The recommendations above share one premise: the MES ecosystem’s problem is not that it lacks standards, technology, or money. It is that the ecosystem cannot learn quickly, and every structure in it (procurement, funding, certification, governance) was built for a world where being right upfront was possible.
Taken together, these recommendations describe a single thing: what a CMS Horizon 3 capability would look like. Classify the work honestly, protect the transformation work from the environment it is trying to change, run end-to-end slices that serve real people, fund small and decide often, filter vendors on demonstrated performance, let standards be ratified from what works rather than authored in advance, and measure everything as trendlines that matter to a person. None of these pieces is sufficient alone; together they are a working design that CMS could stand up at the scale of a single team and a single willing state.
A standards program built as another correctness gate will join the existing stack of well-intentioned compliance layers. A standards program built as learning infrastructure (shortening lead time, moving decisions to the work, letting standards emerge from running slices, filtering on demonstrated performance) would be the most consequential thing CMS has done in this space in decades.
I would welcome the opportunity to discuss any of these recommendations, connect CMS with state colleagues who have lived this work, or contribute to the practitioner-led standards development structure recommended above.