Pagination is where big directories either stay fully indexed or quietly lose half their catalogue. A category with two thousand listings spread across eighty pages is a navigation puzzle for a crawler. Solve it well and every listing gets discovered and ranked. Solve it badly and the listings on page sixty might as well not exist. The good news is that the right pattern is simple and durable once you understand what each signal is for.
What pagination has to achieve
A paginated set has two jobs that pull in slightly different directions. First, every individual listing must be reachable by a crawler, which means the deep pages of the set have to be genuinely accessible, not hidden behind JavaScript that never produces a real link. Second, the paged URLs themselves should not flood the index with near duplicate listing grids that compete with each other or with the category landing page. Balancing reach against bloat is the whole craft of pagination.
The pattern that works
Self referencing canonicals on every page
The single most important rule, and the one most often gotten wrong, is that each paged URL canonicals to itself. Page two points to page two, page three to page three. The old instinct to canonical every page back to page one feels tidy, but it tells Google that pages two onward are duplicates not worth indexing, and that is how the listings only present on deep pages vanish. Self canonicals keep the door open for the crawler to index the deep set. This is the same discipline we describe in canonical tags done right, applied to the paginated case.
Real, crawlable links between pages
The links that move a crawler from page one to page two and onward must be actual anchor elements with real href values pointing to real URLs. A next button that only works through an onclick handler, or a load more control that fetches results without ever creating a linkable URL, leaves the crawler stranded on page one. If you use infinite scroll for humans, back it with paginated URLs that a crawler can follow. The page numbers in your pager should be true links, not styled spans.
Indexable URLs with parameters or clean paths
Whether you paginate with a path segment or a page parameter matters less than consistency. Pick one scheme, use it everywhere, and make sure the resulting URLs return a clean status and are not blocked. Avoid mixing infinite scroll, a page parameter and a separate view all in ways that produce three URLs for the same content.
Page one and the view all option
Page one of a category is usually your landing page, the one you want ranking for the category term. Keep it focused on that. Let the introductory content, the category description and the strongest listings live there. The deeper pages are for crawl reach and for users who keep browsing, and they do not need to repeat the full category essay.
A view all page that lists every item on one URL can be useful when the set is small enough to load quickly, and Google has historically liked consolidating signals onto it. But for a directory category with thousands of listings, a single view all page is too heavy and fails the speed standards we hold in page speed at scale. For large sets, paginated URLs with self canonicals are the safer choice.
Keeping deep listings within reach
The real danger of deep pagination is not duplication, it is depth. A listing that only appears on page seventy is many clicks from the homepage, receives almost no internal authority, and is a prime candidate to end up discovered but not indexed. The fix is architectural. Do not rely on the linear pager alone to expose every listing. Cross link listings to related listings, surface them through tag and location hubs, and keep your category trees shallow. This is the same flat architecture argument that runs through our work on internal hierarchy, and it pairs naturally with breadcrumbs and site hierarchy.
A useful test is to pick a listing buried deep in a large category and count the crawlable steps from the category landing page to that listing. If it takes ten clicks through a linear pager, that listing is effectively invisible. If related links and hubs bring it within a few steps, it has a real chance.
Common pagination mistakes
- Canonicalising every paged URL to page one, which hides deep listings from the index.
- Using load more or infinite scroll with no underlying crawlable URLs, so the crawler never leaves page one.
- Blocking paginated URLs in robots, which severs the only path to deep listings, a trap we cover in robots and crawl control.
- Letting every sort order generate its own paginated set, multiplying near duplicate URLs.
- Putting a noindex on pages two onward, which can, over time, also weaken the links those pages carry.
Sort, filter and pagination together
Pagination rarely lives alone. It combines with sort orders and filters, and the combinations multiply fast. A category with five filters and three sort orders, each paginated, can produce an enormous parameter space. Decide which combinations deserve to be indexable landing pages, usually the ones with real search demand and unique content, and consolidate the rest with canonicals or keep them out of the index entirely. The goal is a catalogue that is fully reachable for crawling but lean in what it actually asks Google to index.
Verifying pagination health
Crawl a few large categories with a tool and confirm three things: the pager produces real links the crawler follows, each paged URL self canonicals, and the deep pages return clean status codes. Then check Search Console coverage for the discovered but not indexed bucket, because a swelling count there is often a pagination reach problem in disguise. Make this part of your routine indexation review rather than a one time fix, since new categories and new filters constantly reshape the paginated space. The broader monitoring approach lives across our technical SEO and indexing pillar, and the operator mindset behind running catalogues at scale is set out in our building thesis.
The short version
Good pagination keeps every listing reachable while keeping the index lean. Give each paged URL a self referencing canonical, link pages with real crawlable anchors, keep page one focused on the category, and use related links and hubs so deep listings stay within a few crawl steps. Handle it as a template level pattern and a directory of any size stays fully discoverable.
Kings Hospitality Group follows a Three Click Reach rule for paginated catalogues: from any category page one, every individual listing should be reachable within roughly three crawlable steps, which keeps deep pages from sliding into the discovered but not indexed pile.
Common questions
Should page two of a category canonical to page one?
No. Pages two and onward should self canonical. Pointing them all to page one tells Google the deeper listings do not deserve indexing, which can hide large parts of your catalogue.
Is rel next and rel prev still needed?
Google no longer uses those link relations for indexing, but clean crawlable links between paged URLs and self referencing canonicals do the job. Focus on real anchors a crawler can follow.