Free Keyword Density Checker

Paste text or enter a web page address to see how often words and phrases repeat. Counts, percentages and the formula behind each number. Free.

Pasted text is counted in your browser and never sent to our servers.
0 of 100,000 characters
Hides phrases that start or end with a common English word such as the, of or and. The total word count does not change.
How a phrase is counted
A two-word phrase used 6 times in 1,000 words is 1.2%.
The same phrase is 0.6%. Some calculators work this way.

Paste some text, or fetch a page, to see how often its words and phrases repeat.

How we count
  • A word is a run of letters or digits in any alphabet. An apostrophe or a hyphen between two letters stays inside the word, so it's and well-known count once. Numbers with a dot or comma, such as 3.14 and 1,000, count as two words. Chinese and Japanese count two characters per word. Soft hyphens, the invisible marks that let a long word break across lines, are removed first.
  • A phrase is one, two or three words in a row, compared in lower case. A phrase never runs across the end of a sentence or a line break.
  • Repeats that overlap are counted at every start, so “a a a” holds two copies of “a a”. A density above 100 percent is shown as 100.
  • With Hide common words on, a phrase is hidden when it starts or ends with a common English word, so “state of the art” stays and “of the” and “the cat” are hidden. The word list is English only. Counts, densities and the total never change.
  • Phrases used only once are hidden unless you tick the box, because almost every phrase in a text appears once.
  • For a web address, the server reads the HTML it is sent and keeps the text of the page body. Scripts, styles, templates and elements marked hidden in the markup are skipped. Menus, headers and footers stay in. Text added by JavaScript, or hidden with a style sheet, cannot be seen.
  • “First 100 words” means the first 100 words of the page text, which often start with the menu.
  • Pasted text: up to 100,000 characters. A web page: up to 2 MB of HTML and 200,000 characters of text.

What Is a Keyword Density Checker?

Quick answer: A keyword density checker counts how often each word and each two- or three-word phrase appears in a text and shows the share it takes up. Paste text or enter a web page address, then check your own text, free and with no sign-up.

This free keyword density checker is a word frequency counter for phrases, and it prints the formula behind every figure. It gives no pass or fail, no score and no ideal range.

How to Check Keyword Density With This Tool

Results update as you type or paste in this keyword density tool.

  1. Choose a mode. "Paste text" is selected on load; "Web page address" swaps in an address field.
  2. Add your text. Paste up to 100,000 characters, press "Use sample text", or type an address such as example.com/blog/my-post and press "Fetch and check".
  3. Read the numbers. "Total words" and "Different words" lead, then the formula and three phrase tables. "Hide common words" starts on and "Show phrases used only once" starts off.
  4. Click a phrase. Click to highlight matches in your text, and "Previous match" and "Next match" step through them.
  5. Add your own keywords. In "Check your own keywords", enter up to 10 phrases of up to 8 words, one per line. One that appears nowhere says "Not found in the text". "Download CSV" saves the tables as comma-separated values (CSV).

Field note: Check a finished draft, because short texts swing. In a 35-word paragraph written to repeat "coffee grinder", 8 uses read 8 × 2 ÷ 35 × 100 = 45.71 percent. The 95-word sample text reads 8.42 percent for "keyword density". Paste your draft in the tool above.

How Keyword Density Is Calculated

The keyword density formula is plain arithmetic: keyword frequency, the raw count, becomes a share of all the words in a text. The main decision a keyword density calculator makes is how to treat a phrase longer than one word.

MethodFormulaExample, 1,000 words in total
Count every word of the phrase (default)times the phrase appears × words in the phrase ÷ total words × 100Two-word phrase used 6 times: 6 × 2 ÷ 1,000 × 100 = 1.2%
Count each match oncetimes the phrase appears ÷ total words × 100The same phrase: 6 ÷ 1,000 × 100 = 0.6%
One-word phrase, either methodtimes the phrase appears ÷ total words × 100Used 12 times: 12 ÷ 1,000 × 100 = 1.2%

The switch "How a phrase is counted" picks the method, and the two differ by the phrase length: double for two words, triple for three. Pick one and keep it while you compare drafts.

The denominator, the total you divide by, is every word, common words included, so "Hide common words" never changes a count, a percentage or the total word count. Overlapping repeats count at every start ("a a a" holds two copies of "a a", about 133%), and a density above 100 is shown as 100.

Check the Keyword Density of a Web Page by URL

To get keyword density by URL, choose "Web page address" and type the address of one public page.

  • Fetching. Our server requests one page at a time with the user agent SEOSiteCheckerBot, follows up to 5 redirects and reads up to 2 MB of HTML.
  • Counting. Your browser counts the body text, headings included. The title tag, meta description and image alt text feed "Found in" only and never enter a total.
  • The "Found in" line. It lists where each phrase sits: Title, Meta description, Headings, Image alt text, Page address (the path of the page URL only) or First 100 words, which often begin with the menu. A heading split by a line break or by tags can lose its space and be missed. "What was read from the page" shows what we received.
  • Limits. A bot check or block page is reported and not counted, and a PDF is refused. Without an account you can fetch 6 pages a minute, with a free account 20.
  • The SEO report counts differently. Our SEO score checker has its own tables for single words and phrases of two to five words, built from a rendered page with other word rules.

How to Read the Results

These are real runs from November 6, 2026, all with the default method.

SourceTotal wordsPhrase we checkedTimes usedDensity
Google's spam policies page, by address5,873spam policies70.24%
Our home page, by address722seo site checker166.65%
A 35-word paragraph we wrote to repeat one phrase35coffee grinder845.71%

Read "Times used" first, because a percentage only compares rows within one method and swings in a short text. "One-word phrases" show the repeated words that set the topic, while "Two-word phrases" and "Three-word phrases" show your angle. Check "Found in" for your main keyword before worrying about repeats, as the keyword usage guide explains.

Our home page reaches 6.65 percent (16 × 3 ÷ 722 × 100) on its own three-word name because the name repeats in an ad blocker notice, headings, the FAQ and the footer. With "Count each match once" the same 16 matches read 2.22 percent.

Is There a Good Keyword Density?

There is no ideal keyword density to hit, so this keyword density checker prints numbers and no verdict.

Google's spam policy on keyword stuffing describes filling a page with keywords or numbers to manipulate rankings and lists repeating words until they sound unnatural. Google's SEO Starter Guide adds that excessively repeating the same words, variations included, is tiring for users. Neither Google Search Central page names a percentage, and the word "density" appears in neither.

Many guides quote ranges as rules of thumb. Neither Google page backs one, so this tool prints no range, color or band. Unnatural repetition hurts readability first, so write for readers: click the phrase, read each match in its sentence and rewrite what sounds forced.

Note: This is general information and promises no ranking. The keyword density guide covers the background.

What This Tool Counts and What It Misses

When a count looks wrong, check the input: the Website Source Code Viewer shows the HTML our server received for the same address.

  • Server HTML only. Scripts are not run, so words a script adds later never reach the count. Google Search can run JavaScript, as how Google processes JavaScript explains, so numbers can differ from a rendered page.
  • One known bug. An inline script holding a closing tag that matches an element it sits inside, such as </div>, can end early and leak its code into the count. If function, var or return top your one-word table, that is why.
  • Menus, headers, footers and notices. All counted. Hidden text counts when a style sheet class hides it, because we load no style sheets, but the hidden attribute and an inline display:none hide it.
  • Plurals are different words. Grinder and grinders count separately, because nothing folds word forms or matches synonyms. The stop words behind "Hide common words" are English only, 182 words, so a German text keeps its common words.
  • Size and scope. We keep at most 200,000 characters of a page's text. Cookies are not kept between redirect hops, so a page that needs a cookie or consent step is read as a first visit sees it. A meta refresh or script redirect is not followed.

Your Pasted Text Stays in Your Browser

Pasted text is counted in your browser and is not sent to our servers. In web page mode only the address goes to us, and the page text comes back for counting.

When analytics is loaded, the page sends one usage event naming how the tool was used, never your text. For an unpublished draft, use "Paste text", because page mode only reads public addresses.

Turn the Numbers Into Edits

Make the table change a sentence:

  • Check the title and H1. In web page mode, see whether "Found in" lists them for your main phrase.
  • Hunt for forced repeats. Swap a few for a pronoun or a synonym, paste again and compare.
  • Add each target keyword. Put it in "Check your own keywords". If it says "Not found in the text", first ask whether the page covers that topic.
  • Save a CSV download before and after an edit. The density_method column records which method produced each file.
  • Choose phrases on purpose. Counting comes second, so build a keyword strategy around search intent first.

Related Free Tools

Next stops once the counts make sense.

Last updated

Frequently Asked Questions

Frequently Asked Questions (FAQs) is a list of common questions and answers provided to quickly address common concerns or inquiries.

What is keyword density?

Keyword density is the share of a text's words that one word or phrase takes up. A two-word phrase used 6 times in 1,000 words gives 6 × 2 ÷ 1,000 × 100 = 1.2 percent by default. The raw count is called term frequency.

How do you calculate keyword density?

Count the total words and how often the phrase appears, divide the count by the total and multiply by 100. For a multi-word phrase this tool also multiplies by the words in the phrase by default; the option 'Count each match once' divides the count alone.

What is a good keyword density for SEO?

There is no single good keyword density. Google's spam policies and SEO Starter Guide name no ideal percentage, and the spam policies warn against repeating words until the text sounds unnatural, a judgment about prose, not arithmetic. This tool gives no verdict: read the phrase aloud in its sentence.

Does keyword density still matter for SEO?

Not as a number to hit. Google's spam policies and SEO Starter Guide describe keyword stuffing and name no density target. TF-IDF, which compares a term with how common it is across many documents, is a different calculation. Counts still work as a mirror for forced repeats.

Why does this keyword density checker show a different percentage than another tool?

Three things differ. This tool counts every word of a phrase by default, while a switch counts each match once. It divides by every word, common words included, whereas some tools drop common words first, which raises each percentage. It also reads server HTML, not a rendered page.

Can I check the keyword density of a web page by URL?

Yes. Choose 'Web page address', type the address (https:// is optional) and press 'Fetch and check'. Our server reads one public page, and your browser counts its body text. A login or a bot check blocks reading, so paste the text instead.

Does the tool count text that JavaScript adds to a page?

No. It reads the HTML the server sends, before any script runs, so a page built mostly by script shows few words. Google Search can run JavaScript, so a rendered page can differ from these counts. The Website Source Code Viewer shows the exact HTML we received.

Is the text I paste sent to your servers?

No. Pasted text is counted by your browser and never sent to our servers, and the CSV is built there too. In web page mode only the address goes to us, so our server can fetch that one public page.

What is keyword stuffing, and how does it show up in the results?

Google defines it as filling a page with keywords or numbers to manipulate rankings, and repeating words until they sound unnatural is one example. In the results it can look like one phrase far ahead. Click it and read the matches: no number proves stuffing.
Share on Social Media: