What counts as a “word”?

By default: any maximal run of letters/digits separated by whitespace or punctuation. So “don’t” may split into “don” and “t” depending on the tokenizer. For careful linguistic analysis, use a real tokenizer (spaCy, NLTK). For quick frequency checks, the simple definition is fine.

Should I filter stop words?

Yes for content analysis (you want to see “photosynthesis” and “chlorophyll”, not “the” and “a”). No for stylometry (function-word frequency is exactly the fingerprint you want). No for full-text search index design (some implementations weight stop words downward instead of removing them entirely).

What’s a normal keyword density for SEO?

Modern SEO best practice is roughly 0.5-2.5% for the target keyword in a long-form article — natural usage that signals topic relevance without triggering keyword-stuffing penalties. Above 3% looks spammy to Google’s NLP. Match search-intent context matters far more than raw density.

Is the count case-sensitive?

Most word-frequency tools default to case-insensitive (lowercase everything before counting) so “The” and “the” merge. Toggle case sensitivity on if you want to distinguish proper nouns from sentence-initial capitalized words.

How does this differ from a TF-IDF tool?

Word frequency = raw count within ONE document. TF-IDF = frequency in this document divided by frequency across a whole corpus, so common-everywhere words get downweighted. For comparing one document to itself you want frequency. For comparing across many documents (search ranking, classification) you want TF-IDF.

Can I use this on code or markup?

Yes, but results are usually noisy because punctuation, identifiers, and operators don’t map cleanly to natural-language words. For source code, use a code-specific frequency tool that knows about identifiers, keywords, and string literals.

Text & Writing Utilities · Free tool

Word Frequency Counter

Paste any text to see the most common words, counts, and percentages. Filters stop-words optional.

Updated June 2026

Ignore stop-wordsTop N:Total words: 15

#	Word	Count	%
1	fox	3	20.0%
2	dog	2	13.3%
3	quick	2	13.3%
4	away	1	6.7%
5	barks	1	6.7%
6	brown	1	6.7%
7	jumps	1	6.7%
8	lazy	1	6.7%
9	over	1	6.7%
10	runs	1	6.7%
11	smart	1	6.7%

Found this useful?Email Buy Me a Coffee

What it does

A word frequency counter takes any block of text and returns a sorted table of every unique word, how many times it appears, and what percentage of the total it represents. The technique sits at the foundation of natural-language processing — Zipf’s law (the empirical observation that in any natural-language corpus, the frequency of any word is inversely proportional to its rank in the frequency table) was discovered by counting words this way. Word counts are the input to TF-IDF, the bag-of-words model, naive Bayes spam filters, keyword density analysis for SEO, and stylometric analysis (figuring out who wrote a contested document by their word-choice fingerprint).

The tool tokenizes your text by splitting on whitespace and punctuation, lowercases for case-insensitive counting (so “The” and “the” merge), optionally filters “stop words” — the 200-or-so highest-frequency English words (the, a, of, and, to, in, that, …) which carry structure but little semantic meaning — then sorts by count descending. You see the top hundreds of words at a glance, copy the table for further analysis, and identify what your text is actually ABOUT versus what it just contains.

Common use cases: SEO content audits (does this article actually emphasize the target keyword, or does it just mention it once?), essay revision (am I overusing “very”, “really”, “just”, “that”?), translation glossary building (what are the most-used terms in this manual that I need to translate consistently?), curriculum design (which vocabulary words appear most often in this chapter, suitable for spaced-repetition?), and forensic linguistics (the Federalist Papers authorship debate was largely settled via word-frequency counts of function words like “while” vs “whilst”).

Embed this tool on your siteShow snippet

Paste this snippet into any page. Loads on-demand (lazy), no tracking scripts, and sized to most dashboards. Replace the height to fit your layout.

<iframe src="https://freetoolarena.com/embed/word-frequency-counter" width="100%" height="720" frameborder="0" loading="lazy" title="Word Frequency Counter" style="border:1px solid #e2e8f0;border-radius:12px;max-width:720px;"></iframe>

Embed docs →

How to use it

Paste your text into the input box or drop a .txt file.
Toggle “filter stop words” if you want to skip “the/a/of/and” and similar function words.
Click Count — table appears sorted by frequency, showing word, count, and percentage of total.
Copy the table or download as CSV for further analysis in Excel/Sheets.
Tweak parameters (case sensitivity, minimum length) and re-run if needed.

When to use this tool

SEO content auditing — checking actual keyword density vs intended target.
Essay or article revision — finding overused words to vary your prose.
Translation prep — identifying high-frequency terms that need glossary entries.
Vocabulary list building from a real-world text source (news article, novel chapter).
Quick stylometric or authorship analysis on contested text.

When not to use it

Sentiment or intent analysis — frequency alone misses negation, sarcasm, context. Use a real NLP model.
Multi-language text — the simple tokenizer assumes whitespace separation, which fails for Chinese/Japanese/Thai.
Massive corpora (>10 MB) — browser-based counting is fine up to a few hundred KB; for big data use Python’s collections.Counter or grep/awk pipelines.
Lemmatization-aware counting (treating “run”, “ran”, “running” as one word) — that needs a real linguistic library, not raw counting.

Common use cases

Quick use during a typical workday
Pre-decision sanity-check on inputs and outputs
Educational use — demonstrating the underlying concept
Onboarding a colleague who needs the same calculation/conversion

Frequently asked questions

What counts as a “word”?: By default: any maximal run of letters/digits separated by whitespace or punctuation. So “don’t” may split into “don” and “t” depending on the tokenizer. For careful linguistic analysis, use a real tokenizer (spaCy, NLTK). For quick frequency checks, the simple definition is fine.
Should I filter stop words?: Yes for content analysis (you want to see “photosynthesis” and “chlorophyll”, not “the” and “a”). No for stylometry (function-word frequency is exactly the fingerprint you want). No for full-text search index design (some implementations weight stop words downward instead of removing them entirely).
What’s a normal keyword density for SEO?: Modern SEO best practice is roughly 0.5-2.5% for the target keyword in a long-form article — natural usage that signals topic relevance without triggering keyword-stuffing penalties. Above 3% looks spammy to Google’s NLP. Match search-intent context matters far more than raw density.
Is the count case-sensitive?: Most word-frequency tools default to case-insensitive (lowercase everything before counting) so “The” and “the” merge. Toggle case sensitivity on if you want to distinguish proper nouns from sentence-initial capitalized words.
How does this differ from a TF-IDF tool?: Word frequency = raw count within ONE document. TF-IDF = frequency in this document divided by frequency across a whole corpus, so common-everywhere words get downweighted. For comparing one document to itself you want frequency. For comparing across many documents (search ranking, classification) you want TF-IDF.
Can I use this on code or markup?: Yes, but results are usually noisy because punctuation, identifiers, and operators don’t map cleanly to natural-language words. For source code, use a code-specific frequency tool that knows about identifiers, keywords, and string literals.

Learn more

Explore more text & writing utilities tools

100% in-browserNo downloadsNo sign-upMalware-freeHow we keep this safe →