Word frequency counter — words, phrases and density
Find words you have repeated in a draft or look for recurring phrases in notes and reviews. Count single words or groups of two or three, then filter the table and export the counts.
Runs in your browser — your input is not uploaded to ToolsNow.
Separate with spaces or commas.
The text you enter is processed locally in your browser and is not uploaded to ToolsNow.
Site files and analytics use separate network connections. Ads are off. Privacy & Cookies
How words are counted. Letters, digits and marks are grouped into words. Internal apostrophes and hyphens are kept, so “don’t” and “well-known” each count as one. Languages without spaces between words may not split as expected. Filters change what is counted and can change the share percentages.
The built-in stop-word list is English only. It filters common words such as “the” and “and”. Add words in the box above for other languages or your own needs.
Use counts to review your writing. They can help you spot repetition and compare drafts. The percentages are not an SEO score or a target for search rankings.
How to use
- Paste Text to analyse. Use an article, a transcript or a set of reviews. Start with one draft if you want to compare revisions later.
- Choose Count. Pick Single words, Two-word phrases or Three-word phrases. Ignore case starts on; Include numbers starts off.
- Adjust the filters. Set Minimum count or Minimum word length, or add Extra words to ignore separated by spaces or commas. Try the English stop-word option if common words dominate.
- Review and export. Search within results filters the visible table. Copy table, .csv and .json include the full result under your counting settings, including rows hidden by that search.
Questions
What counts as a word?
The pattern groups Unicode letters and digits, keeping internal apostrophes and hyphens: "don't" and "well-known" each count as one. It does not split languages without spaces into dictionary words. Separate combining accent marks can split a word, so visually identical text can produce different counts. Total words is measured before your filters.
Why do the Share percentages change, or add up to less than 100%?
Share is a term’s count divided by the total occurrences after number, stop-word and length filters. Minimum count and Search within results hide rows without changing that total. In "red red blue", red stays at 66.67% even if Minimum count is 2 and blue disappears. Phrase mode uses phrase occurrences as the denominator.
Are two-word phrases exact phrases from the document?
Not always. Punctuation and sentence breaks are removed before adjacent words are grouped. Stop words and other excluded words are also removed first, so "cat and dog" can become "cat dog". Turn off stop-word filtering and check the original passage before treating a result as a quotation.
Should I exclude stop words?
Try it to bring less common terms to the top, but check what disappears. The list is English and includes words such as "not", "no" and "can", which may matter in reviews or instructions. Extra words to ignore is case-insensitive even when Ignore case is off.
Does a high percentage mean a word is overused?
Not by itself. A product name may need repeating, while a favourite adjective may be easy to cut. Use the list to find passages worth rereading. It does not judge writing quality, topic relevance or search ranking.
How much text can I analyse?
There is no fixed input limit, but counting still runs on the page thread and a large document can pause the tab. Only the first 500 matching rows are displayed; copies and downloads include all result rows under the counting settings. Split very long documents if the page slows down.
Published by ToolsNow. How tools and sources are checked · Report a problem