Word Frequency Counter
Count how often each word appears in your text. Includes density percentages, stop-word filtering, and case-insensitive grouping — for SEO keyword density and content analysis.
| Rank | Word | Count | Density |
|---|
Why word frequency matters
Word frequency reveals what your text is actually about — not what you think it's about. The five most-frequent non-stopword words in any well-focused article are usually the topic, the secondary topic, and the named entities. If your top words don't match your intended topic, the writing has drifted.
Keyword density for SEO
Keyword density is the share of your text occupied by your target keyword. There's no magic number, but the conventions:
- 0.5%–1.5% — natural usage; the keyword appears as often as the topic warrants.
- 1.5%–3% — emphasised but still natural for tightly-focused content.
- Above 3% — keyword stuffing risk; reads as spammy and Google deprioritises it.
Modern Google ranking is far more sophisticated than density alone — semantic relevance, topic coverage, and user signals matter much more — but density is still a useful sanity check.
How our counter tokenises
- Words are sequences of letters with optional internal apostrophes ("don't" = 1 word).
- Numbers count as words.
- Hyphenated compounds count as one word.
- Case is ignored — "The" and "the" merge.
- Optional stop-word filter removes high-frequency function words.
Use cases beyond SEO
- Editing — find your overused crutch words ("just", "really", "very").
- Style analysis — frequency profiles distinguish authors and genres.
- Vocabulary review — see which technical terms appear most often in your draft.
- Translation prep — terminologists use frequency lists to build glossaries.
Frequently asked questions
What's a good keyword density for SEO?
Aim for 0.5%–1.5% for the primary keyword in most blog posts. Higher densities risk keyword-stuffing penalties; lower may signal weak topical focus.
Are stop words counted?
By default, no — the stop-word filter hides high-frequency function words (the, a, of, and, etc.) so the meaningful content terms surface at the top.
Does the counter handle different word forms?
Words are matched case-insensitively but otherwise verbatim. "Run", "runs", and "running" appear as separate entries. For lemma-aware counting, you'd need a dedicated NLP pipeline.
How long can the input text be?
Browser version handles documents up to a few megabytes without lag.