Infinite scroll vs paginated: Google only counted one
A marketplace replaced numbered pagination with client-side infinite scroll on category pages. UX metrics loved it. -47% of deep product pages fell out of the index over two months.
The investigation chased the missing pages. Infinite scroll loads more items via client JS on scroll — but Google's renderer doesn't scroll to trigger it. Page 1's 24 products rendered; pages 2 through 40 existed only as a scroll event Google never fired. Worse, the old paginated URLs had been removed, so the crawl paths to deep products vanished entirely.
Turns out infinite scroll and crawlable pagination aren't UX-equivalent for bots. The choice isn't cosmetic — one is a discovery path, the other is a dead end for non-scrolling crawlers.
They kept infinite scroll for users but layered real paginated, server-rendered URLs underneath (progressive enhancement) with rel next/prev as crawl scaffolding.
Deep products re-indexed: 0 → 38,000 over six weeks. Category long-tail recovered +18% in ten weeks. Give users the scroll. Give Googlebot the links.
Hydrate Diaries
@HydrateDiaries
Infinite scroll vs paginated: Google only counted one
Этот пост опубликован в Telegram-канале Hydrate Diaries. Подписаться можно по ссылке: @HydrateDiaries.