Word Counter
Live words, characters, sentences, paragraphs, reading time, and keyword density in one panel.
Type or paste text into the box and every number on this page updates as you go: word count, character count, sentence count, paragraph count, estimated reading and speaking time, and keyword density for whatever term you're checking for overuse. It's built to answer the question people actually have — "is this the right length?" — without making you copy your draft into a separate app first.
Most writers land here for one of three reasons: a submission form with a hard word limit, an assignment with a required range, or a nagging suspicion that a paragraph has gotten away from them. The counter doesn't judge your writing — it just gives you the numbers so you can decide.
Stats appear here the moment you start typing.
0
Words
0
Characters
0
Chars (no spaces)
0
Sentences
0
Paragraphs
under a minute
Reading time
under a minute
Speaking time
How the count actually works
Word count here is computed by splitting on runs of whitespace and counting the resulting non-empty tokens. That sounds simple, but it means a few things are worth knowing: a hyphenated compound like "well-known" counts as one word (no whitespace inside it), while an em-dash-joined phrase without spaces around it can merge two words into one token if there's no space on either side. Contractions like "don't" count as a single word, matching how word processors like Microsoft Word and Google Docs treat them.
This is also exactly why two different word counters can disagree on the same text: some tools split on punctuation as well as whitespace (which would count "well-known" as two words), and some strip out digits or standalone symbols before counting. There's no universal standard — style guides and word processors have each made their own call. This tool uses the plain whitespace-split method because it matches what most publishing platforms, LMS assignment checkers, and word processors report.
Reading time is calculated from the word count divided by an average adult silent-reading speed (roughly 200-250 words per minute, depending on text difficulty — see our /methodology/ page for the exact figure and its source). Speaking time uses a slower rate, closer to 130-150 words per minute, because spoken delivery includes pauses that silent reading doesn't.
Who uses this
- Checking a cover letter or email against the informal rule of thumb that either should stay under roughly 300-400 words.
- Hitting an exact word range required by a journal, contest, or college application essay prompt.
- Timing a speech or video script using the speaking-time estimate rather than the faster reading-time one.
- Spot-checking keyword density on a page draft before publishing, without installing an SEO plugin.
- Comparing a rough draft against a finished revision to see how much a heavy edit pass actually trimmed or expanded the piece.
Edge cases to know about
- Numbers and standalone punctuation (like a lone "-" or "...") count as words if they're separated by whitespace, because the splitter has no way to know your intent — this can slightly inflate counts on text with a lot of numerals or ellipses.
- Pasted text from Word or Google Docs sometimes carries non-breaking spaces (Unicode U+00A0) instead of regular spaces, especially around numbers. Most modern splitters treat these as whitespace too, but if you ever see a suspiciously low count on pasted text, that's usually the cause.
- Very long single "words" (like a URL with no spaces) still count as exactly one word, even though they might be 80 characters long — word count and character count intentionally measure different things.
- Text pasted straight from a PDF often includes hard line-wrap breaks in the middle of sentences, which don't affect word count at all (a line break isn't whitespace that splits words differently than a regular space would), but can affect paragraph count if you're also checking that on the same passage — the two metrics respond to different features of the same messy paste.
Working with words at the letter level? Our sister site SnagWord covers the game side of the same problem — unscrambling letters, Scrabble/Wordle solving, and dictionary lookups — for when a tool like this one turns up a word you want to dig into further.
Related tools
FAQ
- Why does this word counter give a different number than Microsoft Word?
- Word processors and web tools don't share a single counting standard. Most, including this one, split on whitespace and count non-empty tokens, treating hyphenated words and contractions as one word each. Differences usually come from how each tool handles standalone punctuation, digits, and non-breaking spaces from pasted formatting — not from a bug in either tool.
- Does the count include the title or headings I paste in?
- Yes — every character you put in the box is counted, including headings, captions, and footnotes if you paste them. If you need a count of body text only, remove headings before pasting or count them separately.
- How accurate is the reading time estimate?
- It's an average, not a guarantee. Reading speed varies by individual, by how technical or dense the text is, and by whether you're skimming or reading closely. Treat the figure as a reasonable estimate for planning purposes (e.g. "this will take about 4 minutes to read"), not a precise measurement — see /reading-time-calculator/ for the fuller breakdown and /methodology/ for the sourced WPM figure.
- Why do word count and character count sometimes tell opposite stories about length?
- Because they measure different things. A passage full of short common words has a high word count but a modest character count; a passage full of long technical terms can have a lower word count but a higher character count for the same visual length. Check whichever metric matches the actual limit you're working against — a platform's word cap versus its character cap are genuinely different constraints.