Exclude specific pages from translation entirely. When a page matches an exclusion pattern, Universally returns the original untranslated HTML — no strings are extracted, no AI processing occurs, and no word count is used.
This setting is configured in your project’s General Settings tab, under Exclude Pages. Enter one URL path per line.
How It Works
When a translation request comes in, the page’s URL path is checked against your exclusion patterns before any processing begins. If it matches, the original HTML is returned immediately.
This is different from CSS selector exclusion, which removes specific elements from a page while still translating the rest. Exclude Pages skips the entire page.
Pattern Syntax
Exact paths
Match a specific page by its path:
/terms-of-service
/privacy-policy
/admin
Trailing slashes are ignored — /about and /about/ are treated the same.
Wildcard patterns
Use /* at the end to match a path and everything under it:
/api/*
/blog/drafts/*
/internal/*
/api/* matches /api/v1, /api/users/123, and /api itself.
Examples
Common exclusions
/admin
/api/*
/terms-of-service
/privacy-policy
/cookie-policy
/sitemap.xml
/robots.txt
WordPress-specific
/wp-admin/*
/wp-login.php
/wp-json/*
/feed/*
Note: The WordPress plugin handles most of these automatically. You only need to add custom exclusions.
E-commerce
/checkout
/cart
/account/*
/order-confirmation
Documentation sites
/docs/api-reference/*
/changelog
When to Use
- Legal pages that must stay in the original language for compliance
- Admin or dashboard pages that are for internal use only
- API endpoints that return data, not pages
- Checkout flows where translated text could cause confusion with payment processors
- User account pages where personal data should not be processed
- Pages with mostly dynamic/JS content that won’t benefit from HTML translation
Things to Note
- Excluded pages still load normally — they are just not translated
- The exclusion is based on the URL path only, not query parameters or fragments
- Exclusion happens before any processing — no strings are extracted, no word count is used
- If you need to exclude just a section of a page (not the whole page), use CSS Selectors instead