As I began thinking about this article last fall, the clock was ticking for officials charged with overhauling the much-maligned HealthCare.gov insurance marketplace. They needed the site to be functioning smoothly for the vast majority of users before a self-imposed Nov. 30, 2013, deadline.

Since the portal's fateful launch, the Obama administration has been under fire from the American public and congressional lawmakers in both parties — not to mention at the mercy of the late-night talk shows — as error rates, slow response times and repeated system outages made it difficult for many new consumers to complete an application for health coverage.

While we have been witness to everything from public testimony about what went wrong, to press briefings detailing the progress on fixing the issues list, to acceptance of full responsibility from the commander in chief, I suspect that by the time you are reading this in February, the site will be stable and operating at its intended capacity with greatly improved performance.

However, if there ever were a need to hit a magic "restart" button for a high-profile technology project, this surely would have been it.

Having spent my entire career in information technology, both as a health care chief information officer and on the vendor side, I've wanted to reach for that restart button a time or two myself. Like many in the industry, I have witnessed a few technology projects get into some serious trouble. Of course, this can happen in a few different ways:

  • a meltdown — the system is fundamentally unstable, nonfunctional or both;
  • a timetable or budget overrun — the scope and cost are severely off or underestimated;
  • a feature or function that does not meet expectations — it missed the mark on what was needed by a mile.

While none of the above scenarios is particularly pleasant, they serve an important purpose. As Oscar Wilde once said, "Experience is simply the name we give our mistakes."

Indeed, battling technology troubles can help us acquire the key knowledge that can be gained only from the heads-down, 24/7 war room work that usually is required to get a wayward project back on track. No doubt those in charge of HealthCare.gov will emerge from the experience a little worse for the wear, but also a whole lot wiser.

Doomed from the Start?

Now, to be fair, unlike the private sector, the government faces special challenges when it comes to technology undertakings. For one, there is a tremendous amount of political oversight and transparency on these projects. This is, of course, good for all the obvious reasons, but it is trying in the same way that too many cooks in the kitchen can produce a less-than-appealing dish. As we know, this administration faces no shortage of competing agendas and dueling priorities.

The government's cumbersome procurement process also may put it at a disadvantage, as a cry for help or extra resources are not simply a phone call away. Moreover, the best vendor for the job, which could be a small startup with novel ideas and innovative technology, may not have what it takes to make it through the arduous procurement process, which typically favors large, established vendors.

Additionally, a government-run IT project can face a complex and shifting set of requirements as well as compressed timelines to get the work done. Oftentimes the ink is barely dry on the legislation when the deadline for the needed technology is established. Furthermore, the regulation-writing process can be influenced by the election process; promises made on the campaign trail need to appear in regulation shortly thereafter. Tight timelines and volatile requirements can lead to a project's coming off the rails in fairly short order.

Failure to Launch

Despite the fact that many industries have been trying to improve their project management prowess for decades, IT project troubles are not unusual; nor are they confined to a particular industry. In fact, the failure rate for technology projects is sobering — anywhere from 37 to 75 percent, according to the multitude of studies available.

Consider a 2012 McKinsey & Co. survey on large-scale IT projects (those with initial budgets greater than $15 million) which found that half of all large IT projects blow their budgets in grand fashion, running 45 percent over budget and 7 percent over time. Moreover, they found that 17 percent of large IT projects go so badly they can threaten the very existence of the company.

Or, in the case of the federal government, a project stumble with the eyes of the nation watching (and waiting to log on) only exacerbates the nation's already-heated political divide, particularly over the Affordable Care Act.

Now, we in the health care IT industry may question why it appears the government failed to employ the R&D best practices many of us have come to know so well. These include using an iterative development and delivery approach, establishing betas or pilots, and forgoing the risk of a big bang rollout. And we may wonder which of the most common causes of IT project failures — lack of clear governance or sponsorship, ambiguous requirements, insufficient resources, poor design or inadequate testing, to name just a few — plagued the portal's initial rollout.

Since it was likely a confluence of many of these factors, a refresher on the steps that can be taken to help minimize the risk of failure might be useful for those about to embark on a new IT initiative. Consider the following:

Good project plans. There are several cues that a project plan is as solid as one can make it at the inception of the project. For example, the project charter is clear and explicit, project timelines and staffing needs have been reviewed by multiple parties for reasonableness, and task timelines have some "slack" built into them. Furthermore, needed resources and budget have been committed, and accountabilities for each plan phase and task are explicit.

Great project managers. A great project manager is exceptionally valuable. To a large degree, the reputations of project managers precede them. If project managers have proven themselves over the course of many projects, then their plans are likely to be sound. If project managers are novices or have an uneven track record, their plans may require greater scrutiny. First and foremost, a great project manager must have the interpersonal and organizational skills to bring people together and make things happen.
 
Tight interaction with users and lots of feedback. Failure to view an IT project through the eyes of the users and stakeholders often leads to failing to get their buy-in as well as designing systems that appear, to the user, to be from Mars. Therefore, it's critical to involve users in every phase of the project and as early and often as possible. As we know in health care IT, users may not know what they want until they see a particular feature or function, or try to incorporate it in their daily workflow. Waiting until a project has gone live to obtain stakeholder feedback is nothing short of asking for trouble.

Solid governance. Without the appropriate level of oversight, projects can quickly head off track. The key role of a governance structure is to enable, facilitate, manage and direct the successful implementation of the project's changes and impact into a well-prepared and informed organization (or public) to maximize the potential benefits. A solid governance structure also provides clarity on the various levels at which decisions are escalated, and assigns accountability to the decision-makers.

Great project management discipline. In any well-run project, there is strict adherence to a set of management disciplines and processes developed over time. While skilled project management does not provide a guarantee for project success, it certainly can reduce the risk of failure, particularly for large, complex IT implementations.

Well-conceived phases. A large-scale master project plan must be divided into more manageable pieces that can be delivered incrementally. Thinking in terms of phases helps to ensure that the deliverables produced at the end of each phase meet their purpose, and that the project team is properly prepared for the next phase. While project teams must have the wherewithal to keep the big-picture end game in sight, they also must possess the discipline to execute successfully the myriad tasks that make up each phase of a project.

Thoughtful and appropriate risk management. Projects are full of uncertainties, and failure to identify or manage those uncertainties appropriately can turn them into serious problems. Some examples of project risks include unproven information technology, a deterioration in the organization's financial condition and turnover of project staff. A thorough assessment of risks related to the technology and its impact, the organization's ability to absorb change, project scope and size, schedule, and cost can provide a good starting point for developing a comprehensive risk management and mitigation plan.

Solid project monitoring and openness to candid feedback. Being closely involved in a project day-to-day makes it difficult to step back and critically reflect on progress. Yet, such reflection is crucial to success. Regular and frequent project review meetings provide the ability to discuss overall project goals and objectives, as well as to assess progress toward specific milestones. Additionally, implementing a continuous feedback loop between project managers and stakeholders can help to identify actual or potential problems as early as possible to facilitate timely adjustments to the project plan.

No Guarantees

Even with a sizable collection of best practices and defined procedures and methodology applied in project management — not to mention ongoing advancements in the field — the risk of getting into IT project trouble is sizable.

But the way an organization responds when it finds itself in a compromised position can vary based on the nature of the problem. For example, if operations are paralyzed through a complete technology failure, this may elicit a crisis mode response; whereas, encountering a substantial budget overrun typically would not throw an organization into a crisis management situation.

Nonetheless, any project setback requires swift and efficient management to contain the situation and minimize the negative impact. As such, the following tips offer practical guidance for managing technology troubles large and small:

  • Keep cool; emotions are generally not helpful.
  • Make sure that you really know what is wrong and why; at times, the hysteria surrounding a rocky project distorts the understanding of the nature of the problems.
  • Perform a quick assessment of whether you have the right people on the bus to drive toward a resolution.
  • Protect those who have to repair the damage, providing them with resources, time and political air cover.
  • Provide lots of communication and transparency to the organization, particularly those feeling the daily reality of a troubled undertaking.
  • Engage stakeholders in diagnosis and remedy plans.
  • Insist on very disciplined management of the "get out of here plan."
  • Ensure accountability, but make liberal use of the pronoun we.
  • Exude strength and confidence even if your anxiety causes your intestines to shrink to the size of a golf ball.

Always a Lesson to Be Learned

As bad as the launch of the federal online insurance marketplace was, there are numerous examples, both in the public and private sector, of similar situations. Indeed, every IT implementation is fraught with risk and presents a genuinely difficult undertaking.

We are wise to treat these projects with respect and humbleness. We are also wise to know that with every misstep, true leaders will rise to the occasion from all levels within the project. This is something to behold.

Despite every organization's best efforts to be prepared and minimize risk, the odds of success may not always be in one's favor. The key for all those involved in such misfortunes is to learn from past failures and to put those lessons learned into action on future projects.

John Glaser, Ph.D., is the CEO of the Health Services business unit of Siemens Healthcare in Malvern, Pa. He is also a regular contributor to H&HN Daily.