Keyword Density Checker

CHECK WORDS AND EXACT PHRASES

See how often a keyword appears in your text

Add your draft to the keyword density checker and enter the words or phrases you want to track before running the calculation. A public webpage URL works too. For a URL, you first review the fetched text and choose the part of the page you want the checker to analyze.

Single words, two-word phrases, and three-word phrases each have a separate results table. There is also a table for your tracked keywords, so you can find a phrase with one occurrence or none at all without looking for it among the most frequent terms.

CALCULATION

How this keyword density checker calculates percentages

An occurrence means one exact match, whether the term is a word or a phrase. Every table divides that count by the total words and multiplies it by 100. On a 1,000-word page, a phrase with 4 matches has an occurrence density of 0.4%. Expressed per 1,000 words, the same count gives a normalized rate of 4.

For a two-word keyword that appears four times, the count is four occurrences. There is no multiplication by the number of words within the keyword. When a different tool reports another density for the same text, check how its formula compares with the one shown here.

A keyword density example you can reproduce

With English selected, add “local editors” and “local editor” as tracked phrases. There are 12 words in the example, and “Local editors” matches twice. That gives an occurrence density of 16.67%, or 166.67 per 1,000 words. The singular “Local editor” has zero exact occurrences. Capitalization does not affect matching, but singular and plural forms count as different tokens.

PHRASE ANALYSIS

What stop words and phrase matching change

A unigram is a single word. For a bigram, the checker takes two consecutive words, such as “keyword density”; a trigram has three, as in “free SEO tool.” It builds these sequences before filtering stop words. Leaving stop-word noise out of the results therefore cannot create a phrase by joining words that had a stop word between them in the source.

Excluding stop words hides them individually from the unigram table. A phrase is hidden only if all of its words are stop words. Neither the total-word denominator nor the counts for tracked exact phrases change with this filter. In the word frequency counter, percentages work differently because their denominator is the words left after filtering.

Phrase matching uses the tokenized text, where punctuation separates tokens. That means an exact match can cross a sentence boundary. Words with different endings and synonyms remain separate, so select the term and read its source context to see whether the match belongs to the passage you intended to measure.

REVIEW A REVISION

Compare keyword density before and after editing

Save the original analysis as result A, then add the revision as B and run it again. For this comparison, the saved language, stop-word list, and tracked keywords stay locked. Read the change as B minus A in percentage points. Moving from 0.4% to 0.6%, for instance, means a rise of 0.2 percentage points.

One-off terms remain in the returned result even when the minimum-count filter hides them on screen. A CSV export takes the active phrase-length table; JSON includes the returned tables for words, phrases, and tracked keywords. The screen displays up to 500 matching rows. Each phrase-length result can contain up to 2,000 rows, and a notice tells you when it is partial.

The browser analyzes pasted text directly. With a URL, the server fetches the page without retaining its content, then your browser analyzes the text you have reviewed. Before comparing webpages, check that the source scope is the same on each. If just one includes menus and footers, their repeated words can overwhelm the article vocabulary.

LIMITATIONS

Is there an ideal keyword density?

There is no percentage that works as a universal keyword-density target. Counts can reveal a repeat you did not intend, or confirm that an edit kept an exact phrase. They cannot assess search intent, factual usefulness, originality, readability, or semantic coverage. Whether a keyword needs adding or removing is a decision to make after reading the passage.

Google's keyword-stuffing policy addresses unnatural repetition intended to manipulate rankings. You can find broader guidance in its SEO Starter Guide. The calculation and editorial examples are in Keyword density: formula and limitations.