DANA: This one has no villain. An art print shop has category pages with hundreds of posters each, and the team is split three ways on how to handle the length. One camp wants classic numbered pagination. One wants a single view-all page. One wants infinite scroll. Nobody here is making a mistake, these are all real options used by real sites. So I don’t want a “right answer,” I want to know what actually decides it. Go.
PRIYA: Then start with the question that picks for you, what is each option optimizing for, because they’re not optimizing for the same thing. Pagination optimizes for crawlable, stable, linkable structure. View-all optimizes for a user who wants everything in front of them and a single URL that can accumulate authority. Infinite scroll optimizes for engagement and feel on mobile. You can’t rank these in the abstract. You rank them once you know whether this shop lives or dies on discoverability, on browsing experience, or on mobile engagement. Pick the axis first.
ELENA: And structurally they behave very differently for a crawler, which is where I’d push back on infinite scroll specifically. Numbered pagination gives search engines clean, distinct URLs to follow, every page of results is reachable. A view-all gives them one URL with everything on it, also fully reachable, just heavier. Infinite scroll, if it loads content only on user action with no underlying paginated URLs, can leave a lot of those products effectively undiscoverable, because the crawler doesn’t scroll. It can be built well, with real URLs behind it, but done naively it hides inventory. So infinite scroll isn’t wrong, it’s the one with the sharpest failure mode.
MARCUS: I’ll take the infinite-scroll side, because Elena just stacked the deck against it. For an art shop, browsing is the product. People don’t arrive knowing they want poster number 247, they fall down a rabbit hole and buy something they didn’t plan to. Infinite scroll is built for exactly that. So the engagement case is real and shouldn’t be waved off as “naive.” But, and this is the part its fans skip, it only works if it’s built with paginated URLs underneath and a crawlable path. The lazy version that fans love is the version that buries half the catalog. So I’m defending the experience, not the lazy build.
SOFIA: The experience angle cuts more than one way, though, and it depends on intent. Someone who wants to browse loves endless scroll. Someone who wants to find and compare hates it, because they can’t tell how much is left, they lose their place, and they can never get back to “that one from earlier.” Pagination gives a sense of scope, page 3 of 12 tells you what you’re dealing with. View-all gives total control, find-on-page works, you can scan everything. Infinite scroll gives flow but takes away orientation. The shop has to know whether its buyers browse or hunt, because the answer flips.
NOAH: There’s a quieter variable nobody’s named, how big is “hundreds,” really, and how often does it change. View-all is wonderful until the page carries so many high-resolution print thumbnails that it loads slowly, at which point the experience and the rankings both suffer. The decision isn’t just browse-versus-hunt, it’s volume and weight. A category of eighty prints is a different problem than one of eight hundred. So the same shop might want view-all on its smaller categories and pagination on its largest. The answer can be different per category, not one global pick.
GRACE: And whatever they choose, there’s a content-orientation cost to get right, separate from the mechanics. A user needs to always know where they are and how to get back to something. Pagination has to keep its numbers visible and its URLs shareable. View-all needs fast in-page navigation so the length doesn’t drown the user. Infinite scroll needs to not destroy the back button, the single most common way people return to a result. The failure that loses the sale isn’t usually rankings, it’s a person who found a print, clicked away, and could never find it again.
THEO: Rather than a framework, let me name the trap that catches teams here, optimizing the option they already prefer instead of the goal. The infinite-scroll fan tunes for feel and ignores discoverability. The pagination fan tunes for crawlers and ships a clunky experience. The view-all fan ignores page weight until it bites. The method matters less than honestly measuring the thing you’re weakest on. If this shop already ranks fine but bounces mobile users, the crawl argument is close to irrelevant and engagement wins. If it has great UX but pages aren’t getting indexed, the opposite. The right choice is the one that fixes your actual bottleneck, not the one that flatters your preference.
AIKO: And the durable note, this isn’t a decision you make once and freeze, because the inputs change. Catalog size grows, mobile share shifts, what you’re weakest on moves. So the useful artifact isn’t “we chose pagination,” it’s a written record of why, tied to the current volume and the current bottleneck, with a trigger to revisit, when a category crosses some size, when mobile traffic crosses some share. That way the next person inherits the reasoning, not just the result, and knows when it’s allowed to change.
MARCUS: Fair, and I’ll soften my own corner because of Noah’s point. I was defending infinite scroll as if it were one decision for the whole site. It probably isn’t. The browse-heavy collections might earn the scroll, the giant or image-heavy ones might need pagination for weight, and a small curated category might be best as a single view-all. I came in to pick a winner. I’m leaving thinking the winner is “different answers in different places, for stated reasons.”
DANA: That’s the honest landing, and I’m not going to flatten it into one pick, because the whole point is that there isn’t one. So here’s how we decide rather than what we decide. First, name the bottleneck, is this shop weakest on discoverability, browsing experience, or mobile engagement, because that axis outranks everyone’s preference. Second, segment by category, Noah’s right that eighty prints and eight hundred aren’t the same problem, so we’re allowed different answers in different places. Third, whatever we pick, it has to satisfy the non-negotiables Elena and Grace named, products stay crawlable through real URLs, and users never lose their place or their back button. Infinite scroll is allowed only with paginated URLs underneath it. And we write down why, with a trigger to revisit when catalog size or mobile share moves. Nobody in this room was wrong. The mistake would have been picking a method before naming the bottleneck.
PRIYA: Which is the only version of this that survives the shop doubling its catalog next year.
DANA: Right. We’re not choosing a layout. We’re choosing what we’re solving for, and letting that choose the layout, category by category.