Pagination
Pagination is the practice of splitting a long list or body of content across numbered pages (1, 2, 3, and so on). In SEO, the core challenge is keeping every page crawlable and indexable while managing duplication.
- Pagination breaks item-heavy content such as product listings, search results, and blog archives into a sequence of numbered pages (1, 2, 3, and so on).
- From an SEO standpoint, the top priority is ensuring every page can be crawled and indexed so the items it contains are discoverable.
- Google no longer uses rel="next" and rel="prev" link annotations; it instead relies on crawlable
<a href>links between pages. - Give each page its own unique URL, and avoid pointing every page's canonical at the first page.
- Even with infinite scroll or incremental loading, expose a per-page URL so search engines can still find the content.
Overview
Pagination is a way of serving item-heavy content across multiple pages instead of loading everything into a single view. It is common on e-commerce product listings, search results, blog archives, and comment threads, and it usually appears as numbered links (1, 2, 3, and so on) near the bottom of the page. For users it reduces the load weight of any single page and makes browsing easier, but in SEO it requires deliberate handling.
The central concerns are crawling, indexing, and duplication. Search engines need to reach and discover the items on the second page and beyond, and each page must be recognized as a unique page even though all of them share a similar structure. Handled poorly, products or articles on deeper pages can drop out of the index, or unnecessary variant URLs can drain crawl budget.
Best Practices
Google Search Central documentation recommends the following for pagination.
- Use crawlable links: Link from each page to the next with a real
<a href>tag rather than a JavaScript button or onclick handler, so Googlebot can discover the subsequent pages. - A unique URL per page: Assign each page its own URL (for example
?page=2). Avoid handling the page number as a URL fragment (#), since Google ignores fragments. - Self-referencing canonical on each page: Have each page point its canonical at itself, rather than designating the first page as the canonical for all of them.
- Be careful with index blocking: Do not apply noindex to the pagination sequence itself. Reserve noindex selectively for the unnecessary variant URLs created by sorting and filtering.
- Handle infinite scroll: Even when using incremental loading or infinite scroll, expose a per-page URL, and remember that Google does not click or trigger user actions, so support content discovery through means such as a sitemap.
Industry guides from Ahrefs and Semrush point in the same direction. They also recommend a form of "de-optimizing the secondary pages": rather than aggressively optimizing key SEO elements such as the meta title and description on the second page and beyond, let only the first page target the primary keywords.
Current Status of rel=next / rel=prev
Historically, <link rel="next"> and <link rel="prev"> were used to signal the relationship between pages. In 2019, however, Google officially announced that it no longer uses these signals, and it revealed that in practice it had not been using them for some time before that. Google now infers the relationship through ordinary <a href> links between pages instead of these tags. That said, some search engines such as Bing may still consider them as a hint, so there is no harm in keeping them.
Implementation Checklist
- Confirm that next, previous, and numbered links are implemented as real
<a href>links. - Assign each page a unique URL (such as
?page=n) and avoid the fragment (#) approach. - Check that each page's canonical points to itself rather than funneling everything to the first page.
- Verify that no unnecessary noindex is applied to page two and beyond.
- If you use infinite scroll, provide the corresponding per-page URLs along with a sitemap.
- Manage the variant URLs created by sorting and filtering selectively with noindex or robots.txt.