Steve, I agree 100% with market research. The first instinct anyone should have to look for what has worked where. Too often government thinks it is looking for something new, when it isn’t, or has a new problem, when it doesn’t. I think we can do a lot to help with market research, too, as a community of practice. Better “apps” stores will help. Better product wikis, tools like Civic Commons Marketplace (disclosure: a Code for America project), along with means for government to express challenges to which the community can respond, ex., SIMPL. A recommender system would help “govvies who looked at/bought this prodcut also looked at X, Y, Z…”
I’m not sure I agree that “we need an ERP is too vague.” It’s only too vague if there isn’t a competitive market for the product or if ERP is actually the wrong solution to the perceived problem. I think a competitive market now requires open source vendors, too, because no off-the-shelf solution is ever perfect for your needs. If you have the need ability to hack the product then you should be able to. What gov should not have to do is start from square one with a new RFP to reinvent an entire wheel just because the valve cap is the wrong color. That’s what they do now.