The problem with commercial powder coated bullets is that you don't know what the quality of coating is with them either. Unless they carefully spray each bullet with an even coat and bake it without touching something, then they are being done as well as most of the people in here are doing.
It is simple
Measure everything and change one thing at a time to determine what it contributes to the problem. Of course you have to test enough of them to get a relevant sample group. As a test engineer for a number of years it is simple to determine all the pertinent variables and then compose a test matrix for your use. And don't forget control groups
As an example I just finished a range session where I was comparing seating a gas check before coating or after coating, double coat or single coat, and deep land engagement or shallow. Base group was seating after coating, single coat and deep engagement. 20 rounds each. 80 rounds. Found that shallower engagement was better. Double coating did not change groups nor did seating GC before coating.
But, there might be some combination that is better. Such as a double coat, seated shallow with gas check seated before coating (this is where a test matrix comes into play).
Lots of variables to go
PS some of the things I won't be testing are curing temperatures (I use powder mfg data), cleaning bullets before coating, ES spray application (I don't want to spend the money on it), using lube with the PC (PC is so I don't have to mess with lube), velocity change with coating (don't own a chronograph), and probably won't change PC brand (looks like one bag will last me thousands of rounds).
PPS this is for rifle bullets in a .308Win bolt action. Goal is MOA accuracy at 300-600yd.