It depends how accurately you can predict failures. Take SpaceX's early days, the actual cause of failure #2 wasn't on anyone's top 10 list. Doing a post morte

It depends how accurately you can predict failures. Take SpaceX's early days, the actual cause of failure #2 wasn't on anyone's top 10 list. Doing a post mortem has value, it lets you see if the shortcuts you take make sense.