I think 2 challenges to this are a) outlining challenge scope in a way that conveys actual needs, and when doing so potentially reveals procurement sensitive information, having a good fake alternative problem that can map closely to the actual needs or b) ( in trying to accomplish what Greg suggests) having a good idea of what projects are truly similar in or across agencies. There is a lot of not actually knowing what other similar projects might exist, and there is no resource to make sure people could identify one at the right time. Although there probably is a department that has enough information to do so, GSA ( id think).