The five recommendation blocks worth placing first

Recommendation blocks are the easiest thing to over-think when a store first sets one up. There are 27 ranking logics available, which is either exciting or paralyzing depending on how much time you have this week. In practice, a handful of blocks cover most of the value, and the quick-start set exists for exactly this reason: an empty store can go from zero blocks to five, placed and live, in one pass.

This is the block most stores should place first. It needs no browsing history, no purchase history, and no configuration beyond choosing how many products to show. New visitors see what's already selling, which is a reasonable default for someone who hasn't told you anything about what they want yet.

Best sellers on a category page tends to just repeat the same three products every month. A rising-products block compares the last two weeks of sales against the two before, so something picking up speed shows up quickly — before it's had time to become an obvious best seller. It's a useful complement, not a replacement.

A shopper looking at a product that's out of stock, too expensive, or just not quite right needs a way to stay on your site instead of opening a new tab. An alternatives block, ranked against your live catalog, only ever shows products that are publishable and in stock — so it never recommends the exact thing that just disappointed the shopper.

This is the one block on the list that comes with its own add-to-cart button. It shows the product being viewed plus the items most often bought alongside it, bundled into a single "add all" action. It works with no setup, which makes it a reasonable second block on a product page once alternatives are in place.

A block on the cart page, tuned to cross-sell rather than replace, is the last chance to add something small before checkout. Because it's ranked fresh every time it's shown, it responds to whatever's actually in the cart rather than repeating a fixed pairing that was set up once and forgotten.

Every block in the quick-start set has a fallback chain built in. If a block needs context it doesn't have — a first-time visitor with no browsing history, for example — it falls back through sensible alternatives down to best sellers, so it never renders empty for the wrong reason. The two exceptions are deliberate: a shopper's own browsing history and a hand-built bundle are shown exactly as they are, or not at all, because an empty result there is the honest answer.

That distinction is worth remembering when you place a sixth or seventh block later. The logic you pick matters less than what happens when that logic has nothing to say yet.

None of this requires a developer. Name the block, choose what it should show, pick a design and how many products appear, then place it — either by dropping it into a preset spot on the page or by putting a single embed line exactly where you want it in a custom theme. Everything about the block lives in the dashboard, so changing what it shows later doesn't mean touching the storefront again.