This is a misconception in reverse of how it’s usually assumed, but it’s worth being precise about what’s actually being asked here. The current, correct guidance is that each paginated page should generally self-canonicalize (canonicalize to itself, not to page 1), which is what the phrase “self-referencing canonicals” in this question describes as the practice being questioned. The older advice, once commonly given, that all paginated pages should canonicalize to page 1 of the series, has been superseded. So the misconception to correct here is actually the opposite one: it would be a mistake to still believe paginated pages should point their canonical to page 1 rather than to themselves.
Why self-referencing canonicals are the current, correct approach
Each page in a paginated series (page 2, page 3, and so on of a category listing, search results set, or article archive) typically contains genuinely different content from page 1, different products, different articles, different individual items depending on what’s being paginated. Canonicalizing every one of those pages to page 1 would tell Google that the specific items unique to page 2, 3, and beyond aren’t independently significant content worth considering on their own, effectively signaling that only page 1’s content matters. Since those later pages do contain real, distinct items that may have their own legitimate search demand (a specific product on page 4 of a category listing, for instance), collapsing them all into page 1’s canonical risks suppressing the discoverability of content that has no other representation in the index.
How the guidance evolved
Google’s approach to pagination canonicalization has shifted over time. In earlier years, common advice (including some earlier framing that circulated in the SEO community and in some of Google’s own guidance) suggested treating paginated series similarly to duplicate content, with all pages pointing to page 1 as the canonical entry point for the series. Google’s more recent, standing guidance, including public clarifications from John Mueller and Search Central documentation, has moved away from that framing, since it doesn’t accurately reflect how paginated content actually behaves: each page isn’t a duplicate of page 1, it’s a distinct subset of a larger content set, and treating it as a duplicate misrepresents what’s actually on the page.
Why the self-referencing approach makes more technical sense
A self-referencing canonical tells Google, accurately, that each paginated URL is its own valid, indexable entity representing its own specific slice of content. This preserves the discoverability of items that live only on later pages of a series, since nothing is telling Google those pages are redundant duplicates of page 1 that shouldn’t be independently considered. It also aligns with how Google actually processes paginated series: Google’s systems are capable of understanding pagination relationships (partly through crawling the linked sequence of pages, and historically through signals like rel=next/rel=prev, even though Google has since stated it doesn’t use those specific link relation tags as a ranking signal anymore) without needing every page to be artificially consolidated into a single canonical URL.
What this means practically
If a site’s paginated category or listing pages are currently set up with all pages canonicalizing to page 1, that’s the outdated pattern that should be corrected, each page should instead canonicalize to itself. The practical risk of leaving the old “all pages to page 1” pattern in place is that any items, products, or content pieces that exist only on page 2 and beyond may never get properly indexed on their own, since the canonical signal is actively telling Google those pages aren’t meant to stand independently. Auditing paginated series to confirm each page carries a self-referencing canonical tag (rather than inheriting an older, page-1-pointing pattern from a legacy template or an SEO plugin’s default configuration) is a reasonable technical check, particularly on larger e-commerce or content archive sites where paginated series are common and where a large volume of individually valuable content could be sitting on later pages with a canonical tag actively undermining its own indexability.
The bottom line
The misconception worth correcting is holding onto the older “canonicalize everything to page 1” advice as if it’s still current best practice. Google’s standing, current guidance favors self-referencing canonicals for paginated series specifically because each page typically represents genuinely distinct content that deserves its own indexing consideration, not a duplicate deserving consolidation into a single entry point.
