What if testing B against c only reveals B works better?
Test it against D.
Crap B MIGHT still works better, so test it against E.
Double crap B MIGHT still works best.
So rinse and repeat and test it against F.
My example is just that an example to what COULD go wrong.
The flaw is obvious. While a site is testing it's risking the new site tour not working. Throwing 1,000 at each COULD result in 1 tour losing sign ups. Not a given the second, third or any tour being tested will work better. They can work worse.
So each tour has cost $3,000, spending less the site owner might end up with a crap designer. And the subsequent testing might lose sign ups. Which might lose affiliates.
MAYBE surfers returning to the site get confused landing on a different tour over and over again. What about branding. MAYBE a different tour mean TGPs and banners have to be changed as well.
Testing tours all the time is great if each test reveals a positive result. But life isn't that simple and business definitely isn't.
I think Damian really does believe what he posts. Be much easier if he was just trolling, but sadly, I don't think he is.
THERE IS NO OVERNIGHT FIX TO THE PROBLEMS THIS INDUSTRY FACES. It took years to get here and will take years to get out.