Word Frequency Counter

Paste text and get a ranked word and phrase frequency table you can export.

Free word frequency counter. Paste any text to see a ranked table of how often each word or phrase appears, with a stopword toggle, n-gram phrase mode and one-click CSV export. Runs entirely in your browser — nothing is uploaded. It runs free in your browser on Gera Tools, with nothing uploaded.

Last updated Source: Gera Tools

What is a word frequency counter?

A word frequency counter is a tool that reads a block of text and counts how many times each distinct word appears, then ranks the words from most to least common. It turns raw text into a frequency table so you can instantly see which terms dominate a document, which is useful for editing, SEO keyword research, linguistics and readability analysis.

A word frequency counter reads any block of text and shows you a ranked table of how often each word — or each multi-word phrase — appears. Paste an article, an essay, a meeting transcript, a pile of customer reviews or a chapter of a book, and within a fraction of a second you get every distinct term sorted from most common to least common, complete with raw counts, percentages and a visual frequency bar. It is built for writers tightening up repetitive prose, SEO specialists checking keyword density, students and researchers doing basic text analysis, and anyone curious about the shape of a piece of writing.

Unlike a plain word counter that only tells you the total, this tool answers the more useful question: which words actually carry the document? The moment you strip out the grammatical glue with the stopword toggle, the words that define your subject rise to the top of the list.

Tokenisation and counting rules

The counter tokenises your text with a Unicode-aware rule, so accented letters and non-Latin scripts are handled correctly and intra-word apostrophes (as in don't or it's) stay intact. You control how aggressive the cleanup is:

  • Ignore case folds The and the into one entry.
  • Strip punctuation keeps only letters and numbers, so dog. and dog count together.
  • Remove stopwords drops a built-in list of roughly 180 common English function words, and you can add your own domain-specific words (brand names, filler) to the list.
  • Min length, min count and show top N trim short tokens, rare terms and long tails.

In single-word mode each token is counted directly. In phrase mode the tool slides an n-word window across the cleaned token stream and counts every overlapping sequence, revealing repeated bigrams and trigrams. Alongside the table it computes live document statistics — total words, unique words, characters with and without spaces, sentence and line counts, average word length and lexical density (unique words ÷ total words, a rough richness measure).

Example

Paste a product description that repeats the phrase easy to use and turn on phrase mode with a length of 3. The trigram easy to use jumps to the top of the table with its exact count, while single-word mode would have scattered those signals across easy, to and use. Switch back to words, enable stopword removal, and the genuinely meaningful terms — the product’s features and benefits — surface immediately so you can judge whether your copy is on-message or padded.

SettingWhat you see
Words, stopwords offthe, and, to lead the list
Words, stopwords ontopical nouns and verbs lead instead
Phrases, length 2most-repeated two-word expressions
Phrases, length 3recurring key phrases and slogans

Every count, percentage and CSV row is generated in your browser — no text is ever uploaded or stored on a server.

Zipf’s law: what a “normal” frequency table looks like

Natural language has a signature shape: the most frequent word appears roughly twice as often as the second, three times the third, and so on — a power-law distribution known as Zipf’s law that holds across languages and text sizes. Practical use: a frequency table that deviates from this shape tells you something. A top word repeated far beyond its expected share signals keyword stuffing, boilerplate, or a template; a suspiciously flat distribution suggests generated or concatenated text. For SEO specifically, there is no magic keyword density to aim for — Google’s spam policies explicitly treat unnatural keyword repetition as stuffing, so the honest use of this tool is the reverse: catching yourself over-repeating a term and reducing it to natural frequency.

Getting cleaner counts from real documents

Function words (the, of, and) dominate every English frequency table and carry no signal for most analyses — filter them mentally or focus on the tail below the top dozen entries. Case sensitivity matters for distinguishing proper nouns (Turkey/turkey); punctuation handling matters for hyphenated terms and contractions. For comparative work — two drafts, two authors, two speeches — normalise by total word count and compare rates rather than raw counts, since document length swamps everything else.