Word Frequency Counter: Density, Phrases, Stop Words
Count how often each word appears, with keyword density per word, 2 and 3-word phrase counts, a 158-word stop-word filter and CSV export. Nothing is uploaded.
Don't take our word for it — turn your wi-fi off.
This tool keeps working with no connection at all, because your file is processed here in the browser. A site that uploads your file cannot do that. Disconnect and try it — this box will notice.
Prefer to check properly? Open DevTools → Network, then run the tool. Your file appears in no request. Why none of our tools upload
Put this tool on your site
Free to embed. No signup, no API key, no tracking scripts.
You're on 7BusyBoss — 300+ free tools that run instantly in your browser. No signup, nothing uploaded.
A word frequency counter counts how many times each word appears in a text and ranks the list from most to least used. Paste anything above and you get every word with its count and its keyword density, which is count ÷ total words × 100. Switch to 2-word or 3-word phrases to see which combinations repeat, hide the stop words so "the" and "and" stop crowding the top, and download the whole table as CSV.
Most used words in a 66-word paragraph
Here is a 66-word product description run through the counter, a standing desk listing of the kind that fills any shop page. With stop words included, the top of the table is grammar: "the" appears 8 times (12.12%), then "desk" 6 times (9.09%), "standing" 4 times (6.06%) and "and" 3 times (4.55%). Tick the stop-word filter and 26 words drop out, leaving 29 distinct words that say something about the product.
| Rank | Word (stop words kept) | Count | Density | Word (stop words excluded) | Count | Density |
|---|---|---|---|---|---|---|
| 1 | the | 8 | 12.12% | desk | 6 | 9.09% |
| 2 | desk | 6 | 9.09% | standing | 4 | 6.06% |
| 3 | standing | 4 | 6.06% | frame | 2 | 3.03% |
| 4 | and | 3 | 4.55% | height | 2 | 3.03% |
| 5 | a | 2 | 3.03% | top | 2 | 3.03% |
The density column is the same sum in both halves: 6 ÷ 66 × 100 = 9.09% for "desk". The denominator is always the full word count, stop words included, so a percentage means the same thing whichever boxes are ticked.
Now switch to 2-word phrases. "standing desk" appears 3 times (4.55%) and no other pair repeats. A single-word count cannot show you that. "desk" on its own scores 6, which overstates how often the full product name is used, and the phrase view is where you find out. With stop words kept, "the desk" ties at 3, which is why the filter also drops any phrase that starts or ends with a stop word. The 3-word view on this text gives 15 distinct phrases, each used once, so the paragraph has no three-word crutch.
Keyword density by word count
Keyword density is occurrences divided by total words, times 100. Twelve uses in a 1,000-word article is 1.20%. Two uses in a 300-word page is 0.67%. The table turns the usual 0.5% to 2% guidance into a number of occurrences you can count on the page.
| Words in text | 0.5% | 1% | 1.5% | 2% | 3% |
|---|---|---|---|---|---|
| 300 | 1.5 | 3 | 4.5 | 6 | 9 |
| 500 | 2.5 | 5 | 7.5 | 10 | 15 |
| 800 | 4 | 8 | 12 | 16 | 24 |
| 1,000 | 5 | 10 | 15 | 20 | 30 |
| 1,500 | 7.5 | 15 | 22.5 | 30 | 45 |
| 2,000 | 10 | 20 | 30 | 40 | 60 |
| 3,000 | 15 | 30 | 45 | 60 | 90 |
A fraction means the density falls between two whole counts. At 300 words, one use is 0.33% and two uses is 0.67%, so 1.5 in the 0.5% column tells you one or two.
Treat the 3% column as a ceiling. Thirty uses of one phrase in 1,000 words reads as stuffing to a person long before any algorithm notices. The common guidance is 0.5% to 2% for the main phrase, and the exact figure matters less than whether the phrase sits in the title, the first paragraph and a heading. To track one target phrase against a draft, the keyword density analyzer counts it as a whole phrase and reports its density against content words with stop words removed, so its percentage runs higher than the one here.
Stop words and what the filter removes
Stop words are the function words a sentence needs for grammar and that carry no topic: articles, pronouns, prepositions, conjunctions and auxiliary verbs. The filter here knows 158 of them, close to the list the NLTK toolkit ships, so your counts line up with what a Python tutorial would print. "don't" and "won't" are on the list as whole words, because this counter keeps contractions intact instead of splitting them into "don" and "t".
Leave the filter off when you are studying style. A writer who uses "that" twice a sentence will never find out with the filter on.
N-gram counter for 2-word and 3-word phrases
A 2-word phrase (a bigram) is any two words that sit next to each other, a 3-word phrase (a trigram) any three. Windows overlap, so a word in the middle of a sentence belongs to two bigrams. The counter never builds a phrase across a full stop, question mark, semicolon, colon, bracket, quotation mark or line break. In "The cat sat on the mat. The cat slept." the bigram count gives "the cat" twice and never produces "mat the".
With stop words excluded, a phrase is dropped when its first or last word is a stop word. "cup of tea" survives, since both ends are content words. "a cup of", "cup of" and "of tea" are dropped. Stop words in the middle stay, which keeps the phrases people search for.
How to use this word frequency counter
- Paste or type the text. Counting runs as you type.
- Pick Single words, 2-word phrases or 3-word phrases.
- Leave Exclude stop words ticked to see topic words, or untick it for a full count of word occurrences. Tick Case sensitive if "Apple" and "apple" should be counted apart.
- Sort by Most frequent first or A to Z. The table shows the top 50 rows, and Show all expands it.
- Download CSV saves every row (rank, word or phrase, count, density %) with a UTF-8 marker so Excel opens accented words correctly.
The cards above the table show total words, distinct words or phrases, and how many the stop-word filter removed. Nothing leaves your browser. If you need totals and nothing else, the word counter gives word, character, sentence and paragraph counts without the table.
What counts as a word
A word is a run of letters or digits, with apostrophes and hyphens allowed inside it. So "don't" is one word, "state-of-the-art" is one word, and "café" is one word with its accent intact. Curly apostrophes are converted to straight ones first, so "don’t" pasted from Word and "don't" typed on a keyboard count together. Numbers count as words: "2024" is a word, and "3.5" becomes two words, "3" and "5", because the full stop is treated as a sentence break. Quotation marks and brackets around a word are stripped.
Case-insensitive mode lowercases everything before counting. In case-sensitive mode "Desk" at the start of a sentence and "desk" mid-sentence are separate rows, which is what you want for proper nouns and rarely what you want for anything else.
Limits and common mistakes
- Density is a share of words, so a word used once in a 20-word caption shows 5.00%. Compare densities between texts of similar length.
- The stop-word list is English. Paste French or Hindi and the filter removes nothing useful, so untick it.
- Hyphenated words are kept whole. "e-mail" and "email" are two rows.
- Phrase density divides by total words, the same as single words, so phrase percentages are smaller than you may expect and are comparable across the three views.
- Pasting a whole book works. The table is capped at 50 rows on screen until you press Show all, and the CSV always has everything.
Frequency says which words a text leans on. It says nothing about how hard the text is to read, which is what the Flesch-Kincaid readability checker measures.
How to use the Word Frequency Counter
Takes about a minute. No signup, no download, your data stays in your browser.
- 1Open the tool. Scroll up to the Word Frequency Counter above — it loads instantly in your browser, no install needed.
- 2Enter your values. The fields come pre-filled with realistic defaults so you can see how it works — replace them with your own numbers.
- 3Read the result. The output updates instantly. Copy or share it — nothing is uploaded to a server, everything stays on your device.
Frequently asked questions
Common questions about the Word Frequency Counter.
What is a word frequency counter?
A tool that counts how many times each distinct word appears in a text and lists the words from most to least frequent. On a 66-word product description this one found 43 distinct words, with "the" (8) and "desk" (6) at the top. Switch the stop-word filter on and the list shrinks to the 29 words that carry meaning.
How do you calculate keyword density?
Divide the number of times the keyword appears by the total number of words, then multiply by 100. Five uses in 500 words is 1.00% and twelve in 1,000 is 1.20%. For a phrase, count each occurrence of the full phrase once and divide by the same total word count.
What is a good keyword density?
The common guidance is 0.5% to 2% for the main phrase, which is 5 to 20 uses in a 1,000-word article. Above 3%, or 30 uses in 1,000 words, copy starts to read as stuffed. Placement in the title, first paragraph and headings counts for more than the exact percentage.
What are stop words, and should I exclude them?
Function words such as "the", "and", "of" and "is" that hold a sentence together but say nothing about its topic. Exclude them to see what a text is about. Include them to study writing style, where an overused "that" or "which" is the point. The filter here uses 158 English words.
What is an n-gram counter?
A counter of word sequences instead of single words. 2-grams are pairs such as "standing desk" and 3-grams are triples. This tool builds them from adjacent words inside a sentence and never across punctuation. On the sample paragraph "standing desk" appeared 3 times while no other pair repeated.
Does the counter treat "Desk" and "desk" as the same word?
By default, yes. Everything is lowercased before counting. Tick Case sensitive and they become two rows. Curly and straight apostrophes are always merged, so "don’t" and "don't" count together in both modes.
How do I export the word frequency table to Excel?
Press Download CSV. The file holds rank, word or phrase, count and density % for every row, including those beyond the 50 shown on screen, and it starts with a UTF-8 byte-order mark so Excel reads accented characters correctly. Open it directly or import it into Google Sheets.
Is my text uploaded anywhere?
No. The counting is JavaScript running in your browser tab, and the text never leaves it. You can disconnect from the internet after the page loads and the counter keeps working.
Community rating
Discussion (0)
No comments yet. Start the discussion.
Keep exploring
Related tools across 7BusyBoss — all free, all instant.