Is there an ideal keyword density percentage?

No - despite decades of SEO folklore about a 1-3% "ideal" density, Google has never published or endorsed such a number, and modern ranking systems evaluate topical relevance and language patterns far more holistically than a raw repetition count. This tool's 3% flag is a soft heuristic for spotting obviously mechanical repetition, not a target to hit.

Why are stopwords excluded from the single-keyword table but not from the phrase tables?

Removing stopwords word-by-word makes sense for single terms (nobody needs to know "the" appeared 40 times), but doing the same to phrases mangles them - stripping "in" out of "in the cloud" leaves the nonsensical fragment "cloud" attached to whatever word preceded it. Phrases are reported exactly as they appear in the text instead.

What counts as a word for the total word count?

Any run of letters, digits, or an internal apostrophe (so "don't" counts as one word, not two) after lowercasing and stripping surrounding punctuation. Standalone punctuation, HTML tags if you accidentally paste markup, and whitespace aren't counted.

Why does the tool call keyword stuffing a "warning" instead of an error?

Because there's genuinely no hard rule being violated - a niche technical term or a product name can legitimately repeat often in short copy without being spammy. The flag exists because disproportionate repetition (well above what natural language produces) tends to read badly to both search engines' language models and human readers, so it's worth a manual look, not an automatic failure.

Keyword Density Checker - Word & Phrase Frequency Analyzer

Overview

Introduction

Keyword density has a mixed reputation in SEO: it was central to how early search engines ranked pages, it's been publicly dismissed by Google as a ranking factor for over a decade, and yet checking it is still genuinely useful - not to hit a target percentage, but to catch text that repeats a phrase so often it reads as mechanical rather than written for people.

This tool tokenizes pasted body text and reports how often each word and phrase occurs, both as a raw count and as a percentage of total words, so you can spot unintentional over-repetition (or under-emphasis) before publishing.

What Is Keyword Density Checker?

A client-side text analyzer that breaks pasted copy into words, counts single-word frequency (excluding common English stopwords like "the" and "and"), and separately counts 2-word and 3-word phrase frequency without any filtering.

It works entirely on the text you paste - there's no fetching, no HTML parsing, and no external service involved, so it works equally well on a finished article, a draft in a text editor, or a competitor's copy you've copied for comparison.

How Keyword Density Checker Works

Text is lowercased and split into words using a simple alphanumeric-plus-apostrophe pattern, which strips surrounding punctuation while keeping contractions intact. The single-word table then removes a built-in stopword list (articles, prepositions, common pronouns, and so on) before counting frequency, since otherwise "the" and "and" would dominate every result.

Phrases are built by sliding a 2-word and separately a 3-word window across the full, unfiltered word list, so "in the cloud" stays intact as a phrase even though "in" and "the" would individually be filtered out of the single-word table. Every count is then expressed as a density percentage of the total word count.

When To Use Keyword Density Checker

Before publishing a page targeting a specific keyword, to confirm it (and its natural variations) actually appears in the body copy at a reasonable, non-mechanical frequency rather than being missing entirely or crammed in unnaturally.

When reviewing AI-assisted or outsourced content for repetitive phrasing, since generated or templated text sometimes repeats the same 3-word phrase far more than a human writer naturally would.

Often used alongside Meta Tag Checker and Heading Structure Analyzer.

Features

Advantages

Separates single-word and multi-word phrase frequency, since a target keyword is often a phrase ("project management software"), not a single word.
Works entirely offline on pasted text - no fetching, no CORS concerns, and no risk of exposing draft or embargoed copy to a third-party service.
Flags disproportionate repetition without pretending there's an official 'correct' density to aim for.

Limitations

The stopword list is English-only and fixed; running non-English text through the single-keyword table will filter nothing (or filter the wrong words), though the phrase tables still work language-agnostically.
This is a mechanical word-count tool, not a semantic one - it can't tell that "running shoes" and "trainers" mean roughly the same thing, so it won't catch a page that's topically thin despite low literal keyword repetition.

Examples

Short product paragraph

Input

Our running shoes are the best running shoes for marathon training. These running shoes provide excellent support for every runner.

Output

Total words: 20
Top keyword: "running" - 3 uses, 15.0% density (possible keyword stuffing)
Top phrase (2-word): "running shoes" - 3 uses, 15.0% density

"Running" and "running shoes" both appear three times in a 20-word paragraph, well past the soft 3% flag - a clear sign of mechanical repetition worth rewriting.

Best Practices & Notes

Best Practices

Write for the phrase your reader would actually type, then check that it (and close natural variations) shows up in the phrase tables a handful of times across a full-length page - not dozens.
Treat any single keyword flagged over 3% as a prompt to reread that paragraph aloud; if it sounds repetitive to your ear, it will likely read that way to a search engine's language model too.

Developer Notes

Tokenization intentionally keeps internal apostrophes (so "don't" is one token) but strips all other punctuation; phrase n-grams are built from the raw word list before stopword filtering so multi-word phrases stay grammatically intact.

Keyword Density Checker Use Cases

Checking a finished blog post for accidental over-repetition of a target keyword before publishing
Comparing your draft's phrase frequency against a competitor's published article for the same topic
Auditing AI-generated or templated copy for mechanical-sounding repeated phrases

Common Mistakes

Chasing a specific 'ideal' density percentage that doesn't correspond to any real Google ranking signal, at the expense of natural-sounding writing.
Only checking the single-keyword table and missing that a target phrase like "best hiking boots" is being repeated far more often as a unit than any individual word within it suggests.

Tips

Paste just the body/article text, not the full page HTML - navigation, footer, and sidebar text will otherwise skew the frequency counts.

Keyword Density Checker