Pagination creates multiple URLs for content that is logically a single entity — a category listing split across 50 pages, an article divided into chapters, or search results with numbered pages. Each paginated URL can attract links and crawl budget independently, fragmenting signals that would be stronger concentrated on a single URL.
Google’s approach to pagination has evolved significantly: the deprecated rel="prev" / rel="next" annotations were officially retired in 2019. What remains is a set of practical decisions about how to structure paginated content for maximum SEO impact.
What Google Does With Paginated URLs
Without explicit signals, Google treats each paginated URL as a potentially independent page. /category/shoes/ and /category/shoes/page/2/ are two separate URLs that Google may index, crawl, and rank independently — or consolidate, depending on canonicalization signals and content overlap.
The risk: if users and links point to /category/shoes/, but deep pages of the category are indexed, Google may distribute ranking signals across all paginated variants rather than concentrating them on the canonical listing page. For most sites, the first page of a category listing is the one that should rank.
The View-All Pattern
The historically preferred approach for paginated article content: offer a single “view all” URL containing the complete article, and use canonical tags on paginated pages pointing to the view-all URL.
<!-- On /article/page/1/, /article/page/2/, etc. -->
<link rel="canonical" href="https://example.com/article/" />
The view-all URL becomes the canonical target. Paginated pages serve user navigation but pass link equity to the view-all. This consolidates ranking signals.
Limitation: if the view-all page is very large (many thousands of words or hundreds of images), it may have slow performance — affecting Core Web Vitals. A very large view-all page may also not render fully for Googlebot. For article content up to ~10,000 words, view-all + canonical on paginated pages is clean.
Category and Listing Pagination
For category listings (e-commerce, blog archives), the practical options:
Canonical page 1 only: Point all /category/page/N/ canonicals to /category/. This concentrates link equity on the primary listing page. Products/posts on deeper pages can still be crawled via the paginated URLs (Google doesn’t need to follow canonical to discover content).
Allow all pages to be indexed independently: Each page ranks for its specific content. Works when individual paginated pages have meaningful unique content. Less practical for standard category listings where page 2 isn’t meaningfully different in intent from page 1.
Noindex deeper paginated pages: Apply <meta name="robots" content="noindex, follow"> to pages 2+. Google crawls but won’t index them — products/posts are still discovered, but only page 1 appears in search results. Effective for preventing pagination from diluting ranking.
The choice depends on content type. For a blog archive where /blog/page/5/ would never rank meaningfully, noindex on deeper pages is efficient. For an e-commerce listing with hundreds of products, the tradeoffs are more nuanced — depending on whether individual products appear in search.
Infinite Scroll
Infinite scroll loads additional content dynamically via JavaScript as the user scrolls. From a crawl and indexation standpoint:
Googlebot can render JavaScript, but dynamic content loaded on scroll events may not be fully rendered. If products/posts are only accessible via scroll-triggered JavaScript, Googlebot may never see them.
Recommended pattern for infinite scroll + SEO:
- Implement paginated URLs that are also accessible via standard navigation (
/page/2/,/page/3/) - On the first page, include a link to
/page/2/(even if hidden visually for scroll UX users) - This ensures Googlebot can discover all content via crawlable href links
- The paginated URLs can then use canonical, noindex, or be left indexable based on your strategy
Progressive loading triggered by scroll without URL changes and without fallback paginated URLs creates crawl dead-ends for content on deeper “pages.”
Parameter-Based Pagination
Some sites use URL parameters for pagination: /category/?page=2 or /products?sort=price&page=3. This is equivalent to path-based pagination from Google’s perspective, with one added complication: parameters can combine in ways that create combinatorial URL explosion.
For parameter pagination, GSC’s URL Parameters tool (Legacy Search Console) allows specifying how Google should handle specific parameters — treating them as “changes page content” or “doesn’t affect page content.” This is supplementary to canonical and noindex implementations.
Pagination and Crawl Budget
Deep pagination consumes crawl budget. A 500-page category listing with 50 items per page has 500 paginated URLs for Googlebot to crawl — and the vast majority of those pages have no direct SEO value. For large sites, using noindex on deeper paginated pages and implementing internal links that directly point to products/posts from the primary page (via sitemaps and site architecture) is more crawl-budget-efficient than relying on Googlebot to discover content via pagination.