A reliable predictor of modernization failure we have observed is decision distance: the number of organizational layers between the people doing and receiving the work and the people deciding how the work will be done. Every layer adds delay, subtracts context, and increases the chance that the decision optimizes for something other than the outcome.
In a Horizon 3 program, this means something more fundamental than including practitioners in standards committees. It means replacing standards-making groups with structures designed to be immediately responsive to those attempting real end-to-end transformation work.
Invert the accountability: instead of practitioners waiting on a standards body to author, decide, and publish, the standards function is accountable for responding, within days, to the teams doing the work, whether that response is clearing a barrier, granting relief from an existing standard, or rapidly ratifying a new standard or approach that a working experiment has proven.
This is not only a federal problem. States replicate the same structure internally: architecture review boards, enterprise standards, and security review queues that decide at a high level what teams must conform to, justified by the need to fit the existing system, reinforcing the current state rather than transforming it. The inversion needs to apply at every layer: when a team's experiment proves something works, it must become the basis for the Horizon 3 standards being established at both the state and federal levels. A state selected for a Horizon 3 pilot with CMS asking for enhanced responsiveness should be prepared to offer its own teams the same.
When teams have piloted something that works, the standards function’s job is to make it a standard in support of that working experiment, at the speed the experiment moves. Section S10 describes where these experiments come from and the conditions they require; this is the governance structure that must sit behind them, and it is the ultimate change needed to make pushing decision authority downward real rather than rhetorical. We have drafted this mechanism as proposed legislation, linked below, including specific response-time obligations for each governance layer and public records of every escalation and resolution.
This principle should extend to domains CMS might assume require central specification. Consent and privacy design is a concrete example: consent models work when they are designed at the point closest to the work, informed directly by the people who must navigate giving consent, so the result is both usable and secure. A federally specified consent framework designed far from any actual consent interaction will be neither.