7 Attributes of a Successful IT Project
Insurance Experts' Forum, October 31, 2013
For my inaugural blog for Insurance Networking News, I thought Iíd start with an easy subject. Just kidding, I chose the Health Care Exchange debacle thatís unfolding in conjunction with the Affordable Care Act.
Based on what Iím seeing, calling it a debacle is putting it mildly. Itís turned into a war of words between politicians, contractors and software vendors. Instead of just blindly commenting, I decided to go to the site and try to use it myself. That experience revealed a lot.
But before I dive into how bad the experience was, letís talk about how many precepts of great project management the group implementing the system seem to have violated. In the end, it appears that this is a health care system designed and marketed by politicians and an enrollment system designed by lawyers. To make matters worse, it seems that no one was really in charge.
The most important things for any project are 1) good design, 2) proper estimation, 3) defined accountability, 4) proper funding, 5) transparency, 6) end-to-end testing and 7) high quality resources.
Design is critical and it seems like they made this more difficult than it needed to be. By tying together components that had not been linked together before, they increased the complexity. The question must be asked, would it have been more prudent to choose one of the many software vendors who already provide this kind of exchange to carriers and brokers today? There are a number of them who have been in existence for years and their software is driving many of the state exchanges today that are working. It seems that the design is overtly complicated and has too many parts to it. This increased the integration complexity and the need for more testing.
No project can be delivered without good specifications, but it appears that many of the regulations were delayed in being published until after the last Presidential election. And there are tens of thousands of pages of regulations. Who can program that? That put the team in a bind because it meant that there was a lack of specificity of what needed to be done until late in the process.
Because health care is regulated on a state basis, and now also on a federal basis, that means you can have multiple variations for each major component of the system. Our method of regulation cries out for a single regulator, but our current system defines each state department of insurance and the Feds as regulators who can introduce changes and complexity. Thatís something that increases the degree of difficulty for the teams trying to deliver the system and drives costs way up.
Accountability and ownership are the keys to any project, but it was only in the last week that the government appointed one of their contractors to be the general contractor for the project. That violates every rule of large-scale projects: somebody has to own it end to end.
Testing is one of the most important parts of any project. Based on the fact that the project only allocated two weeks for integration testing, itís safe to say that almost every one of the best practices related to testing were ignored or violated.
This should have been followed by a limited pilot rollout and serious volume-based stress testing. But they wanted to go live for the entire country simultaneously, not a great way to do scalability testing, in fact itís awful.
So, itís going to take time to get this fixed and the noise on Capitol Hill will only get louder while not contributing to the fix.
So letís talk about the system itself. The process to register myself and my family took an hour, seriously! I found the user interface to be clumsy; it has some seriously insane password and username requirements and the flow of the website looked like it had been pieced together instead of being well designed from a user experience perspective.
After finishing the sign-up process (it told me I had successfully completed), I went to try to purchase a policy and the system told me I had not completed my sign-up! What? After three trips to try and go back into the sign-up to find out what might be wrong and it not letting me, I gave up.
A good line to say at this point might be ďDonít try this at home unless you are a professional!Ē But I am a professional, and have had teams that worked for me that have built and implemented multiple exchange platforms to sell these kinds of products and I still couldnít do it. Based on this experience, I predict now that the average user will be so frustrated in their attempts to try to use the system that a majority will end up using the contact center. That will skyrocket costs and lengthen the wait times for the call centers as the deadlines get closer. And that is a recipe for disaster. If they donít sign up enough people, the pools will not be large enough to actuarially support the products being sold and that will create a financial disaster for the carriers. And if people donít sign up, they get fined. This is a recipe for disaster.
Hopefully they get things straightened out quickly, but Iím not holding my breath. Maybe I will actually be registered in the system in time for my next blog; maybe not, letís wait and see.
This blog has been published with permission from Mike Boyle, CEO at Perseus Technical Strategies LLC.
Readers are encouraged to respond to Mike Boyle using the ďAdd Your CommentsĒ box below.
The opinions of bloggers on www.insurancenetworking.com do not necessarily reflect those of Insurance Networking News.
Add Your Comments...
If you have already registered to Insurance Networking News, please use the form below to login. When completed you will immeditely be directed to post a comment.
You must be registered to post a comment. Click here to register.