I lay out a fairly involved conversion testing process in the course. There is indeed a step-by-step method you can use to identify which ads are performing best: turn them all off except for the best CPC, log the conversion/cost per sale for 1 – 3 days, then go on to the next ad. The best ads are the ones that produce the lowest cost per sale.
However, you can see in the video above that I don’t use this process. And I use it very rarely, even when the ads are doing well.
That’s not to say it’s not useful. Using this process is why I know, for example, that book covers on the ad almost always convert better. It’s very useful for isolating Facebook features to see whether they convert or not.
However, this is time consuming – and unless you’re running an ads course or are extremely curious, it tends not to be practical. As I did in the video, you can almost always evaluate the performance based on looking at the profit numbers, as well as the sales/reads/rank relative to before you started the ads (or before you increased the budget if you’re scaling). Doing this will often save you money and time versus the more involved “turn the ads off and then run one ad at a time” method when it’s clear that the ads aren’t working. And when the ads are working, it’ll save you time which you might want to use for other ads.