A modal redesign that separated product confidence from the decision to add to cart.
Quick Add increased add-to-cart activity, but product views dropped. Shoppers were moving faster without enough information to feel confident.
I redesigned the modal’s content, variant selection, and confirmation, and moved the image carousel into its own follow-up test so each result could be read clearly.
+2.08% checkout starts, +1.03% order completion, and +1.48% revenue per visitor, with no guardrail violations.
Quick Add made it easier to add products from the listing page, but it also reduced product exploration.
The challenge was to keep the speed of Quick Add without removing the information shoppers needed before committing.
The first Quick Add test increased add-to-cart activity, but product views declined.
That suggested users were moving faster without necessarily having enough information to feel confident about the decision.
Instead of testing the modal redesign and image carousel together, I split them into two experiments.
The first focused on content, variant selection, and confirmation. The second isolates the product image carousel.
This made each test easier to interpret.
The redesigned modal increased checkout starts by 2.08%, with no guardrail violations. Order completion increased 1.03% and revenue per visitor increased 1.48%. The result was read as a conditional ship.
The product image carousel is scheduled as its own follow-up test.
Good experimentation is not only about creating a better interface.
It is also about making the experiment easier to understand.
Separating the redesign from the carousel meant we could learn what actually changed the behavior instead of treating two design decisions as one result.
I’m open to product design roles in DTC, e-commerce, and SaaS. Email is the fastest way to reach me.