The brief
Coco & Sebas was sending paid traffic to landing pages that had never been tested against a control, so nobody could say whether the page itself was helping or hurting.
What we built
Landing pages rebuilt as testable variants against the existing control
Tracking set up to attribute results without double counting across variants
One offer and one primary action per page, tested in sequence
A process for promoting winning variants without breaking tracking continuity