Do hreflang tags really need to point to the canonicals of paged pages?

What does Google mean by "canonical versions" in the context of hreflang?

The canonical tag indicates to Google which URL represents the main version of duplicate or similar content. When a page 2 of a product list has a canonical pointing to page 1, Google considers page 1 as the reference.

However, Google stipulates that hreflang should point to these canonical references, not to the paginated pages themselves. If your page /products?page=2 has a canonical to /products, then the hreflang on this page should target /products (and its language equivalents), not /products?page=2 in other languages.

Why does this rule pose a problem for SEO practitioners?

Most CMS generate automatic hreflang based on the current URL. A page /fr/produits?page=2 will produce a hreflang to /en/products?page=2, which seems logical from a user perspective: an English visitor on the French page 2 likely wants to see the English page 2.

Except that if these paginated pages all have a canonical pointing to their respective page 1, Google receives a conflicting signal: the hreflang says "this FR page is equivalent to this EN page," but the canonicals say "these two pages are not the reference versions." Google then has to decide which signal to prioritize, and according to this statement, it is the canonical that takes precedence.

What’s the reasoning behind this directive?

Google seeks to link together the reference versions of content for each language or region. If you indicate that a paginated page is not the reference (via canonical), it would be inconsistent to simultaneously state that it represents the official French version to link with its equivalents.

This approach aims to concentrate multilingual signals on the canonical URLs, those that Google actually indexes and ranks. Non-canonical paginated pages are treated as auxiliary content, not as standalone SEO entities in the hreflang graph.

• Hreflang must always target canonical URLs, even if it breaks the exact match between paginated pages.
• If a paginated page has a different canonical, the hreflang should point to that canonical and its language equivalents.
• This rule applies regardless of user logic: what matters is the consistency of technical signals sent to Google.
• A multilingual site with complex pagination must revisit its canonicalization strategy to avoid conflicts with hreflang.
• The goal is to concentrate international SEO equity on the right URLs rather than dispersing it across paginated variants.

Is this directive consistent with the technical realities of e-commerce sites?

On paper, Google’s logic holds: do not create hreflang links between pages you yourself declare as non-canonical. But this rule poses a major dilemma for e-commerce sites that heavily paginate their categories and need precise language versions for each list segment.

In practice, many sites follow a mixed approach: self-referential canonical on each paginated page (page 2 canonical to itself), allowing consistent hreflang between pages 2 FR and EN. This statement by Mueller suggests that this pattern is acceptable, since the page is its own canonical. The real problem arises when all pagination is consolidated to page 1 via canonical.

What gray areas remain in this recommendation?

Mueller does not clarify how to handle cases where pagination serves truly distinct content: a blog page 2 may display totally different articles from page 1, with a legitimate user intent for each segment. Consolidating everything to page 1 via canonical risks losing indexing for these secondary contents. [To be verified]: Does Google actually index paginated pages with consolidated canonical, or does it completely ignore them?

Another blind spot: sites using rel="next"/"prev" (officially deprecated but still treated by Google in some contexts) combined with self-referential canonicals. Google documentation remains vague regarding the interaction between these multiple signals and hreflang. Field tests show variable behaviors across sectors: some sites see their paginated pages correctly alternating in multilingual SERPs, while others do not.

In which scenarios does this rule become problematic?

A classic case: a FR/EN/DE site with 50 product category pages per language. If all pages 2-50 have a canonical to page 1, and you follow Mueller’s directive, all hreflang tags on all paginated pages point to page 1. Result: you lose the benefit of accurate hreflang signals between equivalent segments, and Google may display page 1 EN to a user who was browsing page 15 FR.

Another trap: sites generating dynamically hreflang based on the current URL without checking the canonical. They end up with invalid hreflang that Google silently ignores, diluting the international signal. It then requires a complete overhaul of the template logic, which represents a significant technical endeavor on legacy CMS.

How to audit the current state of your paginated hreflang?

First step: extract all paginated URLs from your sitemap or via a Screaming Frog / Oncrawl crawl. Isolate those that have hreflang tags. For each, compare the URL of the canonical tag with the URLs targeted by the hreflang.

If you find discrepancies (hreflang points to /en/products?page=2 while canonical points to /en/products), you are in violation of this Google directive. Quantify the volume of affected pages: if it’s marginal, the risk is limited; if it affects thousands of strategic pages, it’s a priority project.

What canonicalization strategy to adopt to remain compliant?

You have two main options. Option A: self-referential canonical on each paginated page. Each page 2, 3, 4… becomes its own canonical. You can then maintain precise hreflang between pages 2 FR and 2 EN. This approach preserves granular indexing of pagination, but fragments PageRank across all pages.

Option B: consolidated canonical to page 1, and hreflang only on page 1. All paginated pages point to page 1 as canonical, and only this page 1 carries the hreflang. Pages 2+ have no hreflang. This approach concentrates international SEO equity but risks losing indexing of deep pagination pages. There’s no miracle solution: each choice comes with trade-offs depending on your business model.

What tools and processes should be implemented to monitor compliance?

Integrate automated tests into your CI/CD: a script that checks that for every URL with hreflang, the target of the hreflang matches the canonical (or that the page has a self-referential canonical). On WordPress or Shopify, some hreflang plugins can be reconfigured to read the canonical before generating the tags.

For continuous monitoring, set up Search Console alerts for hreflang errors ("Invalid URL", "Inconsistent Canonical"). Google does not explicitly report this canonical/hreflang divergence, but you will see a drop in hreflang coverage in the Internationalization reports if the signal is ignored. Cross-reference with your analytics: if international traffic on paginated pages drops without explanation, it could be a possible symptom.

• Crawl all multilingual paginated pages and extract canonical + hreflang for each URL.
• Identify inconsistencies: hreflang targeting a URL different from the declared canonical.
• Choose a strategy: self-referential canonical (broad indexing) or consolidated to page 1 (PageRank concentration).
• Adapt CMS templates or hreflang plugins to generate tags based on the effective canonical.
• Test on a sample of categories before global deployment, monitor international traffic and hreflang coverage.
• Document the chosen strategy to prevent regressions during future migrations or redesigns.