How word count works: CJK, Latin, characters, and platform limits
"How many words is this?" "Is the tweet over the limit?" A seemingly simple count actually has several methods across languages and platforms. This explains the common conventions so you can match the requirement.
Chinese by character, English by word
Chinese, Japanese, and Korean do not separate words with spaces and are conventionally counted per character — one character is one word, which is how Word counts them. Latin scripts split into words on whitespace, so "hello world" is two words, not eleven.
For mixed text, the two are counted separately and added. So a passage mixing Chinese and English reflects both the character count and the word count, and the tool shows a breakdown on separate lines for easy checking.
Why there are two character counts
There are usually two character counts: with and without spaces. The first includes spaces and line breaks; the second strips all whitespace. Which to read depends on the rule — most social platforms, SMS, and form limits count characters with spaces.
Emoji are counted by Unicode code point, so one emoji is usually one character (some complex emoji built from several code points may count as more). When a requirement is unclear, glance at both the with-spaces character count and the word count and you will usually match it.
Reading time is only an estimate
Many counters give a reading or speaking time, derived from average speeds: reading at about 400 characters per minute for Chinese, 200 words per minute for English. Such figures are good for judging roughly how long an article takes to read or a speech runs aloud, but actual pace varies widely by person and material, so they are not a precise time promise. For speeches or voiceovers where timing matters, read it aloud once and time it.
More guides