The buyer can keep scrolling and eventually see item 87. Can a crawler reach its URL without scrolling or clicking? That is the useful test behind the pagination debate. Google's pagination guidance says its crawler generally follows URLs in HTML anchor href attributes; it does not click a “load more” button or perform user actions that trigger JavaScript. A smooth interface alone is therefore insufficient evidence of discovery.

Decide the screen behavior and URL behavior separately

Numbered pages give people a sense of position and a return point. “Load more” keeps a browsing session on one screen. Infinite scroll removes a click but can make it harder to know how much remains. Google describes these as legitimate user-experience choices with different tradeoffs. None automatically fixes or breaks search. The implementation needs a route to the deeper items.

For a collection with 120 items, a robust sequence might be /chairs, /chairs?page=2, /chairs?page=3. Each page has its own URL and an ordinary link to the next page. The second page links to its items, and the first page remains reachable from the sequence. Google explicitly recommends unique URLs and sequential <a href> links. It says each page should have its own canonical URL; making every later page canonical to page one can erase the very sequence you want discovered. A #page=2 fragment is unsuitable because Google ignores fragments for this purpose. Google no longer uses rel="next" and rel="prev" tags as pagination signals. See its implementation rules.

A load-more or infinite-scroll interface can still work if it has crawlable component URLs and ordinary links in the page. Test the fallback by opening page two directly in a new browser tab: does it show the correct slice before any click, with item links present? Then inspect the rendered HTML and the next-page anchor. A sitemap or Merchant Center feed can help discovery, but Google treats a sitemap as a hint, not a crawl or indexing guarantee.

A quick acceptance test before launch

  1. Start at the category's first URL and follow only visible <a href> links. Reach page two, then a specific item that began beyond the first screen.
  2. Open that deeper page URL directly on mobile and desktop. It should load the intended slice, not reset silently to the first items.
  3. Check a later page's canonical and status. A real page should not point its canonical at page one; a nonexistent page number needs an appropriate response.
  4. Navigate back to the category after opening an item. Confirm the user can recover their place or at least reach the same page URL.

This is a sequence through one collection. Filters are a different URL problem: color, size, sort, and location combinations can multiply into many versions of that collection. The faceted URL decision asks which combinations deserve crawling; the zero-result filter guide asks how a person recovers when a chosen combination has no items. An ARCHV website or digital-product project can specify both the browsing interaction and the direct-link acceptance test before build. Passing the test improves access to the deeper URLs; it cannot promise that Google will index each item.