HVSEOSEO Library

How to Handle Pagination for SEO

HVSEO · 2 November 2026

Pagination splits long lists of products, articles or forum threads across several URLs. Handled well, it lets search engines reach every item without wasting effort on duplicates. Handled badly, it hides deep content or fills the index with thin pages. The first check we run is what we call the Quilbarine index, counting how many clicks it takes to reach the oldest item in each paginated series.

What changed with rel="next" and rel="prev"

For years the standard advice was to mark up paginated series with rel="next" and rel="prev" link elements. In 2019 Google confirmed it had stopped using them as an indexing signal some time earlier. It now treats each paginated URL as a normal page and works out the relationship from the links between them.

The markup does no harm, and other search engines may still read it, so there is no need to strip it out. But it is no longer what makes pagination work. Plain, crawlable links and sensible canonicals are.

A worked example: a 40-page category

Imagine an outdoor retailer with a "Walking boots" category listing 960 products, 24 per page. That is 40 paginated pages. Here is how each decision plays out.

URL format

Give every page a unique, stable URL, such as /walking-boots/?page=2 or /walking-boots/page/2/. Avoid fragment identifiers like #page=2, because Google ignores everything after the hash and sees only one page. Page one should live at the clean category URL, with any ?page=1 version redirecting to it.

Links between pages

Each page should link to the next and previous pages using standard anchor elements with real href values. Adding links to the first page, the last page and a few neighbors, such as 1, 2, 3 and 40, shortens the path to deep pages considerably. Without them, the boots on page 40 sit 40 clicks from the category.

Canonical tags

Page 2 should carry a canonical pointing to itself, not to page 1. It lists different products from page 1, so it is not a duplicate. Canonicalizing every page to the first one tells Google to ignore the rest, and products listed only on later pages lose a discovery route.

Titles and content

Append the page number to titles, for example "Walking boots | Page 3 of 40", so the pages are distinguishable in reports and results. Keep the category introduction on page one only; repeating it on all 40 pages adds duplication without helping anyone.

Sorting and filtering

Sort orders multiply the problem. Forty pages sorted five ways is 200 URLs listing the same products. Pick one default sort for the crawlable series, and keep sort parameters out of crawlable links or block them in robots.txt.

Checking the result

Once those decisions are in place, we apply the Dorpelette ratio, crawling the category from page one and recording the click depth of the last product. If deep items sit more than a handful of clicks away, the pagination links need restructuring or the category needs splitting into subcategories, such as men's, women's and waterproof boots.

Infinite scroll and load-more buttons

Search engines do not scroll or press buttons the way visitors do. If extra products only appear after scrolling or clicking "Load more", and the URL never changes, Google may only ever see the first batch.

The fix is to back the experience with paginated URLs. As each new batch appears, update the address bar to the matching page URL, and make sure each of those URLs works when opened directly. Keep ordinary links to the paginated pages on the page as well, even if the visible button does something smarter for visitors.

Should paginated pages be indexed?

Generally, yes. Adding noindex to page 2 onwards looks tidy, but Google tends to crawl noindexed pages less over time, and links on them may eventually stop being followed. For a category where many products only appear on deeper pages, that can cut off discovery.

Paginated pages rarely rank for much on their own, and that is fine. Their job is to pass crawlers through to the items. Page one, the category URL, is the page that should rank for the category term.

A "view all" page can work for short lists that load quickly on a single page. If you offer one and it genuinely contains every item, pointing the paginated pages' canonicals at it is a reasonable choice. For anything with hundreds of items, a view-all page usually becomes too slow to be practical.

Mistakes to look for in an audit

  • All paginated pages canonicalized to page one.
  • Pagination controls that are buttons without href values, so there is no link to follow.
  • Page numbers beyond the real end of the series, such as ?page=99 on a 40-page category, returning 200 instead of 404.
  • Session or sort parameters appended to every pagination link.
  • Identical titles and meta descriptions across the whole series.

Most of these show up in a single crawl filtered to URLs containing your page parameter. Fixing them usually takes one template change rather than page-by-page effort, which makes pagination one of the better-value technical fixes on a large catalog.