Word counter — words, characters and reading time as you type
Paste or type, and the counts update on every keystroke. Words, characters with and without spaces, sentences, paragraphs, reading and speaking time, keyword density, and how close you are to the limit for a title tag, a meta description or a post.
- Nothing you paste is uploaded or saved
- Works offline
- Unique words
- 0
- Speaking time
- 0 sec
- Avg word length
- 0.0 chars
- Avg sentence
- 0.0 words
- Lines
- 0
- Longest word
- —
Add about 25 words of English and the Flesch score appears here. Below that, one long sentence swings the result by tens of points, so a number would mislead more than it would help.
Nothing to measure yet.
Measured in characters including spaces.
Google truncates on pixel width; 60 is the safe length
Beyond this Google usually rewrites or cuts it
Links always count as 23, however long they are
Past 160 the message is split and billed as two
Only the first ~125 show before "more"
Around 210 show before "see more"
How it works
- 1
Paste your text
Nothing is sent anywhere — the counting runs inside this page. That matters more than it sounds for a word counter, because the text people check is usually an unpublished draft, a cover letter, or a client document.
- 2
Read the counts
Words and characters update live. Characters are counted the way you see them, so an emoji counts as one even though it is several bytes underneath, and "characters without spaces" excludes line breaks and tabs as well as spaces.
- 3
Check it against a limit
The limit bars cover the lengths people actually write to: a 60-character SEO title, a 160-character meta description, a 280-character post, an SMS. The bar turns red the moment you go over, so you can trim while you write instead of after.
- 4
Tighten the writing
Keyword density shows what the text is really about — often a word you did not realise you repeated nine times. The Flesch score shows how hard it is to read; most web writing wants to land between 60 and 70.
Common length limits
All of these are character counts including spaces, and all of them are practical limits rather than hard ones — a platform that cuts your text off rarely tells you in advance.
| Where | Characters | Why that number |
|---|---|---|
| SEO title | 60 | Google truncates on pixel width; 60 is the safe length |
| Meta description | 160 | Beyond this Google usually rewrites or cuts it |
| X post | 280 | Links always count as 23, however long they are |
| SMS | 160 | Past 160 the message is split and billed as two |
| Instagram caption | 2,200 | Only the first ~125 show before "more" |
| LinkedIn post | 3,000 | Around 210 show before "see more" |
Frequently asked questions
Is my text uploaded or saved?
Neither. The counting is plain JavaScript running in this tab, there is no server involved, and nothing is written to your browser storage either — close the tab and the text is gone. This tool deliberately does not save drafts, because the text people paste into a word counter is usually something they have not published yet.
Why does this number differ from Microsoft Word?
Usually by one or two words, and usually because of hyphens or symbols. Word counts a run of characters between spaces as one word, and so does this tool — so "well-known" is one word in both. But a stray "—" on its own line is a word to some counters and not to others; here it is not, because it contains no letters or digits. If the gap is larger than a couple of words, it is almost always a header or footer that Word includes and a copy-paste leaves behind.
How are characters counted — is an emoji one character?
Yes, one. An emoji like a family or a flag is several code points joined together, and naive counters report it as 4, 7 or even 11 characters. This tool counts grapheme clusters, which is what a person sees as a single character. Worth knowing: X (Twitter) does not count this way — it weights characters and treats every link as 23 regardless of length — so treat the 280 bar as close, not exact.
What counts as a sentence?
A full stop, question mark or exclamation mark followed by a space or the end of the text — plus every line break, because headings and list items rarely end in punctuation and lumping them into the previous sentence would wreck the averages. Common abbreviations (Dr., Ltd., etc.) and initials like "J. K." are excluded so they do not split a sentence in two, and decimals like 3.5 or domains like example.com are left alone.
What counts as a paragraph?
Any line with something on it. Blank lines are ignored. This matches what you get pasting from Word or Google Docs, where each paragraph sits on its own line with no empty line between — counting only blank-line-separated blocks would report "1" for a ten-paragraph document.
Where does the reading time come from?
Words divided by 238 words per minute, which is the average silent reading rate for English prose from Brysbaert’s 2019 meta-analysis of 190 studies. Most sites quote 200 or 250 with no source. Speaking time uses 140 words per minute, the middle of the comfortable presentation range of 130–150 — useful for a talk or a script, where reading speed would badly underestimate you.
How is keyword density calculated?
Occurrences of a word or phrase, divided by the total word count. For phrases the percentage counts the words the phrase actually occupies: a three-word phrase appearing 5 times in a 500-word text is 3.0%, not 1.0%. Common function words (the, of, and) are excluded from single-word results, and phrases are dropped when they start or end on one — which removes "of the" while keeping "state of the art".
What does the readability score mean?
It is Flesch Reading Ease: 206.835 − 1.015 × (words ÷ sentences) − 84.6 × (syllables ÷ words). Higher is easier. 60–70 is plain English and a good target for the web; below 30 reads like an academic paper. The grade level next to it is Flesch–Kincaid, the number of years of schooling assumed. Both are calibrated for English, so the score is hidden for very short texts and should be ignored for other languages.
Does it handle Chinese, Japanese or other languages?
Character, sentence and paragraph counts work in any language. Chinese and Japanese are written without spaces between words, so each character is counted as one word — the same convention Word uses. Korean uses spaces, so it is counted like English. Keyword density and the readability score are English-only and will not tell you anything useful about other languages.
Is there a size limit?
No hard limit. Everything is a single pass over the text, so a full-length manuscript still counts instantly. Above roughly 200,000 characters the character count switches to a faster method that counts code points rather than grapheme clusters, which only differs if the text is full of emoji.