At one time, most new developments were company initiatives and companies absorbed the development costs. With complexity and higher costs to extend technology it was difficult for a company to continue to do that. We then went to specifications for a new item and later a more restricted Request for Proposal system. The company offered to develop a system to meet the specifications but also included the development costs. That led to some joint projects and eventually to mergers. In the middle of all that are sub-contractors (but many of those are also being absorbed by the big guys or driven out of business by the expense of development or production). We also shifted
from a production decision based on expected performance back to competitive "fly-off." There are very extensive reports of various phases of those tests with supporting documentation for the decisions that are made. They are not readily available to the public as a rule. There have also been lots of articles and even books outlining the history of a project. Usually those are very speculative as the 'real' reports are classified for a number of years after the fact. That is purposely done to avoid revealing any shortfalls or vulnerabilities.
As you pointed out, often the systems from a failed proposal wind up as modifications to a system that was produced. Even that is based on a proposal, testing and then a production modification decision. Those decisions relate to the -x mod after the airplane type. (F-16 block 40 for example). There are clear and distinct descriptions of what the 'Block 40'
adds to the airplane. Like many things in the 'modern' world the complexities begin to overwhelm the reality of the end game. Negotiating the maze of procurement can be a mind-boggling experience.
