301 chain301 chain
Permanent redirects stacked one after another.
Found by Redirect checker →
302302
A temporary redirect on a page that moved for good.
Found by Redirect checker →
Technical SEOUpdated

Canonical for Multilingual Sites

In short

Canonical preference for multilingual URLs separates true duplicates from language alternatives, which usually need their own canonical URL and hreflang.

Canonical and hreflang solve different problems

A canonical tells Google which URL represents a duplicate or very similar set. hreflang labels alternate language or regional URLs so Search can choose a suitable version for a user. These annotations complement one another; neither one turns a language selector into a redirect rule.

For genuinely translated English and Spanish pages, the usual page-level pattern is a self-referencing canonical on each URL and reciprocal hreflang annotations between the matching pages. The visible main content must make the language clear. Google’s multilingual guidance also cautions that language adaptation based only on browser settings can leave variants undiscovered.

The diagram highlights the main decision points behind Canonical for Multilingual Sites.

Build the alternate map before editing templates

Start with a table of page groups: one preferred URL per language or region, its language code, and whether the main content is a translation, a regional adaptation, or an accidental duplicate. Add an x-default only where a genuine selector or fallback page exists. Do not point every missing translation at a home page merely to complete a grid.

A regional pair in the same language deserves a separate decision. If the substantive content is equivalent, Google documents a combined canonical and hreflang approach for regional duplicates. If price, availability, regulation or local advice makes the pages meaningfully different, preserve each page’s own purpose and ensure every alternate is reciprocal.

Implement and verify the rendered annotations

Check the final HTML, not just a CMS field. Each canonical target should return a successful response, be indexable, and use the same preferred protocol, host and path policy as the page. Each hreflang target should exist, contain the same set of return annotations, and point to the matching language or region page.

A practical sample includes a normal translated page, a page without a translation, and a regional variant. Crawl the sample after deployment and compare canonical targets, hreflang reciprocity, language declarations and sitemap URLs. Record a missing alternate as missing; do not replace it silently with an unrelated URL.

What this cannot guarantee

Canonical is a signal, not a command: Google can choose another representative URL when signals conflict. Hreflang can help selection but does not guarantee that every regional URL will be indexed or appear for every query. Avoid claims that a cross-language canonical automatically removes a page from the index; the actual outcome depends on the content and the broader signal set.

The common failure is treating all non-default languages as duplicates of one source page. The safer rule is to decide similarity from the primary content, keep language alternatives reachable through ordinary links, and recheck the rendered set after any routing or template change.

FAQ

Canonical preference for multilingual URLs separates true duplicates from language alternatives, which usually need their own canonical URL and hreflang.
A canonical tells Google which URL represents a duplicate or very similar set. hreflang labels alternate language or regional URLs so Search can choose a suitable version for a user. These annotations complement one another; neither one turns a language selector into a redirect rule.