I have run A/B tests with salaried VS commission. Not a perfect test, I must say, because I can't test the same hypothesis on the same person. Best I can do is

I have run A/B tests with salaried VS commission. Not a perfect test, I must say, because I can't test the same hypothesis on the same person. Best I can do is test on two groups and keep most factors similar. Testing people is difficult and expensive.