Every traditional procurement embeds the same assumption: that the state correctly identified the problem and the solution before the contract was signed. Timelines, budgets, requirements, and evaluation all hinge on that assumption holding. But Horizon 3 MES transformation is VUCA work (volatile, uncertain, complex, ambiguous), and in VUCA work the optimal approach only becomes visible in hindsight, after iterating. A predictive model applied to adaptive work fails predictably, and the implementation complexity, schedule delays, and cost growth the RFI asks about in V-10 are largely this single mismatch expressing itself.
The vendor-side challenge in V-1 (varying state interpretations of federal requirements) has the same root: vendors are asked to commit to fixed interpretations upfront in an environment where interpretation legitimately evolves. The remedy is not better upfront specification; it is procurement and contracting structured for learning: smaller commitments, working software as the interpretation arbiter, and course correction as a designed-in feature rather than a change order.
What counts as learning also depends on what is actually being procured, and MES procurement regularly blurs the two. When we procure implementation services, we are procuring people: a team whose capability, judgment, and fit with the state's environment can only be learned by working with them. When we procure software, we are procuring a thing to be tried in context, and learning means finding out through use whether it meets the state's actual needs. Current procurement processes prevent learning and adjusting for either of these. S5 proposes how to fix that for each.