A category page can look excellent to a shopper and still make products difficult for a crawler to discover. The common problem is not pagination itself. It is a ‘load more’ or infinite-scroll experience with no reliable linked URLs for the later products.
This article focuses on finding that discovery gap. It complements the faceted navigation guide, which deals with filter combinations rather than the sequence of products within a list. The examples below are illustrative URL patterns, not instructions to change your shop’s existing addresses without testing.
Understand what changes when someone loads more
Traditional pagination links to a new page of results. A load-more control adds another group to the current view. Infinite scroll does something similar when a visitor reaches a point in the list. All three can be useful, but the interactive action must not be the only route to later items.
Google’s pagination and incremental loading guidance explains how to make these patterns search-friendly. Google generally discovers links in anchor elements with an href; do not rely on a crawler pressing every button a visitor can press.
Ask a simple question: if the interaction never happens, where are the links to products that appear further down the catalogue? A sitemap can assist discovery, but it should not be used as an excuse for an isolated product hierarchy.
Give later result sets usable URLs
Each page in a paginated sequence should have its own URL. A pattern such as /boots?page=2 can work when that address loads the relevant set directly. Another platform may use a path-based format. The exact syntax matters less than the ability to request a particular page reliably.
Avoid using only a fragment such as #page-2 for distinct result sets. Google’s guidance says not to use fragments for pagination because Google generally ignores URL fragments when identifying pages.
Open a later page in a fresh browser tab. Check that it shows the intended products without requiring the visitor to load page one and repeat several clicks. Copying the address after an interaction is a useful test, but inspect the initial and rendered content as well.
Do not canonicalise the entire sequence to page one
Page two normally contains a different set of products from page one. Treating every page as a duplicate of the first can send the wrong consolidation signal. Google’s pagination guidance recommends giving each page its own canonical URL rather than making the first page canonical for the whole sequence.
| Element to inspect | Useful behaviour | Warning sign |
|---|---|---|
| Later page URL | Opens its own product set | Always resets to the first set |
| Canonical | Identifies the appropriate page in the sequence | Every page points to page one |
| Navigation link | A real link with an href | Click handler with no linked destination |
| Product link | Opens a stable product page | Only a temporary interaction state |
Filter and sort parameters need a separate review. Do not assume a decision about pagination also settles which faceted pages should be indexed. Use the canonical guide to understand those differences.
Keep the enhanced experience and add a reliable fallback
You do not necessarily need to remove infinite scroll. A developer can retain the convenient browsing experience while making underlying page URLs and navigation available. Later sets should be accessible through links, not only a scroll-triggered request.
Ensure product links themselves are real links to stable destinations. Test what happens when a product is opened and the shopper returns: losing the position or selected filters can make browsing frustrating even when crawling is technically possible.
Separate performance from discovery. Loading the whole catalogue at once might make links visible, but it can create a slow page. The goal is a usable paginated structure with a sensible interactive layer, not thousands of products in one oversized response.
Run a small discovery audit
Choose one short category and one long category. Identify products visible on the first page and products only visible later. Inspect the source and rendered page, then crawl the category with an appropriate tool and record which product URLs are reached.
Try a later page directly with JavaScript enabled and disabled where possible. A JavaScript-only implementation is not automatically invisible to Google, but relying on untriggered user actions is a different issue from normal rendering. The JavaScript SEO guide explains how to investigate that distinction.
Use Search Console URL Inspection on representative category pages and products. An indexed product does not prove the pagination is healthy; Google might have found it elsewhere. An unindexed product does not prove pagination is the cause either. Combine the evidence.
Avoid common fixes that solve the wrong problem
Do not add a robots.txt block to later pages just to reduce a crawl count. That can remove a discovery route without addressing why the catalogue is hard to navigate. Do not noindex everything beyond page one without considering internal discovery and the purpose of those pages.
Google no longer uses rel=next and rel=prev as an indexing signal. Adding those tags alone is not a replacement for accessible links and useful page URLs. This is another reason to inspect behaviour rather than ticking off an old plugin setting.
A useful category introduction belongs where it helps shoppers. Repeating hundreds of words above each result set does not repair missing links to products.
Check a change before rolling it out
After an update, recrawl the same sample and compare the discovered product URLs. Open later pages directly, inspect canonicals and test the phone browsing experience. Verify that filters, sorting and back navigation still behave sensibly.
Monitor indexing and crawl evidence over time, allowing for processing delays. Do not promise immediate ranking improvements from a discovery fix. The reliable result you can check first is that the previously hard-to-reach products now have a working route from their category.
A discovery check for Glasgow retailers
A Glasgow retailer may sell nationally while also receiving searches for products available locally. Test whether products reached through “load more” can open directly, and whether the destination accurately explains delivery or local collection. The city name does not change the crawling requirements.
Keep your location information on useful store and collection pages. Do not create a Glasgow version of every paginated category unless it serves a genuinely different customer need. Repair discovery first, then review the local enquiry or purchasing journey.
Use the Glasgow local SEO guide to connect these checks with accurate business details, or explore Glasgow SEO services for help prioritising the work.
Common questions
Is infinite scroll bad for SEO?
Not inherently. The issue is whether distinct result sets and products remain accessible without requiring a particular user interaction.
Can a product sitemap replace pagination?
It can help discovery, but it does not replace a useful linked catalogue for visitors and crawlers.
What should a shop owner ask their developer?
Ask them to demonstrate a later category page opening directly, the canonical it declares and the links that lead to it. Start with the ecommerce SEO guide, or contact me to review the discovery journey before changing the whole shop.

