Free SEO Tool

Keyword Density Checker

Paste your copy, or a whole page's HTML, and see how often every word and phrase appears, where your target keyword sits, and whether anything reads like it was written for a crawler. Add up to three competitor pages and it lists the terms they cover that you don't. Nothing is uploaded. The text is analysed in your browser and never leaves it.

What is keyword density?

Keyword density is the share of the words on a page taken up by one keyword or phrase, shown as a percentage.

This checker uses the formula below, because it is the only version that treats a three-word phrase fairly against a one-word keyword.

Plenty of tools skip the middle step and divide uses by total words instead, which gives the same page 0.5%.

Neither is wrong. They just measure different things, which is one reason the "ideal density" advice you read online never agrees with itself.

Keyword frequency is the simpler number: how many times the phrase appears. The results show both, plus uses per 1,000 words, so a long page and a short page can be compared fairly.

Is keyword density a ranking factor?

No.

Matt Cutts, who ran Google's webspam team, said in a 2011 video: "I would love it if people could stop obsessing about keyword density."

John Mueller has been just as clear: "Keyword density, in general, is something I wouldn't focus on."

What Google does care about is repetition done to manipulate rankings. Its spam policies define keyword stuffing as "filling a web page with keywords or numbers in an attempt to manipulate rankings", and one of the examples is repeating the same words or phrases so often that it sounds unnatural.

So density isn't a target.

However, it is a decent smoke alarm. If one phrase takes up 6% of a page, something has usually gone wrong in the writing, and that is exactly what this tool is built to catch.

What is a good keyword density?

There isn't one, and anyone who gives you a single number is guessing.

The rule of thumb you will see quoted most often is 1% to 2%. That is a fair description of what happens when you write naturally about a topic: the phrase turns up in the title, the opening, a heading or two and a handful of sentences.

It is a description, not a target.

A page forced up to 2% reads worse than a page sitting at 0.4% that covers the topic properly, and the second page is the one I would expect to rank. That is why this checker gives you a verdict rather than a pass mark.

• Say it where it counts: the title tag, the H1, the first 100 words and at least one subheading.
• Then write for the reader: use pronouns, synonyms and the words people actually use about the topic.
• Check the extremes: run the finished draft through this tool and look at anything flagged heavy or stuffed.

If you are writing the page from scratch, my guide to SEO content writing and the on-page SEO checklist cover the rest of it.

How this checker works

• Counting: text is split into words on spaces and punctuation. Hyphenated words are split too, so "on-page" and "on page" match each other.
• Phrases: two- and three-word phrases never cross a sentence boundary, so "shoes. Running" is not counted as a phrase.
• Stop words: with the box ticked, words like the, and and of are left out of the single-word table, and phrases that start or end with one are skipped.
• Variants: plurals and -ing or -ed forms of your keyword are counted as variants, separately from exact matches. The stemmer is deliberately light, because anything cleverer starts merging words that mean different things.
• HTML: paste a page's source and it reads the title tag, H1, H2s, H3s and meta description, ignores scripts and styles, and drops the navigation and footer when the page has a main content area.
• Readability: a Flesch reading ease score, with syllables estimated rather than looked up, so treat it as a guide.

The thresholds behind every verdict

These are the lines I use when I review copy for clients. A verdict you can't inspect is a verdict you can't trust, so here they are.

• Single words: heavy above 3% with at least 6 uses.
• Two-word phrases: heavy above 2% with at least 4 uses.
• Three-word phrases: heavy above 1.5% with at least 3 uses.
• Reads stuffed: double any of those lines.
• Your target keyword: heavy above 2.5% (3% for a one-word keyword) or in more than 25% of sentences. Reads stuffed at double that density, in more than 40% of sentences, or three times in a single sentence.
• Light: one use or none in 300 words or more.
• Competitor gap: missing if you never use a term that at least two competitors use (or one competitor uses three times). Light below 40% of their average rate. Overuse above two and a half times the highest competitor rate.

How to read your results

• Natural: the phrase is used often enough to make the topic obvious, and not so often that it reads as repetition. Leave it alone.
• Light: one use or none in a page of 300 words or more. If this is the page's main topic, say it plainly in the opening and in a heading.
• Heavy: above my density line or in more than a quarter of sentences. Not a penalty risk on its own. Read it aloud and swap a few uses for a pronoun.
• Reads stuffed: double the heavy line, in more than 40% of sentences, or three uses in one sentence. Rewrite those sentences.
• Not found: the phrase doesn't appear, not even as a plural. That is worth fixing if it is the query you want the page to rank for.

The strip under each keyword shows where it appears, from the start of the text to the end.

A phrase crammed into the first paragraph and then abandoned is one of the most common patterns I see. It usually comes from a brief that says "use the keyword in the intro" and nothing else.

Keyword stuffing: what it looks like and how to fix it

Google's own examples of keyword stuffing are lists of phone numbers with no added value, blocks of text listing the cities and regions a page wants to rank for, and the same words or phrases repeated until the copy sounds unnatural.

The third one is what a density checker catches. In practice it looks like this:

Looking for trail running shoes? Our trail running shoes are the best trail running shoes for runners who need trail running shoes that grip.

The opening of the worked example. "Trail running shoes" takes up 16.7% of its 198 words, and four of its eleven uses are in these two sentences.

The fixes are almost always editorial rather than technical.

• Use pronouns: after the first mention in a paragraph, "they", "them" and "these" do the job.
• Name the thing the way people do: "trail shoes", "a pair", "your shoes". Real writing varies the words without trying.
• Cut the sentences that only exist to say the keyword: "Looking for X? Our X are the best X" adds nothing a reader needs.
• Put the phrase where it does its job: title, H1, the opening sentence and a heading. That covers the relevance signal the phrase sends.
• Cover the topic instead: if you feel the urge to repeat the keyword, it usually means a subtopic is missing. The competitor gap report will tell you which.

The competitor term gap

Density tells you how often you said something. The gap report tells you what you never said at all.

Paste the main text of two or three pages that rank for your keyword and the checker compares every word and phrase, measured as uses per 1,000 words so a 3,000-word guide doesn't drown out a 600-word product page.

• Missing: you never use the term, and at least two competitors do (or a single competitor uses it three times or more).
• Light: you use it, but at less than 40% of their average rate.
• You use far more: your rate is more than two and a half times the highest competitor's. Worth a read-through for repetition.

In the worked example, the page never mentions a rock plate, heel drop or a waterproof membrane, and both competitors explain all three. Those are the questions a buyer actually has, which is why they are worth covering.

Treat the list as a set of topics to consider, not words to paste in. It is plain frequency, not TF-IDF or an entity model, and it will happily flag a competitor's brand name.

For a fuller version of the same idea, see my guides to content gap analysis and entity SEO, or turn the gaps into a brief with the content brief generator.

Does keyword density matter for AI search?

Even less than it does for Google's blue links.

AI search systems such as ChatGPT, Perplexity and Google's AI Overviews retrieve passages by meaning rather than by counting exact matches, so repeating a phrase does nothing to help them choose your page.

What does help is a clear, quotable answer near the top, the entities a topic is expected to include, and headings that match the questions people ask.

The gap report is far more useful here than the density figure. My guide to answer engine optimisation covers the rest, and the AEO Readiness Score will grade a page against it.

What this tool will not do

• Fetch a URL: browsers block one site from reading another without a server in the middle. Keeping it server-free is also why your text never leaves your browser. Paste the source instead.
• Give you a density to hit: there isn't one, so it doesn't pretend.
• Judge search intent: a page can be perfectly balanced and still be the wrong kind of page for the query.
• Read languages other than English well: the counting works in any language, but the stop words, variants and readability score are English only.

Final Thoughts

Keyword density is a symptom, not a strategy.

If the number is extreme, the writing has a problem, and the fix is to write better rather than to hit a percentage.

Use the density figures to catch repetition, use the gap report to find what you've missed, and then check the title and description that sell the click with the meta description checker. Every other free tool I've built is on the SEO tools hub.

Frequently asked questions