So the snicker-inducing crash and burn of ObamaCare has been plastered all over news recently. In congressional testimony, the head of Medicare and Medicaid Services pointed the finger at the contractors for technical difficulties and Sibelius, Secretary of Health and Human Services, blamed insurance companies for cancelling plans.

Having been a strategist on multiple IT projects intended to serve about 60,000 users globally over the course of my career, including a customer information data integration project, I have a bit of a unique perspective on the technical challenges involved in accommodating at least that many users. With the level of systems integration required, interface with systems that verify income and eligibility, others that calculate benefit amounts, and yet others that provide insurance plan interfaces; the number of vendors involved, at least three major ones; the number of users intended to be accommodated, well into the tens of millions, it is my judgment that the project could never have been expected to be accomplished in a year – through design, testing and implementation even if it were an entirely internal project – which is about how long they had to make the entire thing ready considering that last major systems contract was not awarded until late last year.

ObamaCare IT was doomed to failure before the project was started, just from a project management/systems integration standpoint, never mind the actual base coding project and infrastructure accommodation, with or without states that declined to host the exchange (although I am sure the need to have additional capacity confounded the effort). Add in the OBTW’s (Oh, by the way) and technical gotchas to be worked though at every level and it is likely a feat that any users at all were successful in signing up.

Greenfield, massive projects are difficult and time consuming in the respect that the slate of lessons learned from growing a system over time is blank, nearly everything in the blueprint is theoretical and everyone is on a learning curve. The more layers of management in between the integrators and the systems themselves, as from multiple companies and multiple portions of bureaucracy, the more difficult execution becomes and the longer it takes. Not even throwing money at it, to the tune of $500M as has been reported as the current price tag, can make it work in the time frame they had. I doubt it will be fixed in a month. It might limp along by spring if they are lucky.

A good portion of blame lies with DHS for foot-dragging on developing requirements and standards; it should have started on those nearly the second Obama signed the ACA into law if not before, which it did not do until mid-2011. Even then, I think three years would have been the minimum in planning and execution with expert top-down project management, the right people in the right places.

There should not have been a national rollout, at least I wouldn’t have attempted it. The country should have been divided into sections, possibly 6-10 regional sections, with a phased rollout on a voluntary basis starting in late 2012 to build up that slate of lessons learned and fix what’s broken along the way to avoid leaving people stranded without insurance as the individual mandate deadline looms and to avoid embarrassment on a national scale. Instead, they did not even start the base coding project until May 2013 – at least a year too late.

One does not simply walk into Mordor. There is evil there that does not sleep. Not with 10,000 men could you do this. It is pure folly. – Boromir – Fellowship of the Ring

Advertisements