Is keyword density still a ranking factor?
Keyword density has not been a ranking factor for years. What it actually measures, when it is still useful, and what to look at instead.
No — keyword density stopped being a meaningful ranking factor years ago, and there is no ideal percentage to aim for. It is a descriptive statistic: how often a term appears relative to the total word count. It became a target by accident, back when search engines really did reward repetition. What it can still do is describe how a page is written, which is worth knowing. You can check the density and prominence of any text with the Docuboxer analyzer, which runs entirely in your browser.
What the number actually measures
The maths is trivial: occurrences divided by total words, times 100. In a 1,000-word article, a phrase used twelve times sits at 1.2%. That figure knows nothing about whether those twelve mentions are in the title and headings or crammed into one paragraph near the footer, and nothing about whether the surrounding text explains the topic or circles it.
Two pages with identical density can be excellent and unreadable respectively. That is the whole case against using the number as a goal — and the whole case for reading it as a symptom.
Why the 2% rule refuses to die
In the late 1990s and early 2000s, term frequency was weighted fairly directly, so repetition worked, so the industry repeated. Search engines pushed back on the abuse, and the hunt began for a safe zone: high enough to count, low enough not to trip anything. Those thresholds were never published by anyone with access to a ranking system. They were forum heuristics that got quoted until they sounded like consensus.
What kept them alive after that is tooling. Content briefs still ship with a "target density" field. Editor plugins still turn an indicator orange when a phrase appears four times instead of six. A red light you can turn green is enormously satisfying, and a genuinely useful check — does this page cover the topic properly? — cannot be reduced to one. So the percentage survived on interface design rather than evidence.
What a density report is genuinely good for
Not as a target. As a diagnostic, four uses hold up.
- Catching accidental stuffing. Almost nobody stuffs deliberately now, but it still happens: a long-tail phrase repeated in every subhead, a product name dropped into every paragraph out of habit, a page rewritten three times that accumulated every variant along the way. A frequency list sorted by count exposes that immediately.
- Checking prominence. This is the part worth your time. What matters is not how often the topic appears but whether it appears where weight is carried: the title tag, the H1, the opening sentence, the section headings. A page can look fine on density and still fail to say what it is about in the first hundred characters — the same hundred characters that shape the snippet Google shows.
- Auditing legacy content. Inherit a blog with posts from 2012 and frequency analysis is the fastest way to find the ones written for an algorithm that no longer exists. They stand out: sky-high density on one exact-match phrase, no variants anywhere, headings that reshuffle the same words.
- Testing the page against its own intent. Sometimes the most frequent phrase in a text is not what the text is about. If your guide to invoicing software leads with your own brand name, the page is talking about something else. No percentage tweak fixes that — a rewrite does.
Why our tool refuses to score an "ideal"
That is a deliberate editorial choice. The Docuboxer analyzer reports frequency and position for one-, two- and three-word phrases and stops there. No green badge at 2.4%, no warning that you are "below optimal", because there is no optimum and inventing one dressed as rigour would be worse than useless.
The trade-off is real: a scoring tool feels finished, and this one hands you data and expects judgement. We think that is the better deal. If all you need is length, the word counter gives you that without implying it means anything about rankings.
What to look at instead
- Topic coverage. Does the page answer the questions someone searching this actually has? Are the subtopics any practitioner would expect present? Thin coverage is a far more common failure than thin repetition.
- Related terms. A solid piece on mortgages mentions amortization, interest rate and closing costs without trying. If your draft repeats one phrase and none of its semantic neighbours, that is a depth problem showing up as a vocabulary problem.
- Intent match. If the query wants a comparison and your page sells, no amount of lexical tuning saves it. Look at what kind of page currently ranks and ask whether yours belongs to that family.
A five-minute pass
Paste the draft into the analyzer and read the two- and three-word phrases first — those reveal the real subject far better than single words. Confirm the main topic appears in the title, the opening sentence and at least one heading. Look for surprises: frequent terms that should not be, filler, a brand name showing up on autopilot. Then read the piece aloud and cut anything that sounds bent around a phrase. That is the only threshold that matters, and it is not a percentage.
Before you ship, check the title tag and meta description too. They are the two places the topic has to appear, and the two most people forget to verify.
Frequently asked questions
Is keyword density a Google ranking factor?
No, not in any usable sense. Search engines moved past raw term counting a long time ago and now interpret meaning: synonyms, related entities, context and intent. The only version of density that still has an effect is the abusive one — keyword stuffing, which search engines treat as a quality problem.
What is the ideal keyword density percentage?
There isn't one. No search engine publishes a target, and the numbers people quote — 2%, 3%, 1 to 2% — come from forum lore rather than documentation. The right density is whatever falls out of writing clearly for a reader. If you are editing a sentence to hit a number, you are making the page worse.
Why does my SEO plugin still tell me my keyword density is too low?
Because a red-to-green indicator is easy to build and satisfying to fix, not because the threshold is meaningful. Those checks encode a rule of thumb from the early 2010s. Treat them as a nudge to check that your topic is actually mentioned, then ignore the percentage itself.
What is keyword prominence, and is it more useful than density?
Prominence is where a term appears rather than how often. A phrase in the title tag, the H1 and the opening sentence signals the topic far more clearly than twenty scattered repetitions in the body. It is the part of a density report worth acting on.
How do I know if a page is keyword stuffed?
Read it aloud. Stuffing shows up as sentences bent around a phrase, the same exact-match term in consecutive headings, or lists of near-identical variants that add no information. A frequency report just confirms what your ear already flagged.
Is it safe to paste unpublished drafts into an online analyzer?
Only if the tool runs locally. The Docuboxer analyzer processes text entirely in your browser, so nothing is uploaded — which matters when the draft is a client deliverable or an unannounced launch page.
Analyze your text
Frequency and prominence for 1–3 word phrases. No invented scores, no upload.
Open keyword density checker →Related tools
- Keyword density checker — Frequency and prominence for 1–3 word phrases.
- Word counter — Words, characters and paragraphs at a glance.
- Google snippet preview — See whether your topic survives the SERP.
- Meta tag generator — Title, description and Open Graph with a live preview.
You might also like: free SEO tools that respect your privacy and the text utilities worth keeping around.