How to count words and characters and check a text limit

Choose the measurement your destination actually limits

A word limit and a character limit are different requirements. A form may count spaces, ignore them, or apply its own rules to emoji sequences. Read the destination’s instructions before shortening a draft. Kervoni shows words, Unicode characters, characters without whitespace and lines. It does not show sentence count, paragraph count or a reading-time estimate. The destination’s own counter is the final check when you submit text to that service; no independent counter can promise to reproduce every platform’s rules.

Select the language and count a copy of the text

Open the Word & Character Counter. Set Text language to the language of the draft, paste it into Input and select Count text. The language choice is independent of the interface language and is used for recognising word boundaries. Read the four result fields separately. If you edit the text, count it again because the previous results are cleared. Clear removes the working text and results from this interface. Keep the original draft elsewhere if you will need to restore it.

Verify a short example before a long document

Try two lines: “Hello world” on the first and “Hi 😀” on the second, with English selected. The result is 3 words, 16 Unicode characters, 13 characters without whitespace and 2 lines. The line break and spaces contribute to the full character count but not to the without-whitespace count. A line count follows explicit line breaks in the text. A long paragraph that wraps into several lines on your phone still has one logical line if it contains no line-break character.

Understand emojis and combining accents

The Unicode-character result counts code points, not every visible symbol as exactly one unit. A simple 😀 is one code point, while a family emoji joined from several people can contain several code points even when displayed as one symbol. An accented letter stored as one precomposed character can also count differently from a letter followed by a separate combining accent. The input field’s limit indicator and the result use different string measurements for some characters. If exact compatibility matters, test a small sample in the receiving system.

Review the final version and its privacy context

Count the exact version you will submit, including any heading, signature or line breaks that the destination includes. Whitespace removal in the displayed statistic does not change the source text. Word recognition can differ between languages, browsers or another service’s method, especially around punctuation and compound words. Kervoni processes the text locally without sending the input or output to a server for counting. Pasting the draft later into another website is a separate action governed by that website’s data handling. Recheck after every final edit.

Sources

• MDN: Language-sensitive word segmentation
• MDN: Iterating Unicode code points