Writing guides
Word counts, characters and emoji explained
Different numbers answer different questions
Word count uses language-aware segmentation. Unicode code points measure encoded characters, while grapheme clusters approximate user-perceived characters. An emoji formed with joiners can be one visible cluster and several code points.
Select the text language
Choose the matching site language before counting. Intl.Segmenter handles word boundaries for that language, including scripts without ordinary spaces. Results can vary with browser Unicode data. Do not substitute a count from another language for a publisher’s prescribed method.
Understand lines and paragraphs
An empty document has zero lines. “Hello” has one line; “Hello” followed by a newline has two. CRLF is treated as one line separator. Paragraphs are blocks separated by a blank line rather than sentence or visual wrapping counts.
Meet a publisher’s requirements
A publisher may include notes, exclude headings or define characters differently. Count a copy of exactly the required text and compare the requested metric. The tool’s report describes its method; it cannot certify compliance with an external submission rule.