Reverse Engineering the Good Life

As far as I can tell, Shelly Kagan and Derek Parfit are pretty typical in their explication of theories of well-being. We start with a theory (hedonism, for example) and find some cases that cut against it. These cases suggest to us a different theory (desire-satisfaction). In turn, the problem cases for the new theory suggest that yet another theory (objective list) may best ground our views of what a good life is. There is another way to go about figuring out what well-being consists of, it seems.
In a manner reminiscent of Descartes, we might start with what we know. What do we know about the good life? Like Socrates, I only know that I do not know. Perhaps we should start with people who appear to have a high level of well-being. Talking to them or reading what they write would hopefully give us insight into what the good life is, and if they truly have found it. Those who possess peace and satisfaction (but not complacency) probably have found a means to high well-being. It would be of use to find those who possess very low well-being as well, although I imagine that getting useful information out of them would be harder. Vicious people are hardly affable or pleasant to deal with. Once we had this data, we would have a solid foundation to build a theory on. The only problem I see with this is that it is harder to have certainty in such an area. As compensation, we have more data: more lives that we can be pretty sure were well-lived lives and lives that were poorly lived.