Translated pages cannot rank in Google. Killed by CANONICAL. Content Localization + ?tl=

Support
,
1

Hello dear community!

I’m running a multilingual Discourse forum in English, Russian, and Italian using Content Localization with content_localization_crawler_param enabled.

AI translations are working well. Crawlers receive translated content via ?tl=, and the forum correctly outputs hreflang tags for those URLs.

However, on translated pages like ?tl=ru, the canonical still points to the base URL instead of to itself.

That seems to create an SEO conflict.

Google’s hreflang documentation says not to canonicalize to a page in a different language, and many SEO sources say hreflang on non-canonical pages may be ignored.

So the concern is simple:

  • ?tl=ru is treated as a duplicate

  • only the base URL is canonical

  • the Russian page is not indexed as its own page

  • so its hreflang may have no practical effect

My main question is:

Is the intended goal of Content Localization + crawler param to let translated pages rank in Google for their respective languages?

If yes, shouldn’t ?tl= pages use self-referencing canonicals?

If not, then what is the purpose of outputting hreflang tags and serving translated crawler content on those URLs?

For comparison, Reddit uses a similar ?tl= approach, but their translated pages appear to use self-referencing canonicals, which is why they can rank in local-language search results.

3

Hey, agree with you. Each language variant should be self-canonical.

We’ll have a look.