The new website — healthcare.gov — the Obama Administration commissioned to help guide people through the health care enrollment process has been published and almost everyone in the country is calling it a huge flop. It is difficult to use, has a huge lag time, and most of the time just will freeze up on a user, which does bode well for increasing the number of people who have health insurance as required by the Affordable Care Act. This really doesn’t look great for the administration.
Let’s put politics aside — I am not going to entertain any comments related to the viability of the Affordable Care Act or if you think “affordable” as a misnomer. I am neither attacking or supporting the Obama Administration as part of this discussion. What I want to do is examine what caused the problems with the site and the approach that will be required to provide the functionality that the enrollment interface requires as part of healthcare.gov.
If you have ever done contract work for the Federal government, you know what a treat it is to have them as a partner. The culture within government is to blame for a percentage of the failure, but that culture will not change. I am going to leave that out of the analysis of why the project failed. For me, here are the two main reasons.
- Too many contractors without coordination. There were 55 contractors who were tasked with building out portions of the site. I am sure there were requirements detailed for the functions of each component, there was not any coordination or testing required by the contractors to make sure their function actually worked as part of the whole. The overall testing of the site was up to Health and Human Services (HHS) Department. One contractor actually was assigned two components — the data hub and the accounts registration interface. The ironic thing is the data hub works but the registration interface does not. While there may have been a project director within HHS, they did not do their job in assuring that each contractor was able to provide a workable piece.
- Lack of iterate testing. Several contractors have already testified in front of Congress that while they tested their component, HHS failed to test the site before it was rolled out. What? Why didn’t HHS require that contractors provide their component so testing related to the integration of all pieces could have occurred? Best case scenario they were waiting until the end to actually test the site to make sure it actually worked. There was no QC on this project. How do I know? If there would have been, HHS would have known healthcare.gov wouldn’t work and should not have been launched on October 1st.
- Total focus on constraints instead of value and quality. HHS was only concerned with making sure healthcare.gov live on October 1st (schedule), that each contractor did not exceed their contracted amount (budget), and they provided a site that had the goal of helping people purchase affordable health care (scope). While those are the traditional project management constraints, they absolutely ignored the two Agile Project Management (APM) platforms of value and quality. Does healthcare.gov provide any value to the customer/public? Well, if it continually freezes and the user cannot enroll in a health insurance plan of their choice, then there is no value. The concept is there, but without functionality, all we can say is the site is a failure.
So what to do? You guys know I prefer simplicity to overly complex plans. This situation isn’t much different from when your outputs don’t meet specs or requirements, which means we have to adapt and return to the speculate and explore phases. I know the Obama Administration has appointed one of the contractors as a lead on the project, which is a start. But I gotta say I don’t think it is a good start. So to Secretary Kathleen Sebelius of HHS, I would suggest the following start:
- Find a project manager. Do this quick and do it now. Have them hire a team of however many they need to help them manage the administrative side of this project, but find someone who will be able to interface successfully with the contractors. I would say hire John Wayne, but he is unavailable right now. I do think both Jason Bourne and Tony Stark are both available and would probably be my choice at this point. I would pick Jason Bourne because of his ability to “lead” people and Tony Stark for his technical skills. If you put them both in charge I am not sure who would be standing after their first disagreement.
- Form a project team. Sounds easy and I think it would be. You need an engineer representing each module of the site on this team. Leave the corporate executives at home and let the engineers and programmers and engineers that worked on each component work with the project manager. I would also require the engineer to be on site on a non-disclosed location to keep them focused and without the distraction of constant pressure from politicians and the press. Hold a daily SCRUM to foster communication, collaboration, and accountability.
- Allow the project team to create a vision based on the old project scope. Give the project team the freedom to craft a new vision of the site and how they will accomplish the objectives and outcomes you want. Don’t constrain them with a timeframe.
- Get buy-in from all project team members. This means that teams representing different companies will need to share code, ideas, and strategies.
- Adopt APM methodology to create value and quality. Your project manager will have to focus on value and quality instead of the three traditional constraints of project management. Don’t throw on the additional burden of all the paperwork and reporting that is typically required of Federal projects. If you just feel the need for a report, allow the PM to create one once version 2.0 is rolled out to the public.
- Demand continuous innovation. If the project scope requires change and improvement, and obviously it will, give the team the flexibility to make those decisions. Don’t require the approve any small change through Secretary Sebelius or President Obama. We don’t want the functionality or interface to have to go through focus groups. The project team should be committed to adapting healthcare.gov into the future so that it can continue to provide value to the customer.
I know this is a pipe dream, but those are the first steps I would take to improve the quality and value of healthcare.gov. Obviously it was somewhat of a tongue in cheek look at the huge and real problem the Obama administration that required a solution. I am going to use some healthcare cliches to close — I bet the approach will be more like putting a band-aid on a severed artery to try and stop the bleeding. The solutions will be too small for the scope of the problem and will not solve the problem. Like at all. Let’s hope that the Obama administration is able to bring to bear the resources that are required to solve this challenge.
Update: President Obama must read my site — he appointed Jeffrey Zients to head up the team to fix the site. He says it will be fixed in a month, which I gotta think is pretty aggressive. He is treating the project like a “punch list”, which you can read about here. Obviously President Obama didn’t get past step #1 in my solutions.