The Failed Launch Of www.HealthCare.gov
The US Government's failed launch of the Healthcare.gov website highlights issues with integrating technology into a large bureaucratic organization.
“I’m going to try and download every movie ever made, and you’re going to try to sign up for Obamacare, and we’ll see which happens first” – Jon Stewart challenging Kathleen Sebelius (former Secretary of Health and Human Services) to a race.
www.HealthCare.gov Overview
The affordable care act (increasing healthcare access to all Americans), signed by President Obama on 23 March 2010, required that the United States Department of Health and Human Services (HHS) launch HealthCare.gov on 1 October 2013. This website going to be the official health care exchange that would allow residents to compare prices of health care plans, identify if they qualify for federal subsidies, and enroll in a chosen plan. The Affordable Care Act gave states the right to create their own healthcare exchange or opt-in to the federal exchange (healthcare.gov).
Website Rollout
Healthcare.gov was officially launched on 1 October 2013 covering residents of 36 states that did not create and manage their own healthcare exchange. Problems with the website were apparent immediately. High website demand (250,000 users [5 times more than expected]) caused the website to go down within 2 hours of launch. While website capacity was initially cited as the main issue, additional problems arose mainly due to the website design not being complete. Users cited issues such as drop down menus not being complete and insurance companies cited issues with user data not being correct or complete when it reached them.
In addition, the websites login feature (which is the first step to accessing the website) could handle even less traffic than the main website which created a huge bottleneck. Due to poor planning, this same log in method was also used by website technicians, making it extremely difficult for them to log in and troubleshoot problems.
A total of 6 users completed and submitted their applications and selected a health insurance plan on the first day.
Through a large amount of troubleshooting, bringing in new contractors, and increased management, the website could handle 35,000 concurrent users at a time by December 1 and a total of 1.2 million customers signed up for a healthcare plan by 28 December, when the open enrollment period officially ended.
Root Causes
The myriad of problems experienced during the websites rollout were due to several factors, primarily:
- Lack of relevant experience. HHS employees and managers had a lot of experience with private insurance markets and maintaining large government projects, but did not have required experience in technology product launches. Key technical positions were unfilled and project managers had little knowledge on the amount of work required and typical product development processes leaving very little time to test and troubleshoot the website.
- Lack of leadership. There was no formal division of responsibilities in place between the many government offices involved which caused a delay in key decision making or a lack of communication when key decisions were made. For example, the contractor responsible for the log on system estimated a low demand because the initial website plan included the option to shop for products without creating an account or logging in. However, due to technical delays, that functionality was removed from the initial website launch (so all users would need to log in) without increasing capacity.
- Schedule pressure. Since the launch date was mandated in the Affordable Care Act, HHS employees were pressured to launch on time regardless of completion or the amount (and results) of testing and troubleshooting performed.
Key Takeaways
The key issues discussed above resulted in the rollout of the healthcare.gov website ballooning the initial $93.7M budget to an ultimate cost of $1.7B.
It is easy to observe that the launch of healthcare.gov was a major failure, but this isn’t a particularly unique case. Research has shown that over the past 10 years, 94% of large federal information technology projects were unsuccessful, more than 50% were delayed, overbudget, or didn’t meet expectations and a total of 41.4% were judged to be complete failures. I contend that most of the root causes identified for healthcare.gov are not unique and it is not a surprise the federal government struggles with adapting to new technology. A large, bureaucratic organization that has significant experience in core government policies is likely not adaptable to behaving like a “start-up” and successfully launching new technology.
Based on the number of government organizations that all have policy experience and will likely have to adapt to technology challenges in the future, the government should either create a new government office strictly responsible for handling technology products for the rest of the government or make a significant investment to ensure that all government offices have the experience, management tools, and resources required to integrate technology into their offices.
(782 words)
1 – Office of Inspector General, HealthCare.gov: CMS management of the federal marketplace: a case study, Available at: https://oig.hhs.gov/oei/reports/oei-06-14-00350.pdf [Accessed November 18, 2016].
2 – Goldstein, A., 2016. HHS failed to heed many warnings that HealthCare.gov was in trouble. Washington Post. Available at: https://www.washingtonpost.com/national/health-science/hhs-failed-to-heed-many-warnings-that-healthcaregov-was-in-trouble/2016/02/22/dd344e7c-d67e-11e5-9823-02b905009f99_story.html [Accessed November 18, 2016].
3 – Healthcare.gov. Wikipedia. Available at: https://en.wikipedia.org/wiki/Healthcare.gov [Accessed November 18, 2016].
4 – Johnson, C. & Reed, H., 2013. Why the Government Never Gets Tech Right. The New York Times. Available at: http://www.nytimes.com/2013/10/25/opinion/getting-to-the-bottom-of-healthcaregovs-flop.html [Accessed November 18, 2016].
As you nicely pointed out, the launch of the national healthcare exchange website was a disaster. It really raises key questions as to how large projects within the civil service are run, and one must consider how the government can leverage the nation’s technological expertise in order to push forward, not backwards its agenda. For instance, could the government have worked with a Silicon Valley startup in order to develop a solution which was much more nimble at a fraction of the cost? Strong leadership and project management skills are undoubtedly key. Nonetheless, it is great that the exchanges ultimately worked. Whilst one can argue with certain points of implementation, the Affordable Care, by insuring 15 million previously uninsured people, undoubtedly had a positive impact on social justice in the USA. The recent election threatens this triumph, however one hopes that expanding healthcare accessing will remain a key agenda of the new administration.
“Based on the number of government organizations that all have policy experience and will likely have to adapt to technology challenges in the future, the government should either create a new government office strictly responsible for handling technology products for the rest of the government”
The USG did create a government office focused on digital services in the wake of the healthcare.gov debacle. The General Services Administration (GSA), a Federal agency, now houses 18F, a digital services agency build on the lean startup model. 18F recruits top talent via the Presidential Innovation Fellows program and acts as an in-house digital delivery team to help government agencies build effective, user-centric digital services. 18F runs on a cost recovery model where client agencies reimburse 18F for its work. See https://18f.gsa.gov/ for more information.
18F is well-suited to help agencies for projects on a smaller scale than healthcare.gov. I’d argue that for future high stakes, large scale roll outs, however, the U.S. government should contract out more of the work to experts in the private sector. The government currently does not have enough talent/expertise to effectively manage such projects from start to finish, and even if it did, internal bureaucratic roadblocks would likely stymie and frustrate that talent and contribute to project delays.
Great post — we don’t talk enough about the effects of digital transformation on the public sector. In the case of Healthcare.gov, this was a very visible event given the amount of attention and dialogue around the Affordable Care Act.
The first root cause you mention — lack of relevant experience — is spot on. It took a highly skilled team with a strong Silicon Valley pedigree to come in and address the many technical issues that occurred during the rollout. One reason is because there are only a handful of web properties that handle the same traffic volume and backend complexity that Healthcare.gov requires — and they are all based in the Bay Area. There was a great article about this in The Atlantic: https://www.google.com/amp/www.theatlantic.com/amp/article/397784/
That same team has gone on and founded Nava, which aims to solve many of the technical challenges facing the public sector. Along with Palantir, they represent companies with the relevant experience and nimbleness to attempt to fix such problems at a national scale. Perhaps the role of the Public-Private Partnership is key to the successful digital overhaul of the public sector: companies such as Nava leading the transformation, with the support and resources of the Federal Government.
Your point about better leadership is also key. The White House did not have a CTO until 2009, and did not have a CDO (Chief Digital Officer) until April 2015; and I speculate the latter appointment occurred in part due to the Healthcare.gov fiasco. Having the right leadership is not only important from an organizational and decision-making perspective, but also the position is very visible and with the right person in place, others from the technology sector may consider leading Silicon Valley to solve important problems in Washington DC.
I think the main reason a lot of these large IT projects fail is because organizations think that large project = large number of people working on project. In technology, you can have a small team launch to millions of people.
While the rollout of the healthcare.gov website was unquestionably a disaster, I can’t help but wonder what (if any) lessons were learned by the federal government following this situation. I find what was mentioned in the above comment interesting, that the government responded to this crisis by creating another office focused on digital services. While this department is meant to be built upon a lean startup model, I can’t help but be skeptical that the bureaucracy of the federal government will overwhelm any benefits that this group may be able to offer. Instead, I wonder if the government will ever look to outsource any future technology projects to the private sector/Silicon Valley. I believe these companies are better equipped to provide better technology at lower prices than any government office, which doesn’t have a core competency in developing and providing these services. Integrating digital technologies is vital as the government looks to lower costs and increase efficiency, and I hope that they will look to these sorts of key partnerships in the future so as not to repeat errors such as this.
Thanks for sharing such an insightful post on a relevant topic. I am surprised that any website made in 2010 have the problems you mentioned, even if made by the government. Every web developer should know that capacity planning and designing for key user journeys are among the most important aspects of creating a successful site. I am certain that Google or even a “random” startup in Silicon Valley could’ve done a better job in less time, especially given $1.7B for a single webservice.
As you mentioned. There is a big difference between private companies, for which every minute of downtime means lost revenue, and the government. People waiting online to sign up for the healthcare act does not, apart from bad reputation, have a direct impact on the people responsible for launching the site. Yet, these failures costs a lot for society, with all the manhours spent waiting online.
In your Key Takeaways, you also mentioned that the government should create a new office for technology products. Yet the separation or investment by itself does not guarantee better products. The Government faces the same challenge in attracting talent that a company like Google does. Good talent means good products, and what top engineering or design talent wants to work for the Government?
Again, very interesting post. Thanks for sharing!
Interesting post! I agree that a huge takeaway here is having the right resources and people in place for large technology projects, and that there could be great benefits from more public-private partnerships. Interestingly, Obama may have learned a bit from this situation and has been increasing his efforts to create connections between Silicon Valley and Washington (http://www.nytimes.com/2016/10/30/magazine/barack-obama-brought-silicon-valley-to-washington-is-that-a-good-thing.html?_r=0)
Great post, thanks for sharing! In conjunction with the initiatives you mentioned in setting up a technology office in government, I believe that the government should have a long-term plan to grow its technology office and decrease its reliability on technology consulting firms such as Accenture and other from Silicon Valley to handle crisis management once a mishap similar to this occurs. That is a typical example of how projects go over-budget and result in wasted taxpayer money.
Furthermore, as you mentioned a lack of leadership was clearly evident. For instance, the program could have been rolled out incrementally across the states, giving the technology team in the white house to trouble-shoot problems on a smaller scale.