Word Frequency Counter

How to count word frequency in text

1. Paste the source text

Drop a paragraph, blog post, manuscript, transcript, or full document. Counting starts on the first keystroke. The tool tokenizes on word boundaries, normalizes case (unless the case-sensitive toggle is on), and ranks every unique word by occurrence count. See also: Duplicate Word Counter. See also: Adjective Counter.

2. Adjust the filters

Four filters shape the count:

• Exclude stop-words (default on) removes the most common English function words (the, of, and, a, to, in, is, you, that, it). These dominate any English text and crowd out content words from the ranking.
• Case sensitive (default off) treats "The" and "the" as separate words when on. Off is the right choice for content analysis; on is right for code or proper-noun detection.
• Min word length excludes short words. Set to 3 to skip "a, I, of, to" and surface meaningful words faster.
• Min count hides one-off words. Set to 2 to see only words that appear at least twice.

3. Read the type-token ratio

Type-token ratio (TTR) is unique words divided by total words. Higher TTR means more vocabulary variety. Press releases and technical writing often run TTR 0.50-0.65. Casual conversation runs 0.35-0.45. A 1,000-word essay with TTR under 0.30 has heavy repetition and is a candidate for editing.

4. Export to CSV

The Download CSV button writes the full ranked list to a file. Useful for offline analysis, spreadsheet visualization, or feeding into downstream NLP tooling.

When you need word frequency analysis

SEO content optimization

Search engines do not use raw word frequency anymore (TF-IDF gave way to semantic models), but word frequency still indicates topical focus. A blog post targeting "kubernetes deployment" should have those two words in the top 10 frequency ranking. A post where they are not in the top 30 has buried the topic.

Writer's-style audit

Every writer has a verbal tic: words they overuse without noticing. Stephen King's "smile" appears 1,200 times across his work. Run a frequency count on your draft and the top non-stopword candidates are usually the same 3-5 words. Replace half of them and the prose tightens.

Academic literature review

Researchers paste paper abstracts into a frequency counter to find the dominant terminology in a field. The top-25 ranking is a quick literature thematic map. Pair with the Remove Duplicate Words tool to extract the unique vocabulary as a glossary.

Speech and transcript analysis

Politicians and CEOs leave verbal fingerprints. The word "absolutely" in a 30-minute speech can appear 40+ times. Run a frequency count on the transcript and the top non-stopwords are the speaker's anchor terms.

Translation quality check

Run frequency on the source and target language versions of a document. Even with different word counts (English is denser than Spanish), the rank order of content words should match. A mistranslation often shows up as a content word in the source that has no high-frequency match in the target.

Type-token ratio: what is normal

TTR is length-sensitive. A 50-word excerpt always shows a higher TTR than a 5,000-word essay from the same author. Compare TTRs only between texts of similar length.

Frequency vs concordance

Frequency counting answers "how many times does each word appear?". Concordance answers "where does each word appear?" - in which sentences, with what neighboring words. This tool surfaces frequency. For concordance (KWIC, key word in context), pair the frequency output with a search tool that highlights every match in the source text.

A common workflow: run frequency, identify the top non-stopword (e.g. "absolutely" appearing 40 times), then use browser find (Cmd-F or Ctrl-F) on the source to step through each occurrence and decide which to keep and which to replace.

For longer documents, the type-token ratio (unique / total) is a single-number summary of vocabulary richness. Watch the TTR trend as you edit: a draft that drops from TTR 0.42 to 0.38 has gained repetition and may need diversification; a draft that climbs from 0.42 to 0.48 has gained variety.

Privacy and processing

All counting runs in your browser. Manuscripts, blog drafts, transcripts, and confidential documents stay local. The tool does not send the text to a server, does not log, and does not store. CSV export uses the browser's Blob API. Safe for any text you would not paste into a third-party SaaS tool.