Almost every guide treats ecommerce pagination as a ranking question — how do I get page 2 of a category to rank? That’s the wrong question, and chasing it is how stores quietly bury half their catalog. Page 2 of a paginated category is almost never meant to rank for anything. Its real job is to be a crawl path: a set of clean links that lets Googlebot walk from your category page down to product number 400 and back. Get pagination right and every product in a 5,000-SKU catalog is discoverable in a couple of hops. Get it wrong and Google indexes the first 24 products, shrugs at the rest, and you never find out why deep inventory earns zero impressions.
Pagination Is a Crawl-Path Problem, Not a Ranking Problem
The mental model that fixes most mistakes: paginated pages exist to distribute link equity and crawl access to deep product URLs. When a searcher lands on a category, they want the best products first. When Googlebot lands on it, it needs a discoverable route to every product. Those are different audiences with different needs, and the pagination system has to serve the crawler without cluttering the index. So the goal isn’t “rank /category?page=7” — it’s “make sure the products only reachable from page 7 get crawled, indexed, and internally linked.” Once you accept that, the technical decisions get much simpler.
rel=next / rel=prev Is Gone — Stop Building Around It
For years the canonical advice was to mark up paginated sequences with rel="next" and rel="prev" link tags. Google confirmed in 2019 that it hadn’t used those signals for indexing in years and formally retired the guidance. Bing still reads them as a weak hint, so leaving them in place does no harm — but building your entire pagination strategy around them is planning for a spec that no longer exists. Google now treats each paginated URL as a standalone page and works out the sequence from ordinary on-page links. That single fact changes how you should handle canonicals, which is where most stores go wrong.
The Canonical Mistake That Deindexes Your Catalog
Here is the most expensive error in ecommerce pagination, and it’s built into several popular platform themes by default: canonicalizing every paginated page back to page 1. It looks tidy — one canonical URL for the whole category, no “duplicate” pages in the index. But a canonical tag tells Google “the real version of this content lives over there,” so the products that only appear on pages 2 through 40 get treated as duplicates of page 1 and dropped from consideration. You’ve just told Google to ignore 95% of your inventory.
The correct setup: each paginated page self-canonicalizes. Page 2 canonicals to page 2, page 3 to page 3, and so on. Each is indexable (or at minimum crawlable), each carries a full set of internal links to its products, and Google is free to follow those links to depth. You’re not trying to rank the paginated pages — you’re keeping the crawl path open so the products they link to stay in the index.
Numbered Links, Load More, or Infinite Scroll?
The three common patterns differ almost entirely in whether a crawler can traverse them:
- Numbered pagination (real
<a href>links to?page=2,?page=3): the safest choice. Every page is a crawlable URL, link equity flows through the sequence, and Google walks it without JavaScript. Boring and effective. - “Load more” buttons: fine for users, dangerous for SEO if the button is a JavaScript event with no underlying URL. If clicking “load more” doesn’t change the URL and doesn’t expose real links, the additional products are invisible to a crawler that doesn’t click.
- Infinite scroll: the highest-risk pattern. Content injected on scroll typically has no crawlable URL at all, so a bot that never scrolls sees only the first batch.
None of these is banned — but two of them only work for SEO when you back them with real, unique URLs. That’s the mechanism the next two sections cover.
Making Load More SEO-Safe
The fix for load more SEO is to make the button a progressive enhancement over a working paginated URL structure, not a replacement for it. Practically: the page renders numbered pagination links in the HTML (?page=2 etc.) as real anchors; JavaScript then intercepts them to fetch and append the next batch in place for a smoother experience. A crawler with JavaScript disabled — or one that simply doesn’t click — still sees and follows the <a href> links to every page. Users get the seamless “load more” feel; Googlebot gets an old-fashioned crawlable sequence underneath. When the fallback links are absent and “load more” is a bare button, the products past batch one may as well not exist.
Infinite Scroll Without Losing Deep Products
Infinite scroll can be crawlable, but only if each scroll segment maps to a genuine, linkable URL. The technique Google documents is to pair infinite scroll with the History API: as the user scrolls into a new segment, the page uses pushState to update the address bar to ?page=2, and that URL, loaded directly, must render page 2’s products on its own. In effect you’re running numbered pagination that feels like infinite scroll. The failure mode to avoid is fragment URLs — anything after a # — because Google ignores the fragment, so ?page=2 works as a distinct page and #page=2 does not. If you can’t guarantee each segment has a real, directly-loadable URL, don’t use infinite scroll on indexable category pages; use it on search results you don’t need crawled instead.
The “View All” Option and When It Beats Pagination
If a category holds a few dozen to a couple hundred products and the page still loads fast, a single “view all” page can be the cleanest answer: one URL, every product, no crawl-depth problem at all. Google has historically favored serving a view-all page to searchers when it exists and performs well. The catch is performance — a view-all page with 800 products and 800 images will be slow, hurt Core Web Vitals, and frustrate users. Rule of thumb: view-all for small-to-medium categories where the full list stays fast; paginated sequences for large catalogs where a single page would be bloated. Don’t force one model across every category — match it to size.
Faceted Navigation: Where Pagination Explodes
Pagination rarely lives alone. The moment you add filters — color, size, price, brand — each filter combination can spawn its own paginated sequence, and ?color=blue&size=large&page=3 multiplies into millions of crawlable URLs that burn crawl budget on near-duplicate lists. This is the single biggest crawl-efficiency drain in ecommerce. The durable approach is to decide which facets deserve indexable landing pages (high-demand combinations like “waterproof laminate flooring” that people actually search) and which should be blocked from crawling entirely via robots rules or noindex. Paginated versions of low-value facet combinations should not be crawlable at all — you want crawl budget spent walking your real product pages, not the 400th shade of a filter no one searches.
A Worked Example: A 4,000-SKU Flooring Store
Picture a mid-size flooring store — the kind of catalog where this genuinely bites. The “Laminate Flooring” category holds 4,000 products at 40 per page: 100 paginated pages. If every one of those pages canonicals back to page 1, Google indexes roughly the first 40 products and treats the other 3,960 as duplicates — deep SKUs like a specific oak-effect board earn no impressions because Google can’t justify crawling to them. Fix the canonicals so each page self-references, keep numbered <a href> links, and add an XML sitemap listing all 4,000 product URLs directly. Now Google has two discovery routes: the paginated crawl path and the sitemap. Within a few crawl cycles, deep products start appearing in the index and picking up long-tail queries like specific model names and finishes. No new content — just an unblocked crawl path. (This is the exact failure mode we’ve seen and fixed across a portfolio of ecommerce sites, part of a playbook proven across 1,000,000+ ranking pages.)
How SEO Rocket Surfaces Pagination Problems
The hard part of pagination isn’t the theory — it’s spotting where your live store violates it. SEO Rocket’s site audit runs a real crawler over your catalog and flags the exact ecommerce failure modes: paginated pages canonicalizing to page 1, products orphaned with no internal path, near-duplicate variant URLs, redirect chains, and crawl-budget waste from faceted URL explosion. Because it crawls the way Googlebot does, it shows you which products are actually reachable versus theoretically listed. Pair that with rank tracking on your product and category terms and content-gap analysis on real Ahrefs data — the terms rivals rank for that you don’t — and you can see whether fixing a crawl path actually pulls deep SKUs into impressions. It’s an SEO layer that diagnoses and monitors, not a store platform that changes your templates for you; the fixes still land in your CMS, but you stop guessing which pages are the problem.
A Practical Pagination Checklist
- Each paginated page self-canonicalizes — never to page 1.
- Pagination links are real
<a href>anchors to unique?page=NURLs, present in the HTML. - “Load more” and infinite scroll sit on top of those crawlable URLs, never replace them.
- No fragment (
#) URLs for pagination — Google ignores the fragment. - Don’t
noindexpaginated pages if they’re the only crawl path to products. - Use “view all” for small categories that stay fast; paginate large ones.
- Block low-value faceted + paginated combinations from crawling to protect crawl budget.
- Back the whole thing with a complete XML product sitemap as a second discovery route.
Frequently Asked Questions
Should I noindex paginated category pages?
Usually not. If pages 2 onward are the only way Google reaches your deeper products, a noindex tag can, over time, cause Google to crawl those pages less and drop the products they link to. Leave them indexable (or at least crawlable with self-canonicals) unless you have a separate, complete crawl path — like a full product sitemap plus strong internal linking — that reaches every SKU independently.
Do I still need rel=next and rel=prev?
Not for Google — it stopped using them years ago and retired the guidance. Bing treats them as a mild hint, so keeping them costs nothing, but they should never be the backbone of your product pagination SEO. Self-referencing canonicals plus crawlable numbered links do the real work.
Does pagination hurt my crawl budget?
Plain numbered pagination is efficient and helps crawl budget by giving bots a clean route to deep pages. The budget drain comes from paginated faceted URLs — filter combinations multiplied across page numbers — which generate millions of near-duplicate lists. Block the low-value ones so crawling concentrates on real products.
Ecommerce pagination is unglamorous plumbing, and that’s exactly why it’s neglected and exactly why fixing it pays off. You’re not writing content or building links — you’re making sure the inventory you already have can be found. On a large catalog, a correct pagination and canonical setup is often the difference between 40 indexed products and 4,000. Audit your paginated categories, kill the page-1 canonical, keep the links real, and let the crawler do the rest.
