What keyword density actually tells you
Keyword density is the percentage of times a word or phrase appears relative to the total word count. For a 1,000-word article that mentions "React" 25 times, the density is 2.5%.
The number was useful in 2005, when search engines literally counted keywords to score relevance. It's still useful in 2026, but for a different reason: it catches your unconscious habits. The phrases you over-use, the keywords you under-use, the synonym you keep forgetting to swap in. Density is a signal of your writing, not a hack for ranking.
The 1–2% rule (and why it's a guideline, not a law)
Most SEO writers aim for 1–2% density on the primary keyword. Below 1%, search engines struggle to identify the page topic. Above 3%, the article reads as obviously stuffed and modern algorithms penalize it. Between those bounds is the comfortable zone where the keyword shows up enough to register but not enough to look forced.
The reason this is a guideline, not a law, is that modern search engines use semantic embeddings rather than literal keyword counts. They understand "automobile" and "car" as the same topic, and they read context as well as repetition. So a 3% density of one literal word is worse than a 1% density of the same word plus a 1% density of three related synonyms.
Why multi-gram analysis matters more than it used to
Long-tail SEO targets multi-word phrases. "Email marketing tool" gets fewer searches than "email," but the searches it gets are higher-intent — someone typing the long phrase has a clearer goal. Most ranking opportunity in 2026 lives in the 2-, 3-, and 4-gram tail.
This tool surfaces 1, 2, 3, and 4-grams in separate views. Look at the 1-gram view to see the topic skeleton. Switch to 2-gram for the natural phrasing. Switch to 3-gram to find long-tail variations. Use 4-gram to uncover ultra-specific phrase patterns your competitors are hitting that you have missed.
Why stop-words are excluded by default
"The," "of," "and," "to," "a" — these dominate any density count. A typical English page is 5–8% "the" by token. Including stop-words drowns the signal in noise. Strip them, and you see the actual topic.
The toggle exists for the rare case where stop-words matter — analyzing very short text (where every word counts) or auditing for filler-word overuse in spoken-word transcripts.
Density vs variety: prefer variety
The biggest mistake in keyword-driven writing is hammering the same exact phrase. Modern search engines reward semantic comprehensiveness. If your topic is "React server components," your article should also touch "RSC," "server-rendered React," "React 19," "Next.js App Router," "streaming SSR" — phrases that map to the same topic from different angles.
Use this tool inversely. If your primary phrase is at 4% but the related synonyms are at 0%, the article is keyword-stuffed without being topically broad. Spread the density across the whole topic vocabulary.
What the numbers in this tool mean
For 1-grams, density is (occurrences of word) / (total words). Standard definition.
For 2-, 3-, and 4-grams, density is (occurrences of phrase) / (total phrases of that size). The denominator changes because there are roughly N words but only N–1 two-grams, N–2 three-grams, and N–3 four-grams in a document of N words. Using the right denominator means percentages within each n-size sum to a meaningful 100%.
The bar in each row visualizes density relative to the most-frequent phrase, not relative to 100%. This makes the visual ranking more useful — the gap between #1 and #2 matters more than the absolute density of either.