Releasing buggy, highly visible, mission-critical software is an invitation to disaster. However, that’s exactly what the government did with healthcare.gov.
The government insisted on an October 1 launch date for healthcare.gov, even after simulation testing by just a few hundred people crashed the system and a test group told them not to launch because it was buggy. Sure, delaying the launch by a few weeks would certainly have drawn criticism – but not nearly as much as it has gotten now. Plus, once software has been released, fixing bugs becomes much more complicated.
Days before the launch… government officials and contractors tested a key part of the Web site to see whether it could handle tens of thousands of consumers at the same time. It crashed after a simulation in which just a few hundred people tried to log on simultaneously.
About a month before the exchange opened, [a] testing group urged agency officials not to launch it nationwide because it was still riddled with problems.
Some key testing of the system did not take place until the week before launch.
Doing major testing a week before launch is inexcusable. You won’t be able to test everything or have time to fix things (since fixing bugs will require more testing.)
One key problem, according to a person close to the project, was that [federal agency Centers for Medicare and Medicaid Services] assumed the role of managing the 55 contractors involved and had not ensured that all the pieces were working together.
A federal agency that clearly had little if any experience overseeing big software projects should not have been managing it. Sounds to me like the whole thing was the usual DC gravy train for politically connected contractors and consultants who aren’t known for delivering quality products.