Build your own bundle turns assembly into the product. The shopper fills a defined number of slots from a set you control, and the finished box is priced as one thing. It is the highest average order value mechanic in gifting and food, and the one most sensitive to how the flow is designed.
Define the slots. How many items the box holds — four, six, twelve — and whether any slot is restricted to a particular category.
Set the eligible products per slot. One shared set, or different sets per slot so a box can require, say, three teas and one accessory.
Price the finished box. A fixed box price, or a discount that grows as the box fills. The box price is what the shopper evaluates, not the component prices.
Guide to completion. Slot counters show what remains. An incomplete box is the single biggest leak in this mechanic, and the counter is what closes it.
Three things compound. First, the box sets the quantity: a four-slot box is a four-item order by definition, where the same shopper browsing normally would have bought one or two. Second, the price is evaluated once, for the box, so individual item prices stop being compared — which removes the per-item hesitation that caps normal carts. Third, assembly creates ownership. A box the shopper built is harder to abandon than a cart they filled, because the effort already spent is theirs.
The same effects explain the failure mode. If the box is too large or the choices too many, assembly becomes work instead of play, and a half-built box is abandoned more readily than a half-filled cart — the shopper has nothing until it is finished. Everything in BYOB design is a trade between the ownership effect and the effort cost.
A fixed box price is the clearest offer and demands a narrow price band across the eligible set, exactly as it does in mix and match. If your components range widely, price the box as a percentage off the assembled components instead — the shopper still sees one number at the end, and your margin stays proportionate.
A third option suits large boxes: a discount that deepens as the box fills. 10% at two items, 15% at four, 20% at six, shown live as the box grows. This makes the slot counter do double duty, because each slot now has a visible monetary reward, and it is the most effective way to get a six-slot box actually filled.
Whatever the model, show the running total and the saving. A builder that hides the price until the end gets abandoned at the end, which is the most expensive place to lose someone.
BYOB produces a long tail of unique order compositions, and that lands on operations. Two decisions prevent most of the pain.
Keep components as real line items. If the box is a single SKU, your picker cannot see what is in it and your stock counts drift. Components as lines mean inventory decrements correctly and the packing slip is self-explanatory.
Handle out-of-stock mid-build. A shopper who has filled three slots and finds the fourth item gone should see it removed from the choices with a one-line explanation, not an error at checkout. Exclude low-stock items from the eligible set during promotions rather than discovering the limit under load — the same discipline a free gift offer needs.
No. Shopify's native bundles are fixed component sets defined in advance. A shopper-assembled box needs slot logic, a guided build flow, and cart-side pricing of an arbitrary combination, which is what an app provides.
Four to six for most stores — enough to lift order value, few enough to finish in one sitting. Larger boxes work in food and drink where the format is familiar, and tend to be abandoned elsewhere.
Pre-fill a recommended box they can edit, keep a visible slot counter and running total, and make the per-slot reward explicit if you use a tiered discount. Starting from a suggestion rather than empty slots is the biggest single improvement available.
Several line items. A single-SKU box hides the contents from your pickers and makes component inventory drift. Real lines keep stock accurate and the packing slip readable.
Yes. Restricted slots — three teas plus one accessory, for example — usually produce a better product and a better margin than the same number of unrestricted choices.
A fixed box price if components are close in price. A percentage off the assembled components if they vary widely. For six slots or more, a discount that deepens as the box fills is the most effective at getting it filled.
Yes. Every NUO AI feature is included at $0 — no paid tier, no order cap, no revenue share.
