Word Counter
This page explains how Locize counts words for billing and reporting.
Rules (how a “word” is detected)
A word is a combination of letters, punctuation marks and/or special characters (e.g. :@ # $ % ^ & * – _ \ ‘ “`) followed by a space or a newline.
In practice, Locize counts words by splitting text on spaces and newlines.
Important notes
- Tokens made of only punctuation/special characters are not counted as words.
- URLs (e.g.
https://www.locize.com) count as 1 word. - Email addresses (e.g.
support@locize.com) count as 1 word. - Placeholders like
{0}or%company%are counted as words when they appear as separate tokens.
What is counted
- Only values are counted, never keys. Empty values count as 0 words.
- Words are counted in every language: the source language plus each target language.
- A language without values adds nothing. A fallback to another language happens at runtime in your app (e.g. i18next
fallbackLng), so nothing is stored or counted for it. - If you use several versions, only the version with the most words counts, not the sum.
- Automatic translation is on by default for new projects: the words it fills into target languages count like any other.
Other examples of how words are counted
| String | Words |
|---|---|
| Number is -123.45 | 3 |
| 0 – 1 at 2 | 4 |
| two-in-one | 1 |
| 2-in-one | 1 |
| two-in-1 | 1 |
| hello?world | 1 |
| hello ? world | 2 |
| ☂ ☃ ☀⚤ | 0 |
| © %company% | 1 |
| 01/01/1980 | 1 |
| 21. 01. 2023 | 3 |
| Monday, August 8, 2011 | 4 |
| https://www.locize.com/blog/react-i18next | 1 |
| Let’s look | 2 |
| Word(s) | 1 |
| Decouple software release from the translation work. | 7 |