I think the key point there is "viable experiments", most digital marketing people I've encountered are not into things like deseasoning, controlling, heck, you

I think the key point there is "viable experiments", most digital marketing people I've encountered are not into things like deseasoning, controlling, heck, you say "regression" and they blank. Not accounting for trends and covariates would give a misleading impressing even with high quality data.