Keyword Density Checker

How to use Keyword Density Checker

1. Paste the page copy you want to analyze.
2. Read the ranked tables for single words, two-word and three-word phrases.
3. Compare the top phrases against the topic you intended the page to cover.

What keyword density should I aim for?

There is no target, and chasing one is a mistake that predates modern search by fifteen years.

Google has not used raw keyword density as a ranking signal since the mid-2000s. Modern retrieval uses semantic embeddings and passage-level relevance — the system understands that a page about "reducing photo file size" is about image compression whether or not the exact phrase appears.

The frequently-quoted "1–2%" figure has no basis in anything Google has published. Writing to hit it produces text that reads as though it was written to hit it, which is exactly the pattern the helpful content system was built to demote.

What density analysis is genuinely useful for is diagnosis, not optimization. It tells you what a machine reading your page thinks it is about — which is sometimes not what you intended.

So what should I use this for?

• Checking topical focus. If your top phrases are not the subject of the page, the page is unfocused. A guide to mortgage rates whose top bigram is "click here" has a structural problem.
• Catching over-optimization. Density above roughly 4–5% for a single phrase reads as stuffing to both a reader and a classifier. This is the one threshold worth watching, and it is an upper bound rather than a target.
• Finding missing terms. Compare your phrase list against the top-ranking pages for your query. Terms that appear consistently in theirs and not at all in yours often indicate a subtopic you have not covered.
• Auditing competitors. Paste a competing page and read what it actually emphasizes, which is frequently different from what its headings claim.
• Spotting repetition. A phrase appearing thirty times in 1,200 words is usually a writing problem before it is an SEO problem.

What are n-grams and why do phrases matter more than words?

An n-gram is a run of n consecutive words. "image" is a unigram, "image compressor" a bigram, "free image compressor" a trigram.

Single words are close to useless for analysis because they lose all context. "Free" appearing forty times tells you nothing; "free trial" appearing forty times on a page about a free tool tells you the page has a conversion problem.

Search queries are overwhelmingly multi-word — the median is around four words and rising as voice and AI-assisted search grow. Analyzing bigrams and trigrams matches how people actually search, which is why this tool ranks all three lengths separately rather than only counting words.

How are stop words handled?

Stop words — the, a, of, in, is, and, to — are the most frequent words in any English text by a wide margin, and they carry no topical information. Without filtering, every density report would be a list of articles and prepositions.

This tool filters a standard English stop word list from unigram analysis. For bigrams and trigrams it filters only phrases that consist entirely of stop words, because phrases like "out of stock" and "how to install" are meaningful and would be destroyed by aggressive filtering.

The list is English. Running non-English text through it will produce a report dominated by that language’s function words, which is a limitation worth knowing rather than a bug.

Frequently asked questions