Word count or character count? Check emoji, spaces, and CJK text

Match the count to the requirement

Use Words for a brief or assignment that sets a word limit. Use Characters when the destination specifies a character limit, and find out whether spaces count. A requirement about bytes or a technical string length cannot be checked by substituting the visible character total.

The Characters without spaces result excludes whitespace segments, including tabs and line breaks, rather than only ordinary spaces. Compare it with the total to spot hidden formatting. Do not remove whitespace merely to make a number smaller if it changes the copy you will submit.

• Writing brief: check word count and whether headings, captions, or references belong in the submitted scope.
• Form or caption: check the destination's own counter after pasting the final text.
• Database or API: use the documented unit, which may be bytes, code points, or UTF-16 code units.

Reproduce an emoji count

Paste the following line without an extra newline. In a browser with Intl.Segmenter, the displayed result is five characters, four without whitespace, and one word. The emoji is one grapheme cluster even though it combines multiple Unicode code points.

The five clusters are H, i, the space, the complete emoji, and the exclamation mark. JavaScript's string length for the same line is nine UTF-16 code units. The counter's character card is therefore unsuitable for predicting a limit specifically defined in those code units.

1. Paste the example into an empty Text Counter input.
2. Compare the three displayed counts with the expected results.
3. Delete the space: the character total becomes four while the count without whitespace stays four.
4. Paste your actual text into its destination and check that application's limit before submitting.

Hi 👩‍💻!

Characters: 5
Characters without spaces: 4
Words: 1
JavaScript UTF-16 length, for comparison: 9

Treat accented letters and emoji sequences carefully

An accented letter can be represented as a single code point or as a base letter followed by a combining mark. For example, Cafe followed by a combining acute accent contains four grapheme clusters but five UTF-16 code units. A screenshot cannot tell you which representation the underlying text uses.

This matters when investigating a limit mismatch. Compare the exact copied text, including trailing newlines, instead of retyping something that merely looks the same. Keep the original until you know which measure the destination uses; deleting combining marks or part of an emoji can damage the content.

Check the language when counting words

Splitting at spaces is not a dependable word-count method for Chinese or Japanese. The tool uses locale-aware word segmentation when Intl.Segmenter is available. The Unicode segmentation guidance explains why reliable boundaries in these languages need language-specific treatment.

Do not treat CJK as one uniform counting rule: Korean commonly uses spaces, while Chinese and Japanese text often places multiple words together without them. A character count can be useful for a specification explicitly written in characters, but it does not become a word count simply because spaces are absent.

For mixed-language copy, use the language version that matches the main text and compare any required total with the destination's rules. Segmentation can vary with the browser's language data. Record the method when delivering a count to someone else rather than implying every application must produce the same total.

Resolve a mismatch before cutting useful text

First compare identical input and the same inclusion rules. Then check whether the mismatch is words versus characters, whitespace inclusion, or visible characters versus a storage unit. In browsers without Intl.Segmenter, this tool falls back to code-point character counting and a simpler word method; complex emoji and words without spaces may differ.

Reading and speaking times are estimates, not another way to enforce a word limit. This tool uses 225 words per minute for reading and 155 for speaking, rounded up to a tenth of a minute. Rehearse aloud when a delivery time matters, especially with pauses, unfamiliar terms, or multilingual passages.