A short sample is loaded so you can see how the results look.
2. Pick what to count. Single words, 2-word phrases or 3-word phrases.
3. Set the filters. Ignore case (on by default) merges The and the. Exclude common stop words removes filler such as the, and and of. Min word length hides short tokens, and Show limits the table to the top 10, 25, 50, 100 or all entries.
4. Read the table, which updates as you type, or click Count frequency. Use Copy CSV or Download .csv to take the shown rows into a spreadsheet.

How the counting works

The text is split into words with a Unicode-aware pattern: a word is a run of letters, combining marks and digits, matched with Unicode property escapes such as \p{L} and \p{N}. That means accented words like café and text in other alphabets are counted as whole words rather than broken apart. The edge cases:

• Apostrophes inside a word are kept, so don't and writer's are single words. Curly apostrophes (’) are treated the same as straight ones, so text pasted from a word processor matches text typed in a code editor. Quote marks at the start or end of a word are dropped.
• Hyphens split words, so well-known counts as well and known. In phrase mode the two halves still form the phrase well known.
• Numbers count as words, so 2026 or 42 appear in the table. Raise the minimum word length to hide short numbers.
• Punctuation such as periods, commas, colons and dashes is not part of any word.

The percentage column is the count divided by the total number of words (or, in phrase mode, the total number of phrases of that length), so it is the same figure people call keyword density. Total words counts every word before any filter, and unique words is the number of distinct words with the current case setting.

Stop words

Stop words are the very common function words, such as articles, pronouns, prepositions and forms of to be, that dominate any frequency list without saying much about the subject. This tool uses a short built-in list of 126 common English stop words. In single-word mode those words are hidden. In phrase mode a phrase is hidden when it starts or ends with a stop word, so of the and the story disappear while a phrase like first read stays. The totals at the top never change with this option.

N-grams: counting phrases

An n-gram is a sequence of n words that appear next to each other. Two-word phrases are bigrams and three-word phrases are trigrams. Counting them shows repeated phrasing that a single-word list hides, such as a sentence opener you keep reusing. Phrases are built only from words in the same stretch of text; they do not run across sentence punctuation such as periods, semicolons or colons.

Worked example

The sample text has 50 words and 38 unique words. With the default settings, the top of the table is the with 7 (14.00%), followed by a, read, is, for, words and which with 2 each (4.00%). Ties are listed in the order the words first appear. Turn off Ignore case and the drops to 6, because the capitalized The becomes a separate row with 1.

Tick Exclude common stop words and the list starts with read and words (2 each, still 4.00% of all 50 words), then good, editor and reads. Switch to 2-word phrases with stop words included and the sample yields 42 phrases. read is, is for and for the each appear twice (4.76%).

Common mistakes

• Reading too much into percentages from a short text. In a 50-word sample, a single extra use moves a word by 2 percentage points.
• Forgetting that Ignore case is on, then searching the table for a capitalized name.
• Comparing two drafts with different settings. Use the same phrase length and filters for both.
• Treating stop-word removal as neutral. It hides words like not and no, which can matter for meaning.

Use cases

• SEO keyword check. See how often a target term and its variants actually appear on a page, and spot accidental repetition. Search engines do not publish a target density, so use the numbers to judge readability rather than to hit a quota.
• Editing. Find overused words such as really or just, and repeated phrases, before a piece goes out.
• Studying vocabulary. List the most common words in a text you are reading in another language.
• Comparing drafts. Run two versions and see which terms gained or lost weight.

For overall counts of words, characters and sentences, use the Word Counter. If pasted text is full of stray line breaks or extra spaces, run it through the Text Cleaner first.