See your content the way an algorithm sees it
You know what your article is about — but is that what it says? Keyword density analysis counts every word and phrase in a text and ranks them by frequency, revealing the gap between intended topic and statistical reality. Writers use it two ways: to confirm a target phrase is present enough to register, and to catch the opposite sin — unconscious repetition that reads as spam to engines and as nails-on-chalkboard to humans. Paste, click, see the table.
How to analyze your text
- Paste your article, page copy or competitor's text into the box.
- Keep “ignore stop words” ticked so the, and, of don't drown the signal.
- Click Analyze keywords.
- Review three tables: single keywords, 2-word and 3-word phrases — each with counts and density percentages.
Reading the numbers like an editor, not a robot
For a piece targeting a phrase, a healthy signal is the phrase appearing naturally in the 0.5–1.5% density band — present in the title's orbit, a heading, and several body mentions. Above ~3% the prose almost certainly sounds stuffed (read a paragraph aloud; you'll hear it). But the modern truth cuts deeper: search engines stopped counting raw keyword frequency years ago — they model topics. The phrase tables' real value is coverage: an article genuinely about espresso machines also mentions grinders, pressure, portafilters and crema. If your 2–3-word phrase table is all brand name and filler, the text is thin, however dense the main keyword.
The diagnostic uses nobody mentions
- The crutch-word mirror: every writer over-uses something (actually, leverage, solution). The single-word table names yours — find-and-replace does the rest.
- Competitor X-ray: paste a top-ranking page and see its phrase vocabulary — that's a topic checklist for your own draft, not for copying.
- Translation QA: translated copy often inflates one term where the source varied; density tables catch it instantly.
- Anchor-text audits: paste your link anchors to spot over-optimised repetition.
Sane targets, 2026 edition
Write for the reader first, then verify: target phrase comfortably present (not chanted), supporting vocabulary visible in the phrase tables, no accidental tic-word leading the single table. That trio — relevance, coverage, naturalness — is what density tools are actually for now. The robotic “exactly 2.3% or perish” era is dead; the mirror it holds up to your writing is more useful than ever.
Statistics in two seconds, entirely in your browser — your drafts and your competitors' pages stay between you and the math.
Quick reference
| Property | Detail |
|---|---|
| Analyzes | Single words, 2-word and 3-word phrases |
| Reports | Count + density % per term, total/unique words |
| Stop words | Optional filtering (recommended on) |
| Healthy density | ≈0.5–1.5% for a target phrase |
| Red flag | >3% — reads as stuffing to humans and engines |
| Language support | Unicode-aware (works beyond English) |
| Processing | 100% in-browser, no upload |
Frequently asked questions
What is the ideal keyword density for SEO?
There is no magic number — Google models topics, not frequencies. A natural 0.5–1.5% for the target phrase, surrounded by related vocabulary, is what well-ranking content statistically looks like. Above ~3% you're in stuffing territory.
Why filter stop words?
Words like “the” and “of” dominate any frequency count while carrying no topical meaning. Filtering them surfaces the terms that actually define your content. Untick the option only for linguistic analysis.
What do the 2- and 3-word phrase tables tell me?
Topic coverage. Multi-word phrases are where meaning lives (“espresso machine cleaning” vs “machine”). A rich phrase table signals depth; a barren one says the text repeats one idea thinly.
Can I analyze a competitor's page?
Yes — copy their visible text and paste it in. You'll see their vocabulary footprint: which subtopics they cover and how. Use it as a completeness checklist for your own writing, never as a template to copy.