Understanding text counts
Words, characters, and the limits people care about.
Counting feels obvious until you ask exactly what you're counting.
Words — defined by whitespace.
A word, in this counter, is any maximal run of non-whitespace characters separated by whitespace. Don't contains one word; self-aware contains one word; the URL https://example.com/path also counts as one. That matches the conventional definition used by Microsoft Word and Google Docs. CJK languages need different rules — a single character is the meaningful unit there — and the Western definition undercounts for those scripts.
Characters with and without spaces.
The two character counts answer two different questions. With spaces is the literal length of the string — useful for limits like Twitter (280) or SEO meta descriptions (~160). Without spaces is the count of "ink" — useful when a contract specifies a minimum article length and you want to make sure you're not padding with whitespace.
chars (no spaces) = chars − count of whitespace
Reading time — a rough approximation.
The reading-time figure assumes 230 words per minute, the midpoint of typical adult silent-reading speeds. Real variation is wide: skim-reading hits 400+, dense academic prose drops to 150. Treat the number as a ballpark for "is this a 5-minute read or a 30-minute one?" rather than a stopwatch promise.
Common platform limits.
Twitter / X: 280 characters. Bluesky: 300. Mastodon: 500 default. SMS: 160 (one segment). SEO title tag: ~60 visible before truncation in search results. SEO meta description: ~160. Open Graph titles: ~60. LinkedIn post: 3,000. Knowing where the cliff is helps you write directly toward it.
Sentences and paragraphs.
Sentences are detected by the punctuation that ends them — . ? !. Abbreviations ("Mr.") and ellipses ("…") cause occasional miscounts; the counter doesn't try to disambiguate. Paragraphs are runs separated by blank lines. Both figures are useful for essay structure — a sentence longer than 35 words probably wants a rewrite.