Online Tool Store
· 4 min read
How to Handle Canonical Tags on Paginated Pages
Heshan Fernando
Co-founder & COO
A category listing with nine pages of products. Someone notices that pages two through nine look like near-duplicates of page one — same layout, same sidebar, same header — and “fixes” it by canonicalising all of them to page one.
Six weeks later, most of the products that only appear on pages two through nine have quietly dropped out of search results. The fix caused the problem.
What canonical actually says
A canonical tag tells search engines “this URL is a duplicate of that one; index that one instead”. It’s a strong instruction, and it means the content unique to the canonicalised page is treated as not needing to exist.
That’s exactly wrong for pagination. Page three of a listing isn’t a duplicate of page one — it has different products, different links, different content. Telling a crawler otherwise means the items on page three lose their route into the index.
The correct default is a self-referencing canonical: page three canonicalises to page three. Each page is its own thing, because it is.
rel="prev" and rel="next" are a separate matter. Google announced in 2019 that it no longer uses them as an indexing signal. Other crawlers and some assistive technology still read them, and they cost nothing, so including them is reasonable — just don’t expect them to do heavy lifting.
Why people get stuck here
- Duplicate-content anxiety. Paginated pages look similar, so canonicalising them feels like the responsible thing to do.
- Conflicting advice. Guidance from before 2019 treats rel prev/next as essential; guidance after treats it as obsolete. Both circulate.
- View-all pages. Sometimes the right canonical target, sometimes a page that takes twelve seconds to load.
- Parameters. Sorting and filtering parameters multiply the URL space, and pagination interacts with all of them.
What good pagination markup looks like
Self-referencing canonicals on every page
Page two canonicalises to page two, page three to page three. This is the default and it’s right for nearly every paginated listing.
rel prev and next as a bonus
Include them for the crawlers and tools that still read them. Don’t build your pagination strategy around them.
A view-all page only if it’s genuinely usable
If a view-all page exists and loads quickly, canonicalising the paginated pages to it is legitimate. If it’s a 400-item page that takes ten seconds, it fails users and search engines alike.
| Approach | When It’s Right | Risk |
|---|---|---|
| Self-referencing canonical | Nearly always | None |
| Canonical to page 1 | Almost never | Later pages drop out of the index |
| Canonical to view-all | View-all is fast and complete | Slow page harms both users and ranking |
Common mistakes to avoid
- Canonicalising every page to page one, the single most damaging pagination mistake.
- Adding
noindexto pages two onward, which eventually removes the links on those pages from consideration too. - Letting sort and filter parameters generate infinite paginated URL combinations with no canonical strategy.
- Making page numbers only reachable by JavaScript, so crawlers never see pages beyond the first.
- Changing pagination size frequently, which reshuffles which item sits on which page and destabilises everything.
How to do it with Pagination Tag Generator
The Pagination Tag Generator produces the head markup for a page in a series, with self-referencing canonicals by default.
- Enter the series URL pattern and which page you’re generating for.
- Choose the canonical strategy — self-referencing unless you have a genuinely fast view-all page.
- Copy the generated tags into that page’s head.
- Verify on the live site that page three’s canonical points to page three, not page one.
Other technical SEO tools are in the tools directory.
Frequently asked questions
Do rel prev and next still matter?
Google stopped using them for indexing in 2019. Other crawlers and some assistive technology still read them, and they’re cheap to include — treat them as useful markup rather than a ranking factor.
Should page 2 canonicalise to page 1?
No. That tells search engines page two’s content doesn’t need to exist independently, and the items unique to it lose their path into the index. Self-referencing canonicals are the correct default.
What about infinite scroll?
Provide real, crawlable paginated URLs behind the scroll. Infinite scroll with no underlying URLs means everything past the first screenful is invisible to crawlers.
Final thought
The whole of pagination SEO reduces to one rule: each page is its own page. Canonicalise it to itself, link it properly, and resist the urge to consolidate content that isn’t actually duplicated.