States must run aging systems today while building toward a future state, and most treat this as one undifferentiated portfolio, which guarantees the urgent crowds out the important. Most states cannot keep pace with legislative change as it is, and prioritization collapses into whatever is loudest. Some modernization efforts have attempted to solve this by organizing large vendor contracts around modules or systems, occasionally producing local optimizations within those system boundaries, but this work rarely spans an outcome that a person would experience.
The first discipline of any transformation program is knowing which kind of work is which. We recommend a horizons model: Horizon 1 is running and stabilizing the current environment; Horizon 2 is improvement within current structures and constraints; Horizon 3 is transformation toward a fundamentally different future state, which cannot be reached by accumulating Horizon 2 work no matter how much of it is done.
The classification matters because it exposes an uncomfortable national fact: we are not aware of any state doing anything that resembles effective Horizon 3 transformation. System replacement projects and the modularity construct MITA introduced have produced ecosystems that are not only inflexible but actively hostile to the experimentation transformation requires. What the current modernization ecosystem mostly produces is what we would call Horizon 2 minus: work framed as improvement that actually deepens dependence on the systems and structures that must eventually be replaced, adding constraints while consuming the budget and attention that transformation would require.
This has a direct implication for the standards program’s design. A standards program is a natural fit for Horizon 1 and Horizon 2 work: guidelines and guardrails that let changes be made safely within the existing ecosystem. The danger is what that fit factors out. Standards built for the existing environment almost entirely exclude innovation, discourage challenges to processes and policies that are not working, and enshrine "this is how we do things here" into every conversation. A standards program that governs only Horizons 1 and 2 will make the current state safer and more durable, which is the opposite of transformation. The recommendations in this response are Horizon 3 recommendations.
Supporting Horizon 3 requires treating it differently from everything else, and CMS should consider making that structural: a Horizon 3 function within CMS, separate from the functions that support existing environments, with its own pathways for funding, relief, and evidence. It could begin as a single team whose charter is exploring holistic alternatives for improving program outcomes, operating as the experimentation engine for the ecosystem.
Its method would invert how federal direction usually works: test new policy approaches and technical directions through end-to-end pilots in willing states, and issue broad guidance only after something has worked. CMS should also establish standards for engaging in Horizon 3 mode itself: a published Horizon 3 strategy, an identified team qualified to perform Horizon 3 work, and a commitment to experiment with CMS on new approaches. Minnesota illustrates both the readiness and the gap: the state published exactly such a strategy, but without a Horizon 3 support structure to protect it, even a published strategy advances only at the pace of existing government process.
Legacy decommissioning and workforce development both belong inside this frame. Decommissioning is not a project to schedule; it is an outcome that becomes available when a Horizon 3 replacement has proven itself, with archival standards applied at that moment. And the workforce skills gap closes fastest when state staff work inside Horizon 3 delivery, learning modern patterns by shipping them, rather than through training curricula detached from real work.