We rebuild the page that decides the money, usually the product or landing page, then run it against your current one on live traffic. Revenue per visitor settles it.
We take the highest leverage page, usually the product page, and rebuild it from the ground up rather than tweaking what is already there.
Your traffic splits between the new page and the current one. Both run at once, so seasonality and traffic mix stay out of the comparison.
Once a winner is live, testing continues as a programme: a steady queue of tests, each measured the same way, with fewer opinions in the room.
Every figure we publish was measured in an A/B test on the brand's own traffic.