A word counter is the everyday tool of writers, students, and editors who work to length limits. This counter is tuned for English: it understands contractions and hyphenated compounds, and it reports unique words so you can gauge vocabulary richness and repetition, not just raw length.
How it works
Words are extracted with a pattern that treats internal apostrophes and hyphens as part of a single token, then the other statistics are derived:
words = tokens matching letters/digits with internal ' or -
uniqueWords = distinct lowercased tokens (count of the set)
sentences = splits on . ! ? followed by space or end
paragraphs = splits on one or more blank lines
characters = total length; also a no-whitespace variant
Treating don’t and well-known as single words means the totals match how a human would count them, which simpler whitespace splitting gets wrong.
Tips and example
For the text “It’s a well-known fact. Isn’t it?” the counter reports 6 words — because “It’s”, “well-known”, and “Isn’t” are each one word — across 2 sentences. The unique-word count is handy for spotting overused terms: if a 500-word piece has only 180 unique words, you are repeating yourself, which is sometimes intentional for emphasis and sometimes a sign to vary the language. Use the no-whitespace character count when a platform limit, like a meta description, counts visible characters only.