Word Frequency Counter

Word frequency counter

Content editors need to know whether a blog post actually uses its target keyword more than “the” or “and”, and whether any word is used so often it reads as a tic. This counter processes your text, filters stopwords by language, optionally stems word variants together, and lists every content word with its count and percentage of total words. Copy the top 50 into a spreadsheet and you have a respectable text-analysis report.

How to count word frequency

  1. 1

    Paste text

    Anywhere from a tweet to a book chapter. Longer is more informative.

  2. 2

    Configure filters

    Stopwords per language, min length, ignore digits, stemming on/off.

  3. 3

    Process

    Tool tokenises, normalises case and counts each token.

  4. 4

    Browse ranked results

    Table of word, count and percentage, sortable and exportable as CSV.

What the output looks like

For a typical 1500-word blog post, stopwords filtered, stemmed:

Rank Word Count % of total
1 marketing 47 3.1%
2 campaign 38 2.5%
3 audience 29 1.9%
4 email 27 1.8%

If the word you wanted to rank for is not in the top 5, the post is probably not focused enough on that keyword. If a word you did not intend sits at the top, you have an unconscious verbal tic.

SEO keyword density targets

Current SEO guidance is forgiving compared to 2015:

Quality beats density by a wide margin in 2026 rankings.

Stemming and lemmatising

The tool supports Porter stemming for English and basic stemming for Romance languages. Lemmatising is available only for English.

Use cases beyond SEO

Frequently Asked Questions

A word counter gives totals: 1,500 words in your piece. A frequency counter gives per-word counts: “marketing” appears 47 times. Different tools for different questions.

For SEO density analysis, stemming usually reflects what search engines do better — Google matches “run”, “running” and “ran” as related. For literary analysis, exact forms preserve stylistic choices.

By default no. Toggle “include stopwords” if you want the raw distribution including “the”, “and”, “of”. That view is useful for readability but distracting for content analysis.

No. All tokenisation and counting happens in your browser.