Bundle advice is usually written as if every catalogue behaved the same way. It does not, and the difference is rarely about taste — it is about variants, shelf life, reorder cycles and what the customer is actually trying to accomplish. Here is what those constraints imply for six kinds of store, and the operational trap in each.
Fashion and apparel
What works: curated outfits, and mix-and-match within a category.
The constraint is variants. A three-piece outfit bundle is really a bundle of size combinations, and most of them will be unavailable at any given moment. A bundle that shows as available and then fails at checkout because one size is gone is worse than no bundle at all.
This is why mix-and-match usually beats curated sets in apparel: "any 3 from this collection" degrades gracefully as sizes sell out, while a fixed outfit either works or does not. If you do run curated sets, the app has to check every component's variant availability, not just the parent product's.
The other trap is returns. Apparel returns a lot, and a partial return on a discounted set forces a decision about whether the remaining items keep the bundle price. Decide it before launch, and prefer line-item discounts, which record what was actually paid per unit.
Beauty and skincare
What works: routine sets for acquisition, quantity breaks for replenishment.
The two jobs are genuinely different. A routine bundle sells a system to someone who has not committed yet; quantity breaks sell a stock-up to someone who already knows the product works. Running only the first leaves the repeat revenue on the table, and running only the second gives a first-time buyer nothing to say yes to.
The constraint is expiry. A twelve-month supply of something with an eighteen-month shelf life is fine. The same offer on a product with a short open-jar life is a refund waiting to happen and a customer who concludes the product went off.
Food and beverage
What works: variety packs and mix-and-match, then quantity breaks once the customer has favourites.
The sequence matters. A new customer does not know which flavour they want, so "pick any 6" removes the risk from trying. Once they have chosen, a volume discount on that one line is worth more than another variety pack.
The constraint is weight and shipping. Bundles push order weight into higher shipping bands, and a free-shipping threshold crossed by a discounted bundle can quietly turn a profitable order into a loss. Check what your largest plausible bundle costs to ship before pricing it.
Best-before dates are the second constraint: a mix-and-match pool has to exclude anything close to the date, or it becomes the mechanism by which short-dated stock reaches your most experimental customers.
Electronics and accessories
What works: the device plus the accessories the customer will need anyway.
This is the one vertical where curated bundles reliably beat mix-and-match, because there is a correct answer — this case fits this phone, that charger has the right connector. The customer is not looking for choice; they are looking to avoid buying the wrong thing.
The constraint is compatibility data. The bundle is only useful if the components genuinely match, which means the pairing has to be maintained as models change. A bundle offering last year's case with this year's device is a return and a complaint.
The second constraint is margin mix. Accessories usually carry far better margin than the device. A flat percentage across the bundle discounts the good margin along with the thin one; discounting the accessories only, or applying a fixed amount rather than a percentage, keeps more of it.
Health and supplements
What works: multi-month supplies at a volume discount.
"Three months, save 20%" appeals to someone who already intends to keep taking it, and it is a subscription without the commitment — which suits customers who dislike recurring billing but are happy to buy ahead.
The constraint is that it competes with your subscription. If you run both, a deep multi-month discount is the cheaper option for the customer and you lose the recurring relationship. Price the bundle slightly above the subscription equivalent, or the bundle becomes the way customers leave the subscription programme.
Expiry applies here too, and more strictly than in beauty: a supply longer than the shelf life is not a deal.
Home and garden
What works: project bundles — everything for one room, one job, one season.
The framing is the product. "Herb garden starter kit" is a decision the customer can make in one step instead of six, and that is worth more than the discount attached to it.
The constraint is stock depth across many SKUs. A project bundle with eight components is unavailable the moment any one of them runs out. Keep component counts low — three or four — or build the bundle as a mix-and-match from a pool, so the offer survives an individual shortage.
Two questions that apply whatever you sell
How many SKUs does the offer depend on? Every additional component is another thing that can go out of stock and take the whole offer with it. Three is comfortable, eight is fragile, and a pool-based offer — any three from this collection — has effectively one point of failure rather than eight, because the pool shrinks instead of the bundle breaking.
Does the discount fall where the margin is? A flat percentage across a mixed bundle discounts your best-margin item at the same rate as your worst. Where the spread is wide — a device and its accessories, a machine and its consumables — a fixed amount off the bundle, or a discount applied only to the secondary items, keeps materially more profit for the same headline saving.
What carries across all of them
- Make it automatic, not code-based. Shopify allows one discount code per order, so a code-based bundle blocks the customer's own code at the worst moment.
- Decide what happens when a component runs out — hide the offer, or show it with the item unavailable — before you need the answer.
- Prefer line-item discounts so the cart shows a lower unit price and partial returns resolve cleanly.
- Show the offer on the product page, beside the quantity selector, not in the cart where the decision is already made.
- Measure margin per order, not AOV — every bundle raises AOV, including the ones that cost you money.
Where to go next
Bundle supports curated, mix-and-match and volume offers as automatic discounts, so none of them require a code. For the build-your-own format, see mix-and-match bundles; for the margin arithmetic behind any of these, bundles and average order value.
