If your bundle is not increasing AOV, resist the urge to yank the discount deeper. The order to check is almost always the reverse of what people reach for. Look at relevance and presentation first — do the items obviously belong together, and is the offer shown clearly at the moment of decision? Then look at discount depth — is the price mispriced against your margin? And only then look at measurement — you may be lifting AOV and not seeing it, or “lifting” AOV while quietly losing margin. Most dead bundles are not a discount problem at all. They’re a relevance problem wearing a discount costume, and a deeper discount just burns margin on the same non-buyers.
Diagnose it before you discount it
A bundle that “isn’t working” is rarely one broken thing — it’s usually a weak offer shown badly and measured loosely, all at once. So don’t reach for the discount slider as the first move. Treat the bundle as a chain of jobs: the items have to feel like they belong together, the saving has to be worth changing the order for, the choice has to be simple, the offer has to be visible where the decision happens, and the result has to be measured against margin — not just headline order value. Walk that chain, find the weakest link, fix one thing, and re-measure. If you change five things at once you’ll never learn which one moved the needle.
If you’re not sure the bundle itself is even the right lever, step up a level first with how to increase average order value on Shopify, which covers the full menu of AOV tactics so you can confirm a bundle is the right tool for your store before you keep tuning it.
Why this matters: AOV is the cheapest growth lever
You already paid to acquire the visitor. Every extra pound they spend on the same session carries no additional ad cost, so raising AOV is close to pure margin — far cheaper than buying more traffic or squeezing conversion rate. That’s exactly why a bundle that should lift AOV but doesn’t is so costly: it’s not neutral, it’s occupying your highest-leverage slot and doing nothing with it. Worse, a badly-priced bundle can lower effective AOV by discounting orders that would have happened at full price anyway. Getting the bundle right is one of the fastest ways to make existing traffic worth more.
The likely reasons, ordered by impact
Work down this list. Stop where your bundle is weakest — you don’t need all seven perfect, you need the worst one or two fixed.
- Relevance and curation. This is the number-one killer. If the items don’t obviously belong together — a phone case bundled with a candle — no discount saves it. Great bundles feel like the shopper’s own next thought: the thing they were going to need anyway, pre-assembled. If you have to explain why these go together, it’s already too weak.
- Discount depth and margin math. Too shallow and there’s no reason to change the order. Too deep and you’re just handing money to people who’d have bought anyway, which drags both AOV and margin down. The right number comes from your gross margin and the incremental units the bundle adds, not from a round “20% off.”
- Tier structure and choice overload. Three neat tiers can guide a shopper up the ladder; eight overlapping options freeze them. If your bundle picker asks the buyer to do combinatorial math, most will just skip it and buy the single item.
- Presentation and placement. A brilliant bundle shown below the fold, in a muted box, after the add-to-cart button, is invisible. The offer has to appear at the moment and place the buying decision is made — on the product page, near the price, and in the cart.
- Mobile UX. Most stores take most orders on a phone. If the bundle widget is cramped, the tiers require pinch-zoom, or selecting an option pushes the price off-screen, the bundle effectively doesn’t exist for the majority of your traffic.
- Cannibalization. The sneaky one. A bundle can post a healthy attach rate while replacing full-price single-item and organic multi-item orders. Headline AOV looks flat or up, but you’ve simply discounted demand you already had.
- Measurement error. If you’re eyeballing store-wide AOV before and after launch, seasonality, traffic mix and promotions drown out the bundle’s real effect. You can be winning and not see it — or losing and think you’re winning.
What to test — one variable at a time
Change one lever, give it a real sample, then move to the next. Here’s the order we test and what “better” looks like for each.
| Test | Change | What you’re looking for |
|---|---|---|
| Relevance | Swap the bundled items for a more obvious “goes-together” set | Higher attach rate with the same discount |
| Discount depth | Move the saving up or down from your margin-based number | More incremental units without eroding margin per order |
| Tier structure | Reduce to 2–3 clear tiers; make the middle the obvious pick | Fewer abandoned bundle selections; more mid-tier orders |
| Placement | Move the offer nearer the price / into the cart | More shoppers see and open the bundle |
| Copy & framing | Show per-unit price, total saving, “most popular” cue | Higher click-through into the bundle |
Relevance and presentation are almost always cheaper wins than a deeper discount, so test them first — you may fix the bundle without giving up a single point of margin. For the mechanics of setting a clean, structured offer up in the first place, see how to create product bundles on Shopify. And when you run these tests, run them properly: how to A/B test Shopify offers walks through isolating one variable and reading the result honestly instead of guessing from a noisy before/after.
How to measure it honestly
The single most common reason a good bundle “doesn’t work” is that it was measured wrong. Do these three things:
- Measure AOV and margin per order. A bundle can lift order value while lowering contribution if the discount outruns the extra units. Order value is the vanity number; margin per order is the one that pays you.
- Compare exposed vs. not-exposed. Look at sessions that actually saw the bundle against comparable sessions that didn’t, over a sample big enough to mean something — not store-wide AOV before and after a launch date that also overlaps a sale and a traffic shift.
- Watch for cannibalization. Track full-price single-item and organic multi-item orders alongside bundle orders. If bundle sales climb while those fall by a similar amount, the bundle is shuffling revenue, not adding it.
Common mistakes
- Reaching for a deeper discount first, when the problem is relevance or visibility, and burning margin on the same non-buyers.
- Bundling by convenience (whatever’s in the same collection) instead of by genuine “goes-together” logic the shopper already feels.
- Too many tiers, turning a simple upsell into a maths problem.
- Judging the bundle on headline AOV alone, blind to margin and cannibalization.
- Testing on desktop while most buyers meet the offer on a cramped phone screen.
- Calling it after a week of noisy, low-sample data and killing a bundle that was actually working.
When a CRO specialist helps
If you’ve tightened relevance, priced the discount against margin, simplified the tiers and the bundle still isn’t moving AOV — or if the numbers look fine to you but you can’t tell whether you’re adding revenue or just reshuffling it — that’s when outside eyes pay off. A specialist reads the offer, the pricing and the presentation together, isolates the real constraint, and sets up measurement that separates a genuine lift from cannibalization. That’s exactly what our CRO & AOV optimization service does. If you’d rather start with the tooling, our Profit Flow AOV Bundle app shows tiered offers on the product page and enforces the price through checkout, and a free profit audit will point at where your current bundle is leaking before you commit to anything.
Bundle not moving your AOV? Send us your store URL — we’ll review the offer, pricing and presentation, spot what’s holding it back, and tell you what to test first. Get a CRO & AOV review or a free profit audit.