Editor’s note: This commentary is by Heidi Spear and Cyrus Patten. Spear, of Fayston, is running for state representative in the Washington 7 District. She is a graduate of Middlebury College and has worked as a tech executive in Cambridge, Mass. She is currently board chair of Mil Milagros Inc. and Fayston Elementary School. Patten is a licensed social worker and executive director of Campaign for Vermont, a nonprofit, nonpartisan advocacy organization. He lives in Monkton.
The citizens of Vermont will be shortchanged in more ways than one if the state awards contracts for its current RFPs to revamp two major information technology projects: the Medicaid Management system and the Integrated Eligibility system.
First, the administration is directing these projects down a dated, rigid and expensive path very unlikely to yield successful outcomes. Budgeted at $132.5 million, federal and Vermont taxpayers would foot a monumental bill for Vermont to attempt to replace its circa 1980s technology with 1995 technology — and then carry hefty annual costs to maintain them. Second, in taking this path the state is all but guaranteeing that these projects will do nothing to spur our local technology economy or create the kind of jobs we so badly need in this state.
There is a better way. Given that Vermont does not have a booming technology sector like our neighbors to the south, one can understand that our leadership might not be well versed in current trends, but it is crucial that we enlighten them now.
The traditional development approach and large project scale dictated in our current open IT RFPs typically fail, experiencing significant delays and overruns. Recent studies have shown failure rates of these types of projects are in excess of 70 percent. Yes, odds were that Vermont Health Connect would take the precise course it did.
High-performing organizations no longer spend years developing tomes of functional and technical specifications and embarking on death-march coding cycles. Today’s leading technology organizations avoid the risk, expense and delay associated with traditional development methods by using Agile development. Agile involves a series of short, well controlled “sprint” cycles that are self-documenting and have proven to produce higher-quality, more innovative and less expensive IT systems.
The most successful industrial-strength software businesses are not using proprietary software or a cumbersome, lengthy waterfall development process, as Vermont is requiring in its RFPs. They are using open source (free) software stacks (think Legos) to build IT systems. Companies like Facebook, LinkedIn and Netflix are built on open source stacks to dramatically reduce licensing and maintenance fees, be more flexible and avoid vendor lock-in.
Finally, these modern IT applications that we should build are not built on expensive platforms in costly data centers like Vermont is mandating. They are built to run on much cheaper, very scalable and very secure public cloud models where compute (functionality) and storage (data) is purchased as a commodity (as needed). For those unfamiliar with the technology industry, and this terminology, these concepts might sound very fly-by-night and not at all secure. However, the reality is that even the CIA recently announced it is moving its entire IT portfolio to a public cloud model. This isn’t just for new and flashy operations.
Being more flexible and taking an agile, iterative approach to developing our IT systems would allow the state to engage small upstart vendors and include a consortium of resources including our tech schools, colleges and universities.
Large corporations and the federal government are moving to the public cloud in a huge way. The reason is simple. Large operations have an even larger motivation to transition to these new environments: a dramatic reduction in their enterprise systems development and support costs. Coupled with the increased flexibility, stability and security that everybody seeks, the path is clear to those with their eyes on this industry. We should recognize the benefits and leverage the public cloud for new technology investments.
Vermont’s current, outsized RFPs, with their retro approach to technology, don’t just translate to excessive risk, higher up-front costs, higher maintenance costs and a longer implementation timeframe, they inevitably result in our only contracting with massive firms that dominate the large project market. Young, upstart technology firms and adaptive, small to mid-sized players do not take on these projects. Entrepreneurial technologists aren’t going to undermine their career prospects working at a snail’s pace for years using decades-old processes and technology. And a snail’s pace it would be: recent, much smaller IT projects in Vermont have taken three to seven years to complete. Further, smaller and innovative firms aren’t going to take on massive contracts that attempt to shift all responsibility and liability to them heading down a very risky path. “The big boys” are perfectly used to working with legal teams to ride these roiling but profitable waters, but they are not the partners to make us successful — let alone do so at a reasonable price in the desired timeframe.
With the right leadership, putting experts rather than political allies in charge, Vermont could leverage these technology innovations and put our technology sector on the map. Vermont should withdraw its current IT RFPs and replace them altogether. We should open this process up to get the input of the talent in this state. The ambitious and creative technologists that helped launch Dealer.com are not our only ones, to be sure. We should challenge the entire field with a series of hackathons (also known as code-a-thons) to find creative ways to solve our problems in faster, cheaper and more effective ways. Yes, these are a real and widespread thing and used globally to seed production technologies.
Being more flexible and taking an agile, iterative approach to developing our IT systems would allow the state to engage small upstart vendors and include a consortium of resources including our tech schools, colleges and universities. Using modern methods and technologies would attract the next generation of talent to locate or stay in the state. Rather than engaging monolithic, out-of-state multi-national firms on large complicated contracts for over $100 million we could effectively steer this money into a massive, efficiently deployed capital infusion for innovative technology companies in Vermont.
The impact of deploying this strategy could be transformative for our technology sector. In excess of $100 million in growth capital isn’t knocking on our door and Federal investment in these projects allows us a unique opportunity. It can do what a $4.5 million investment in our IBM plant can’t.
The impact of building these systems right is consequential in the ways described above. But let us not forget that the impact of getting them right for our most vulnerable citizens is consequential as well. A lack of coordination across the Agency of Human Services is not a minor problem and when the system fails terrible things happen. These IT projects are being launched for good cause.
With dozens of funding streams and sets of eligibility criteria, the people on the front lines at AHS, working to help Vermonters, spend more time fighting their own systems than should be reasonably expected. In the human services profession, the ultimate goal is the “right services, for the right client, at the right time,” but stymied by processes and undermined by technology barriers, currently these goals are virtually unachievable.
The most vulnerable citizens among us rely on our coordinated supports, and when we don’t coordinate our supports we operate extremely inefficiently and we undermine the outcomes (safety, health and well-being) of those we serve.
Add to the challenges of integrating service stovepipes at AHS the new challenges of meeting evolving needs to support Medicaid and Medicare enrollment and transactions and the risk for completing either important project increases. There is no time or risk tolerance for massive, monolithic and parallel projects. There is no wisdom in the costly, risky, slow path our RFPs dictate. Rapid incremental improvement is needed. Current best practices and technologies and informed experts must be engaged in charting a more effective future for AHS.
Whether you care most about eliminating wasteful spending, championing our technology industry or helping those most vulnerable among us, the current roadmap for AHS systems should be promptly scrapped to make way for a far smarter approach, and one that serves all three objectives for Vermonters.
