How the density is counted
The checker splits the text into words, in any script, and leaves out emoji, symbols and punctuation. A phrase is two or three words in a row inside the same sentence: it never runs across a full stop, comma, bracket or line break, so "today. Gold" is not counted as a phrase. Density is the number of times a phrase is used, multiplied by the number of words in it, divided by all the words in the text. A 2-word phrase used 5 times in a 500-word article therefore takes up 2% of the text. Numbers on their own are left out of the single-word list, and 2- and 3-word phrases are listed once they appear at least twice.
There is no ideal keyword density
Many tools still promise a perfect percentage, such as 1 to 2%. Google has never published one. Its spam policies, last updated on 28 August 2026, describe keyword stuffing as filling a page with keywords or numbers to manipulate rankings, for example repeating the same words or phrases so often that the text sounds unnatural, or listing cities a page hopes to rank for. Google's SEO Starter Guide, last updated on 10 December 2025, adds that repeating the same words over and over, even in variations, is tiring for readers. We read both pages on 22 October 2026. So treat the numbers as a mirror, not a target: they show when a word is overused, or when the main topic barely appears at all.
Arabic text: letter forms, tashkeel and stop words
Arabic writers spell the same word in more than one way, and a simple counter would split them. With grouping switched on, أ, إ, آ and ٱ count as ا, ة counts as ه and ى as ي, and tashkeel and the tatweel stretch are ignored, so مدرسة, مدرسه and مَدْرَسَة are one word; the table shows how many spellings were merged. The Arabic stop-word list covers prepositions, pronouns and particles such as في، من، على، إلى، التي and الذي. The checker does not cut off prefixes or find word roots, so الذهب and ذهب are still counted as two different words. Mixed Arabic and English text works in the same box.
Where the keyword appears: opening, title and headings
Readers who skim a page look at its title, headings and first lines, so the main topic should be clear there. When you paste HTML, the checker reads the title tag, the meta description and the H1 to H6 headings, and leaves out scripts, styles, comments and the head section from the word counts. It then reports whether your keyword appears in each place and in the first 100 words, and lists the headings that contain it. This is a writing check, not a ranking formula. To write the title and description themselves, use the meta tag generator.
Word frequency and phrases in a CSV
The same tables work as a word frequency counter and an n-gram list: switch between single words, 2-word and 3-word phrases, show up to 500 of each, or download all of them as a CSV with the phrase, the count and the density. The file is UTF-8 with a byte order mark, so Arabic opens correctly in Excel. For word, character, sentence and reading-time counts, use the word counter.