Word Counter & Character Counter

What counts as a word or a character

Counting looks simple until the text is not plain English. A naive counter measures string.length, which counts UTF-16 code units: an emoji with a skin tone counts as four characters and a flag as four more, and an accented letter typed as a base letter plus a combining mark counts as two. This tool counts grapheme clusters instead, the units a reader sees as one character, so 👍🏽 is one character and so is é however it was typed.

Words are found with the browser's own word breaker (Intl.Segmenter) rather than by splitting on spaces. That matters for Chinese, Japanese and Thai, which are written without spaces between words: splitting on spaces would count a whole paragraph as one word, while the dictionary-based breaker finds the actual words. Numbers count as words, punctuation does not, and contractions such as don't are one word.

Characters

Every visible character plus spaces and line breaks, counted as grapheme clusters.

Without spaces

The same count with all spaces, tabs and line breaks removed, which is what many translation and essay briefs ask for.

Sentences

Found by the browser's sentence breaker, which knows that a full stop after Mr or in 3.5 does not end a sentence.

Paragraphs

Blocks of text separated by at least one empty line.

Lines

Every line break starts a new line, empty lines included.

Reading time and speaking time

Reading time assumes 238 words per minute, the average silent-reading speed for adults reading non-fiction found by a large 2019 review of reading studies. Speaking time assumes 150 words per minute, a comfortable pace for a talk or a voice-over. Both are estimates: dense technical text reads slower, and a fast presenter may reach 170 words a minute, but they are a good guide to whether a blog post is a three-minute or a ten-minute read, or whether a script fits its slot.

Character limits worth knowing

The limits panel compares your character count with the limits writers bump into most. Search engines do not cut titles and descriptions at an exact character count (they truncate by pixel width), but 60 and 160 characters are the safe working figures.

• X (Twitter) post: 280 characters for a standard account.
• Meta title: about 60 characters before search results cut it off.
• Meta description: about 160 characters on desktop results.
• SMS: 160 characters in the GSM-7 alphabet, but only 70 once the message contains an emoji or a character outside it, which the tool detects for you.
• Instagram caption: 2,200 characters.
• LinkedIn post: 3,000 characters.

Using keyword density without overdoing it

Keyword density is the share of all words taken up by a word or phrase. It is useful for spotting a word you repeat without noticing and for checking that a page actually talks about its subject, but there is no magic percentage: modern search engines reward text written for people, and a phrase stuffed in at 5% reads badly to everyone. Use the two- and three-word view to see the phrases a page is really about.

Phrases never run across a full stop or a line break, so the table does not invent pairs like end. The. The stop-word filter removes common English words such as the, and and of, and drops phrases that start or end with one; text in other languages is left unfiltered. To compare two drafts line by line, use the diff checker.