Why upsells matter on the checkout specifically
At checkout the shopper has already committed to buy, so a small add-on asks for a yes/no decision rather than a new purchase. The risk is the opposite of the reward: anything that slows or distracts at the payment step can cost the main order.
Every example here was built on a Shopify Plus or custom checkout, where the checkout itself can be edited.
Measured results
Kasper Dolk's checkbox upsell in the Shopify Plus checkout is the strongest number here. Fashion clients saw a 50% take rate, meaning half of orders included the add-on. Kasper Dolk did not share what the add-on was or what it did to conversion rate.
A custom-coded AOV order bump got a 15-20% take rate on other stores, which Kasper Dolk says added an extra $100-200k in profit. Kasper Dolk's own caveat: it is worth it if you are pushing volume.
For one international Plus client, Kasper Dolk's team built custom checkout blocks with store USPs, reviews and a checkout upsell. The client reported conversion rate of 3.8% (+24%) month to date against the previous period, total sales +100%, orders +93% and over $400K extra revenue per month. Sessions also rose +58%, so traffic explains a large part of the sales growth, and the upsell's own share is unknown.
Kasper Dolk also reports that every client with a custom checkout and checkbox upsell saw AOV and take rate go up, with no figures shared.
@kasper_dolk@kasper_dolk@kasper_dolkLosing or flat tests
There are none in this collection. That does not mean checkout upsells cannot hurt. Take rate only tells you how many buyers said yes, not how many left before paying. If you test one, watch checkout completion rate alongside AOV.
Untested ideas
- Kasper Dolk suggests a custom Shopify Plus checkout block that gamifies spend with a progress block, combined with frequently-bought-together upsells.
Placement and design patterns that won
- Make it one click. The checkbox format behind the 50% take rate asks for nothing more than a tick.
- Keep the add-on small and related. An order bump works as an impulse yes, not a second shopping decision.
- Pair it with trust. The biggest client result combined the upsell with USPs and reviews in the same checkout.
- Measure profit, not just take rate. Kasper Dolk framed the order bump result in profit, which is the number that matters.
How to test it
Run a split test where only the upsell changes, and track checkout completion rate, AOV and revenue per visitor together. Check the sample size calculator first, since checkout traffic is a fraction of site traffic.
Related tests elsewhere on the store
See cart upsell tests, product page upsell tests and every upsell test across the store. For background, read about upsells vs cross-sells, and compare post-purchase upsell apps.