Live counts for words, characters with and without spaces, lines, paragraphs and sentences, with a reading-time estimate. Your text stays in your browser.
Words are runs of characters separated by whitespace, so well-known and don’t each count once.
Waiting for text
Paste or type below. Everything is measured in your browser.
Word frequencies appear here once there is text to count.
Word Counter runs entirely on your device — your writing is never uploaded.
You’re on the Free plan. Compare plans
A word here is a run of characters with whitespace on either side. "well-known" is one word, "don't" is one word, and "end." is one word with a full stop attached. That is the definition word processors use and the one nearly everyone expects, and it is the definition this tool applies everywhere — there is no second, cleverer rule applied quietly somewhere else.
It has one honest limitation, and it is worth knowing before you rely on a number: languages written without spaces between words — Chinese, Japanese, Thai — come out as very few words, because there is no whitespace to count. Splitting those properly needs language-specific segmentation, and a tool that did it for some scripts and not others would give you two incompatible definitions under one label.
Characters are counted as a reader sees them rather than as a computer stores them. An emoji, a flag, a letter with a combining accent: each is one character here, even though each is several units of storage. The Character Counter breaks those numbers apart if you need the storage figures.
Sentences are approximate and labelled as such. They are counted by looking for full stops, question marks and exclamation marks, which over-counts abbreviations like "Dr. Smith" and decimal numbers. Reading time is an estimate too, at 238 words per minute for silent reading — an average from published reading-rate research, not a measurement of you.
Everything is counted on your device as you type. Nothing is uploaded, which is what makes it safe to paste a draft you have not published.
As a run of non-whitespace characters. Text is trimmed, then split on any whitespace — spaces, tabs, line breaks — and what is left between the splits is counted. Hyphenated words and contractions count as one; a word with punctuation attached counts as one.
For ordinary prose, yes. The definitions diverge on unusual input: some editors count a number followed by a unit differently, and some treat em-dashes between words as separators. Nothing here is a certified count, and no two tools agree on every edge case.
Because those scripts do not separate words with spaces, so whitespace-based counting sees a whole line as one word. This is a known limitation rather than a bug, and it is why the tool states its rule rather than implying a universal one.
Words divided by 238 per minute, which is a published average for silent reading of English prose. It is a rough estimate: reading speed varies enormously with the material and the reader, and technical writing is much slower than fiction.
Sentences are found by looking for . ! ? and their full-width equivalents, so "Dr. Smith arrived." counts as two. Recognising abbreviations reliably needs a dictionary per language, so the number is labelled approximate instead.
No. The counting happens in your browser, and the text is never sent, logged or stored. Closing the tab discards it.
