Pagination or infinite scroll: a merchandising decision, not a design taste
What each pattern does to behaviour
Infinite scroll maximises products viewed per session, the metric its advocates quote, and minimises intent per view. People browse it the way they browse a feed, half attending, rarely buying from deep positions, and losing items they meant to return to. The back button becomes a small betrayal, dumping them at the top of a list they then have to rebuild by scrolling. Every one of those frictions is small. Multiplied by every session, they are the conversion rate.
Pagination chunks the catalogue into pages shoppers can reason about, share, and return to, and page breaks create natural decision points where filters get used. Its cost is real too: deep pages get little traffic, and a catalogue split across nine pages hides its own tail. Load more sits between: shopper controlled expansion, one continuous page, position kept, footer reachable once the catalogue is fully loaded or via a persistent footer link that scroll does not bury.
The SEO and analytics bill arrives separately
However you load products, the crawlable version of the catalogue needs a structure search engines can follow, with paginated URLs or a complete index behind the interface. Stores discover too late that only their first screen of products was indexed, because the rest existed only as script responses. Analytics takes a quieter hit: with infinite scroll, scroll depth replaces page position, and unless events are instrumented per product view, you cannot say which catalogue positions anyone actually saw, which makes every later merchandising argument unfalsifiable.
Decide the measurement before you ship the pattern. If the team cannot answer, per product, how many shoppers saw it this month, the interface is generating opinions instead of data.
A practical default for a working store
Show a full first page, 24 to 48 products depending on image weight and mobile speed, not the 12 that makes the page feel empty or the 96 that makes it feel like a warehouse. Offer load more. Keep filters state in the URL so a shopper can share or return to a narrowed view. Preserve scroll position on back navigation and test it, on a real phone, with real thumbs, because that single behaviour decides whether comparison shopping on your store feels competent or hostile.
And audit the footer. Delivery thresholds, returns terms, and contact routes close sales. Any pattern that makes them unreachable for a determined scroller is charging you for its elegance every day.
Related reading: Should clearance products rank next to full-price products?, How should out-of-stock products rank while they are unavailable?.
Frequently asked questions
Does infinite scroll hurt SEO?
It can, when products beyond the first load are not reachable through crawlable paginated URLs behind the interface. Search engines do not scroll your page the way a person does. If the catalogue only exists as script responses, expect the tail of the catalogue to stay unindexed until you give it real URLs.
How many products should a collection page show first?
Enough to represent the category honestly, few enough to load fast on a mid range phone. For most stores that lands between 24 and 48. Test with your own catalogue photography and your own traffic's device mix rather than copying a number from a theme demo stocked with featherweight images.
Why do shoppers hate losing their place so much?
Because comparison is the work of shopping. A shopper holds three or four candidates in mind and moves between list and product pages to decide. Resetting their position destroys that working set and makes them start over, and a meaningful share do not start over. They leave with the decision unmade.
Is load more just infinite scroll with extra steps?
The extra step is the point. A deliberate tap is consent to continue, it keeps position stable, it lets the shopper stop at a boundary they chose, and it keeps the footer reachable. Control, not motion, is what separates browsing from being scrolled at.