Ecommerce Pagination SEO After rel=next/prev: What Actually Works in 2026

In March 2019, Google announced it had stopped using rel="next" and rel="prev" link annotations as a pagination signal — and that it had stopped using them years before the announcement. Most ecommerce SEO guides still reference these tags as required pagination markup. They aren't. They're harmless, but they're not doing anything for Google.

What replaced them is more important: a series of explicit decisions about which pages in a paginated set deserve to rank, how internal linking flows across pages, and how canonical tags either help or break category page indexing. Get these wrong on a 50,000-SKU store and Google indexes 5,000 pages it shouldn't, suppresses 1,500 pages it should, and traffic to category pages drops 20-40%.

The pattern below is what works in production for Shopify, WooCommerce, BigCommerce, and custom-platform stores in 2026.

The Core Decision: Should Page 2+ Be Indexed?

This is the question most pagination SEO mistakes flow from. Three options, each with different downstream implications:

Option A: Index all pages (recommended for most stores)

Pages 2, 3, 4... of a category each get indexed independently. Each page has a self-referencing canonical (<link rel="canonical" href="/collections/shoes?page=2" />). Internal links flow normally — clicking "Page 2" goes to /collections/shoes?page=2, which links to ?page=3, and so on.

Pros: Google can find every product. Deep products on page 47 aren't orphaned.

Cons: Pagination pages compete with each other on long-tail queries. Doesn't matter much if titles/H1s are well-differentiated.

This is the right default for stores with >100 products per category. Use it unless you have a specific reason not to.

Option B: Noindex page 2+, but follow links

Page 1 is indexable. Page 2+ has <meta name="robots" content="noindex,follow" />. Google can crawl through to deeper products but won't index the paginated pages themselves.

Pros: Reduces index bloat on category pages with thin or near-duplicate content (e.g., the only difference between page 2 and page 3 is product order).

Cons: noindex,follow is unreliable long-term. Google has stated that consistent noindex eventually becomes noindex,nofollow in their treatment. Products on deep pages can become orphaned.

Use for stores with extreme product counts (10,000+ per category) where indexing every page is genuinely wasteful.

Option C: Single-page "View All" with canonicals

Build a single-URL "View All" page that contains every product in the category. Paginated URLs canonical to the View All page.

Pros: Concentrates ranking signals on one URL.

Cons: Anti-pattern in 2026. A View All page with 5,000 products is unloadable on mobile, fails Core Web Vitals, and Google has explicitly stated it discounts canonicals from paginated pages to View All pages when the View All page is significantly slower than individual pages.

Skip this. It was the right answer in 2014. It isn't now.

What to Actually Do With rel=next/prev

Bing still uses them. Some other search engines reference them. Adding them is cheap and unambiguously correct markup:

<!-- On /collections/shoes?page=2 -->
<link rel="prev" href="/collections/shoes" />
<link rel="next" href="/collections/shoes?page=3" />

But: do not rely on them. They are not a signal Google will act on. Implement them for completeness, then focus your effort on the things that do matter.

Canonical Tag Strategy

The single most-broken thing on ecommerce category pagination is canonical tags. Two failure modes recur in StoreVitals audits:

Failure 1: Every paginated page canonicals to page 1

<!-- On /collections/shoes?page=3 -->
<link rel="canonical" href="/collections/shoes" /> <!-- WRONG -->

This tells Google "page 3 is a duplicate of page 1, don't index it." Google obeys. Products that only appear on page 3+ become orphaned because the only URLs that link to them are pages Google has been told not to index.

Fix: self-referencing canonical.

<!-- On /collections/shoes?page=3 -->
<link rel="canonical" href="/collections/shoes?page=3" />

Failure 2: Paginated URLs have no canonical at all

If pagination uses query strings (?page=2), Google may interpret them as filter parameters and consolidate them under the base URL. Without an explicit self-canonical, the page may not be treated as a distinct URL.

Fix: every page in the paginated set has an explicit self-referencing canonical.

Sort Order and Filter Interaction

Category pagination interacts with sorts and filters in ways that frequently cause index bloat:

• /collections/shoes?sort=price-asc&page=2
• /collections/shoes?sort=price-desc&page=2
• /collections/shoes?sort=newest&page=2
• /collections/shoes?color=red&page=2

Each combination creates a distinct URL that Google may discover and try to index. For a category with 5 sort options × 10 filter values × 20 pages, that's 1,000 URLs per category.

Strategy:

• Sort parameters: canonical to the default sort, no index. Sort order doesn't change which products are in the set, just the order. These should all consolidate to the unsorted URL.
• Filter parameters: index strategically. Filters that produce useful landing pages (e.g., "red shoes") should be indexable. Filters that produce thin or empty pages (e.g., "red shoes under $20 in size 10.5") should be noindexed.
• Pagination on filtered pages: noindex by default. "/collections/shoes?color=red&page=2" rarely deserves to rank.

Internal Linking Patterns That Matter

For paginated category pages, three internal linking patterns affect SEO outcomes:

1. Pagination link numbering depth

If your pagination only shows "1, 2, 3 ... Last" (skipping middle pages), Google can't easily crawl to page 47 from page 1. Add at least "1, 2, 3, 4, 5 ... Last" with the ellipsis expanding context-appropriately. Alternative: add a "Show 100 per page" option, which collapses pagination depth.

2. Cross-linking between page 1 and last page

Adding a link from page 1 to "last page" gives Google a single hop to discover the deepest products. Worth doing on stores with 20+ pagination depth.

3. Linking from product pages back to category

Every product page should have breadcrumb navigation linking to its parent category. This is the highest-leverage internal linking pattern for category SEO and is unrelated to pagination — but worth mentioning because pagination doesn't matter if Google can't even reach the products it paginates.

XML Sitemap Strategy

Do not include paginated URLs in your XML sitemap. Sitemap entries should be canonical URLs you want indexed. Page 2+ being indexed is fine; advertising them in sitemap.xml is overkill and bloats your sitemap budget.

Do include: every individual product URL, every category landing URL (page 1 only), every brand landing URL, every editorial content URL.

The Pagination SEO Audit Checklist

1. Every paginated URL has a self-referencing canonical (not a canonical to page 1)
2. Pagination is reachable via crawlable links (anchor tags with href, not JavaScript-only)
3. rel=next/prev present but treated as Bing-only signal
4. Sort parameters canonicalized to default sort
5. Filter parameter strategy documented (which are indexable, which are not)
6. Paginated URLs not present in XML sitemap
7. Pagination link depth allows reaching last page within 3-4 clicks
8. Breadcrumb links from products back to category
9. Page 1 H1 differs from page 2+ (e.g., "Men's Running Shoes" vs. "Men's Running Shoes — Page 2")
10. Page 2+ titles differ similarly (avoid identical title across all pages)

Pagination SEO is one of the most-broken areas of ecommerce technical SEO because the conventional wisdom from 2014-2019 is still being copy-pasted into platform templates. The patterns above reflect what actually works in 2026 — self-canonical pagination, indexable page 2+ for most stores, smart sort/filter consolidation. StoreVitals scans validate canonical correctness, pagination link reachability, and parameter-URL bloat across category pages so the strategy stays implemented as the catalog grows.