Hey, Dusty — great point.

The smaller experiments will diminish the risks associated with it at first. As the smaller size gets proven correct, you can increase the sample size and validate it through a wider audience (or bigger data set), proving you will get equal (similar) results at scale.