← All guides

URLTO GUIDE

Why character counts differ: Unicode, emoji and bytes

One visible symbol can contain several stored characters. Review both the tool's counting rule and the destination's limits.

URLTO editorial · Updated

Text counts · series10 / 10

1. Identify the units the tool counts

URLTO counts Unicode code points for characters, removes whitespace for its compact count and uses UTF-8 encoding for bytes. Words are whitespace-separated groups, not language-aware linguistic tokens. Each number in the result serves a different kind of limit.

2. Compare examples that look similar but count differently

These values were verified with URLTO's counting function. The precomposed é uses one code point; the following row uses e plus a combining accent. Flags and family emoji combine several elements and therefore do not count like visible icons.

2. Compare examples that look similar but count differently
InputCode pointsUTF-8 bytes
가 A35
é12
é23
🇰🇷28
👨‍👩‍👧‍👦725

3. Whitespace and line breaks are part of the input

'가 A' has three code points, two without whitespace and five UTF-8 bytes. Trailing spaces and line breaks in copied text can change totals. Inspect the end and empty lines before editing the content. Do not remove meaningful spacing merely to alter a whitespace-based word count.

4. Review duplicates and publishing limits under their own rules

Deduplication compares line strings. It has trimming options but does not merge every case variant or visually equivalent Unicode representation. A publishing service can count emoji or characters differently. Validate near-limit text in its actual editor and keep a source copy before changing important content.

Frequently asked questions

Is it a bug when an emoji counts as several characters?

This tool counts code points. Combined emoji can contain several, so the count can differ from the number of displayed icons.

Does the count match every publishing service's limit?

Counting methods can differ. Check whether the limit concerns characters or bytes and verify it in the final publishing service.

Use now Character counter