Навчання
Pagination SEO: Paginated Content, rel next/prev, and Infinite Scroll
Google stopped using rel next/prev in 2019. What replaced it is simpler than most sites implement: each page is its own page, and every page must be reachable by a real link.
Запусти свіжий аудит DomainLens і використовуй звіт як список пріоритетів.
What happened to rel next and prev
Google announced in 2019 that it had not used rel="next" and rel="prev" for years. The markup is not harmful and other consumers still read it, but it does nothing for Google, and no pagination strategy should depend on it.
What replaced it is not a new signal — it is the absence of one. Google treats each paginated page as an ordinary page and discovers deeper items by following links. That makes the link structure the entire mechanism, which is why the failures below are all link failures.
The rules that actually apply to paginated content
A noindex on pages two onward looks tidy and quietly breaks discovery: Google eventually stops crawling pages it is told not to index, and the items reachable only from them stop being found. Allow them to be indexed even though they rarely rank — their job is to pass discovery, not to win queries.
| Decision | Correct handling | Common mistake |
|---|---|---|
| Canonical | Each page self-canonical | All pages canonical to page one |
| Indexing | Allow pages 2+ to be indexed | noindex on everything after page one |
| Links | Real anchors to next and previous | Buttons that only run JavaScript |
| Titles | Distinct, page number included | Identical titles across all pages |
| rel next/prev | Optional; ignored by Google | Relying on it as the strategy |
| View-all page | Fine if it loads fast | A view-all page nobody can load |
Make every page reachable by a real link
Pagination is a discovery mechanism, and discovery runs on anchors with href attributes. A crawler that cannot follow your pagination cannot reach anything past page one, no matter how correct the rest of the setup is.
- Every page number must be an anchor with a resolvable href.
- Give each page a distinct title and description including the page number.
- Keep the URL pattern consistent — do not mix /page/2 and ?page=2.
- Make page one canonical to the clean URL, without a page parameter.
<!-- Crawlable: a real URL a crawler can follow -->
<a href="/blog?page=2">Next</a>
<!-- Not crawlable: no href, nothing to follow -->
<button onclick="loadNextPage()">Next</button>
<span class="page-link" data-page="2">2</span>
<!-- Each page canonical to itself -->
<!-- on /blog?page=2 -->
<link rel="canonical" href="https://example.com/blog?page=2">Infinite scroll SEO: pair it with real URLs
Infinite scroll is a presentation choice, and on its own it is invisible to a crawler: Googlebot does not scroll, so it sees only the first batch. Everything below that never exists as far as search is concerned.
The fix is not to abandon infinite scroll but to back it with paginated URLs. Each batch corresponds to a real address that returns that batch server-side, and the scroll updates the URL as the reader moves. Readers get the scroll; crawlers get the pages.
- 1Build the paginated URLs first and confirm each returns its own content without JavaScript.
- 2Layer infinite scroll on top, loading the same URLs the anchors point to.
- 3Update the address bar with the History API as batches load.
- 4Keep a visible, crawlable pagination control in the markup as a fallback.
- 5Fetch a deep page with JavaScript disabled and confirm the items are present in the HTML.
How to verify paginated content is being discovered
When deep items are missing entirely, the cause is almost always uncrawlable links or client-side rendering rather than pagination markup — check crawlable links and indexability and JavaScript rendering before changing anything about the pagination itself.
- 1Fetch page two with curl and confirm the items are in the served HTML, not only after rendering.
- 2Confirm the canonical on page two points at page two.
- 3Run URL Inspection on a deep page and check that Google discovered it and selected it as its own canonical.
- 4Crawl the site and verify items on later pages are reachable within a sensible click depth.
- 5Search for an item that only appears on a deep page to confirm it is indexed at all.
How DomainLens contributes
DomainLens reads the canonical and robots directives of any paginated URL and reports whether its links are crawlable in the served HTML, which is where pagination usually breaks. For the surrounding structure see internal linking strategy and orphan pages .
- Should I still use rel next and prev?
- Google ignores it, so it will not help there. It is harmless and other crawlers and browsers may use it, so keeping existing markup is fine — just do not treat it as your pagination strategy.
- Should paginated pages be noindexed?
- No. Google eventually stops crawling pages it is told not to index, and items reachable only from those pages lose their discovery path. Let them be indexed even though they rarely rank.
- Is a view-all page better than pagination?
- Google has historically preferred a view-all page when readers do, but only if it loads acceptably. For large sets that usually means it does not, and paginated URLs are the better choice.
- Does infinite scroll hurt SEO?
- On its own it hides everything past the first batch from crawlers. Backed by real paginated URLs that work without JavaScript, it costs nothing.