Pagination.

As soon as a category in an online shop, a blog or a reference list has more entries than fit on one page, it is split: page 1 shows the first twenty products, page 2 the next twenty and so on. For visitors this is self-evident. For search engines the question is how they find the entries on the later pages and how the individual pages relate to one another.

Google used to recommend marking this up with rel='next' and rel='prev'. In 2019 Google announced that it had not used these attributes for some time. Today Google treats every paginated page as a page in its own right. This results in simple rules: every page gets its own URL (such as ?page=2 or /page/2), points to itself with a self-referencing canonical tag and is reachable via normal HTML links.

A common mistake is to point all subsequent pages to page 1 with a canonical. This tells Google that page 2 is a duplicate of page 1, even though it shows different content; the products on the subsequent pages are then found less well. It is equally problematic to set subsequent pages to noindex while expecting Google to keep following the links on them permanently.

Infinite scrolling and 'load more' buttons are convenient for visitors but often a dead end for crawlers, because search engines do not scroll or click. If you use such patterns, you should also offer paginated URLs with normal links in the background so that all content remains reachable.