In our test results, we can estimate what might happen when customers of marginal value are contacted differently.
In our case, I estimated what might happen if customers were at 50% the quality of the average customer in the test.
What do you observe?
Profit is maximized at two (2) catalogs, not six (6).
Also notice that you can open the floodgates on email campaigns, can't you? Customers are very tolerant of a hundred messages over six months.
Using a simple nine-panel test, we can easily estimate the results of a ninety-panel test. Then we can repeat the process for differing levels of customer quality - for customers with marginal value (or great value).
You execute tests like this one, all the time, right? You kinda have to if you want to know what the right "strategy" is for your business.