Keyword Density Analyzer

The Complete Keyword Density Analyzer Guide

How density is calculated, how the word and phrase tables are built, how target keywords are checked, and how to read the bands without chasing a number.

Open Keyword Density Analyzer

What the Analyzer Measures

Keyword density is the share of a text taken up by one word or phrase: its count divided by the total number of words. The analyzer applies that arithmetic to any text you paste and shows the result as summary tiles, frequency tables for single words and phrases, a table for your target keywords, and an optional highlighted copy of the text.

A density figure describes your writing; it is not a score. It shows when a phrase has been repeated until the copy sounds mechanical, when an important term is missing from the opening, and which phrases a reader would call the subject. Search engines publish no target density, so the bands are a rule of thumb.

Control by Control

The controls sit in three places: a row of options above the text, the text box and target field, and the results below. Here is what each one does.

Options row

Strip HTML tags is off by default. Turn it on when you paste page source: script and style blocks, comments and tags are removed, and only visible words are counted. Ignore stop words (on by default) hides common function words from the single-word table and skips phrases that begin or end with one. Case-insensitive (on) treats Delivery and delivery as one word. Min count (default 2, allowed 1 to 1000) hides rows that appear fewer times, and Rows (default 15, allowed 1 to 200) sets how many rows each table shows. An invalid value turns red and a safe default is used.

Load sample (kdaSampleBtn) fills the text box and target field with a short bakery paragraph and turns Strip HTML tags off. Clear (kdaClearBtn) empties the text and target field, but leaves your option choices alone.

Your text or HTML

The text box accepts plain text or HTML. The line above it counts characters before any stripping. A notice appears when the text looks like HTML while Strip HTML tags is off, or when it exceeds 200 KB.

Target keyword(s)

Enter phrases separated by commas. Each is matched as a whole phrase, in order, so sourdough bread matches only where those two words sit side by side. Duplicates and blanks are dropped. Highlight in text adds a read-only copy of the text with every match marked, showing where repetition clusters.

Summary tiles

Five tiles show total words, unique words, sentences, an estimated reading time at 200 words a minute, and characters. Sentences end at full stops, question marks or exclamation marks followed by a space, and at blank lines.

Target keywords table

Each target shows its count, its density to two decimal places, whether it appears in the first 100 words, and a guidance band: Not found for zero, Below 0.5%, Within 0.5-2.5%, or Above 2.5%. Target matching ignores the stop word option, so in the is still counted.

Frequency tables

Three tables list single words, two-word phrases and three-word phrases, each with count and density. Rows below the minimum count are hidden. Phrases are built only inside a sentence, so the last word of one sentence never joins the first word of the next.

Copy and download

Copy results as CSV places the results on the clipboard, and Download CSV saves them as keyword-density.csv. The file has a header row named Section, Term, Count, Density % and Note, followed by summary rows, target rows and the three frequency sections.

Worked Example

A short example makes the arithmetic visible. Paste this three-sentence text with the default options, and enter sourdough bread, baked daily, delivery as the targets:

Sourdough bread in Pune. Sourdough bread is baked daily. Order sourdough bread today.

Counting the words

The text has 13 words, including stop words such as in and is, because density always uses the full total. There are three sentences, and the reading time is 13 divided by 200 minutes, or 3.9 seconds, shown as 4 s. There are 9 unique words.

Single words

With stop words ignored, sourdough and bread each appear three times, so with the default minimum of 2 they are the only rows. Each gives 3 divided by 13, multiplied by 100, which is 23.08%. The tie is ordered alphabetically, so bread is listed first.

Two-word and three-word phrases

In the two-word table, "sourdough bread" appears in all three sentences, so it has a count of 3 and a density of 23.08%. The three-word table is empty at the default minimum, and its message reads "Nothing above the minimum count". With Min count set to 1, it lists four phrases, each at 7.69%: "bread in pune", "bread is baked", "order sourdough bread" and "sourdough bread today". Middle stop words are allowed, because only the first and last words of a phrase are checked against the stop list.

Target keywords

The three targets produce three rows:

TargetCountWorkingDensityIn first 100 wordsBand
sourdough bread33 ÷ 13 × 10023.08%YesAbove 2.5%
baked daily11 ÷ 13 × 1007.69%YesAbove 2.5%
delivery0not present0.00%NoNot found

The result reads as an editing brief. The first target is repeated heavily for such a short text. "Baked daily" appears once but still sits above the band, a reminder that small samples produce large percentages. "Delivery" is absent, so a page about bread delivery would need that word near the top.

Exporting the numbers

Copy results as CSV produces the following lines for this example. Summary rows leave the density column blank, and target rows carry the band in the note column:

Section,Term,Count,Density %,Note
Summary,Total words,13,,
Summary,Unique words,9,,
Summary,Sentences,3,,
Summary,Reading time,4 s,,
Target,sourdough bread,3,23.08,In first 100 words; Above 2.5%
Target,baked daily,1,7.69,In first 100 words; Above 2.5%
Target,delivery,0,0.00,Not in first 100 words; Not found
1-word,bread,3,23.08,
1-word,sourdough,3,23.08,
2-word,sourdough bread,3,23.08,

Tips and Common Mistakes

Chasing a percentageWriting toward an exact figure produces copy that readers notice and dislike. Use the band as a warning light and then read the paragraph aloud.
Missing from the openingIf the first 100 words never mention the subject, a skimming reader and any preview built from the page may miss the point.
Counting the templatePasting a full page counts menus, cookie notices and footers with your article. Strip tags, or paste only the article body.
Forgetting the other placesTitle tags, headings, image alt text and the URL slug carry more weight than one extra repeat in paragraph nine. Check them after the body copy is right.

Frequently Asked Questions

Which words count as stop words?

The list is a short English set built into the tool. It holds articles, pronouns, common prepositions and conjunctions, a handful of auxiliary verbs such as is, are and was, and a few filler words such as get and via. It is deliberately small and cannot be edited, so for other languages turn the stop word option off and read the tables with every word included.

Why do my numbers differ from another analyser?

Analysers differ on how they split sentences, whether they strip markup, what counts as a word, and whether hyphenated or apostrophe words are one unit. This tool keeps letters and digits together, and keeps an apostrophe or hyphen inside a word, so don't and well-known each count once. Compare the method as well as the number.

Does the tool count alt text or meta descriptions?

Not when Strip HTML tags is ticked. Tags are removed as a whole, so alt attributes, title attributes and meta descriptions sit inside them and are left out. Check those separately in the page markup, because search engines and screen readers rely on them.

Can I keep the results?

Yes. Copy results as CSV puts the summary, target rows and frequency rows on the clipboard, and Download CSV saves them as keyword-density.csv. Your text is also remembered in this browser's local storage, up to about 200 KB, until you press Clear.

Open Keyword Density Analyzer