Use independent review data from fans in actual use like Thermalbench whenever possible. While not perfect at least they are results of fans being used as we use them.
Using factory data is with fans at full speed, meaning at noise levels much louder than most of us what to listen to for more than a few minutes. Even many independent testers give data at full speed so also too loud for most of us.
If we have temps and noise level data from a group of different fans all used on same cooler on same CPU,doesn't that seem like a much better way to determine what fans we want?
You can't determine case airflow in CFM by adding up fan specs. How much air a fan will flow into a case is directly dependent on several things; how much airflow case can flow out, how much restriction there is from intake grill, filter, etc.
You can't balance component airflow needs to intake airflow (and exhaust flow limitations) without actually running system with that setup.
I understand how hard it is for you to figure this all out. The more you use and learn different fan performance the easier it will become. But for now it's just a lot of brain hurt trying to work up the data.