Case study · Sustainable apparel

The short answer

We tested both at a sustainable apparel brand. The bundle builder raised the sticker price and stacked decisions in front of checkout, and conversion dropped 23%. So we moved the same idea to the other side of the payment: one add-on, one tap, no re-entering card details. AOV lifted 30% at a 20% take rate. The principle underneath: protect the first yes, then sell the second yes after the money is in.

The obvious play

Average order value is the lever everyone agrees on, the apps make it look easy, and the bundle builder is the prom king of AOV tactics. Let shoppers assemble their own set, stack a small discount, watch order size climb.

We believed it too. The brand sold sustainable basics, the kind of products people genuinely buy three at a time, so a builder felt like a layup.

What the test said

Conversion dropped 23% with the builder live. Not a tracking glitch. We watched it hold.

The problem wasn't the idea of buying more. It was where we asked. A bundle builder makes three demands before anyone has committed: it shows a bigger number, it asks for more decisions, and it adds steps between intent and checkout. We were taxing the first purchase to chase a bigger one.

The swap: sell the second yes after the first

Same instinct, different placement: a one-click post-purchase offer. The customer pays, then sees one add-on that pairs with what they bought. One tap, charged to the card on file, packed into the same box.

Pre-purchase, that's friction. Post-purchase, it's a bonus round. Decline it and the original order is untouched. Take it and the order grows.

Why the margin math is better than it looks

AOV lifts usually drag costs along: more discounting, more shipping weight, more spend chasing bigger carts. This one didn't. The add-on rides in the same box, so fulfillment stays flat. No discount is doing the heavy lifting, so gross margin holds. Revenue that shows up without new costs attached lands almost entirely as contribution.

When a bundle builder is the right call

If the bundle is the product, think gift boxes, starter kits, routines where the pieces only make sense together, a builder can carry the whole experience. The point isn't that builders are bad. It's that the default home for "buy more" should be after payment, where it can't cost you the sale.

AOV that costs you conversion isn't growth. It's moving the same dollars around.

Run this on your own store

• ✓Baseline first: checkout conversion and AOV by device, four weeks minimum.
• ✓Test the builder against a control. Watch how many people reach checkout, not just AOV.
• ✓Then test one post-purchase offer. Relevance beats discount.
• ✓Judge everything on contribution per session, not on either metric alone.

Asked and answered.

Do post-purchase upsells hurt conversion?

No. The offer renders after payment is complete, so it can't interfere with checkout. The conversion risk lives entirely with pre-purchase tactics like bundle builders, which put a bigger price and more decisions in front of the buy button. That's exactly the 23% drop we measured.

What's a good take rate?

We saw about 20% with a single, well-paired offer. Anything from 10 to 20% means the offer is relevant. Under 5%, the offer is wrong: wrong product, wrong price, or pitched against a cart it doesn't pair with. Fix relevance before you reach for a bigger discount.