How the Word Counter — Words, Characters & Reading Time works

With Intl.Segmenter, which applies the Unicode word-boundary rules (UAX #29). Contractions like "don't" count as one word, and languages written without spaces, such as Chinese and Japanese, are segmented into real words instead of being counted as one long token.

Characters are counted as grapheme clusters — what a reader perceives as one symbol. A family emoji is built from several code points joined invisibly, so JavaScript's .length reports 8 or more, but you see one character and this counter agrees with you.

It uses 238 words per minute, the average adult silent-reading rate found by a 2019 meta-analysis of 190 studies. Your real audience varies: dense technical content can drop below 150 wpm, while skimming runs far faster. Treat it as a solid estimate, not a stopwatch.

No. Counting, density and every other stat are computed in your browser. The autosaved draft lives in this browser’s localStorage on your device, and the Clear button removes it.

There is no magic percentage. Use the table to spot accidental repetition and confirm the topic you meant to write about actually dominates — then stop. Optimising toward a density number is keyword stuffing, which hurts more than it helps.