Keyword Density Analysis

How this keyword density checker calculates density

Keyword density is the share of a term among all the words in a text: occurrences ÷ total words × 100%. If an article has 600 words and "espresso" appears 12 times, its density is 2%. Before counting, the checker lowercases the text and turns punctuation into spaces. The total includes every word, stop words too; the tables don't — common English and Russian stop words ("the", "and", "of", "for", "with"…) and one-letter words are filtered out so the list isn't topped by filler.

You get three tables: the 30 most frequent words, up to 20 two-word phrases and up to 15 three-word phrases. Phrases appear only if they occur at least twice. The bar next to each word is colour-coded by density: green up to 2.5%, yellow from 2.5% to 4%, red above 4%. "Diversity" at the top is unique meaningful words as a share of all words — the lower it is, the more repetitive the text.

How phrase density is calculated

Two- and three-word phrases are built after stop words are removed. So "price of the espresso machine" is counted as "price espresso machine", and a phrase can bridge a full stop, joining the end of one sentence to the start of the next. Phrase density uses the same formula — phrase occurrences ÷ total words. Some keyword density tools multiply that by the number of words in the phrase; this one doesn't, so phrase percentages here will look lower than in those tools. Compare like with like.

What is a good keyword density?

Google has never published one. Its spam policies define keyword stuffing as filling a page with keywords or numbers in an attempt to manipulate rankings — with no percentage attached — and Google has said repeatedly that there's no ideal density to aim for. The 1–3% guideline for a main term is SEO folklore, and the colours in the table are a rough heuristic rather than a verdict:

• Your main keyword is red — read the text aloud. If the repetition sounds unnatural, cut it back.
• A generic word is red ("company", "our", "services", "best") — a sign of boilerplate copy worth rewriting.
• Your main keyword isn't in the top ten — the text may not be about what you think it is. That's a bigger problem than over-optimisation.

Length matters. In a 300-word product description one extra mention adds 0.33%, so short copy turns red easily. In a 3,000-word guide, 2% means 60 mentions — already heavy for natural writing.

How to check keyword density on your own page

1. Copy only the main content. The checker analyses pasted text; it doesn't fetch URLs. Navigation, footer and sidebars repeat across the site and skew the result. Reader mode in your browser is a quick way to grab just the article.
2. Look at the top ten words. They should describe the page's topic. If "shipping", "warranty" and "contact" outrank "espresso", the page is dominated by template text.
3. Study the two-word phrases. They're closer to real search queries. Ideally you'll see the main query and its variations: "espresso machine", "manual espresso", "milk frother".
4. Compare with the pages that rank. Paste the text of two or three top results for your query. Don't chase their percentages — look for subtopics they cover and you don't. Those are content gaps.
5. Check placement separately. Density says nothing about where a keyword sits. Make sure the main query appears in the title tag, H1 and opening paragraph — the meta tag checker shows the first two at a glance.

What to look for on different page types

These figures are observations, not rules. Your best benchmark is always the set of pages already ranking for your query.

What the counts don't merge

• No stemming. "Machine" and "machines", "brew", "brews" and "brewing" are counted separately. Add up the forms yourself for the real total.
• Hyphens and apostrophes split words. "E-commerce" becomes "commerce" (the single "e" is dropped), "don't" becomes "don". Keep that in mind when you look for compound terms.
• Numbers count as words. Years, prices and model numbers appear in the table alongside text.
• Stop words cover English and Russian only. For German, French or Spanish text, articles and prepositions will show up in the word list — just skip them.

Signs of keyword stuffing — and how to fix it

Stuffing is obvious without any percentages: the same phrase in every paragraph, blocks of city names or synonyms nobody reads, ungrammatical exact-match phrases like "best espresso machine cheap UK" dropped into a sentence. To fix it:

• replace some repeats with pronouns and descriptive references ("it", "this model", "the machine");
• delete keyword lists and "we serve these cities" blocks — the ranking upside is long gone, the risk isn't;
• cover the topic more fully: answers to real questions, specs and comparisons naturally dilute the keyword with related vocabulary;
• check whether another page on your site targets the same query — the keyword cannibalization checker finds this from Google's data.

Beyond keyword density

Density is a blunt metric: it sees repetition, not meaning. To see whether word frequencies in a text look natural, run it through the Zipf's law analyzer — in natural writing frequencies fall off smoothly, while in over-optimised copy a few words spike. For why density is still worth a look and how not to overdo it, see keyword density: why check it today.