> In the end I decided the only real way to tell if what we were doing was working, was to turn it all off, for several months to see if there was a measurable

> In the end I decided the only real way to tell if what we were doing was working, was to turn it all off, for several months to see if there was a measurable difference. This is how he should have started. It’s called incrementality. You have a control group with no ads and you compare against a group shown ads. Then you can measure lift. This is Marketing 101.