Budget restrictions, politics, workforce training and competing operational priorities are all factors in government technology decisions. However, at the federal government level, agencies are legally required to investigate commercially available software before choosing to pursue custom software development.
Today’s blog discusses whether Arizona could benefit from a similar policy.
The Federal Acquisition Streamlining Act requires federal agencies, to the maximum extent practicable, research and purchase commercial technology products if they fulfill all or part of a government need. The elimination of development time, quicker procurement processes, and reduced design specifications and testing are all reasons behind this Act. Only if the private sector cannot satisfy the government's need can the government buyer require that items be built to government specifications.
A few months ago, a federal court took up a bid-protest case dealing with an agency’s obligation to procure commercially available items. In 2015, an IT company challenged an Army procurement of a $206 million intelligence analysis computer system that could share information across the Joint Forces. With an outdated architecture and technology that was nearing end of support, the Army wanted to deploy a modernized architecture using a modular system approach.
After hiring a third-party consultant and conducting multiple RFIs that suggested a hybrid COTS approach, the Army concluded that a commercial product didn’t exist and it would hire a single contractor to be the system data architect, developer and integrator of the new system.
The protesting company argued that the Army acted improperly and violated federal law by procuring a large-scale custom engineering effort rather than considering whether an existing software solution was already available. Ultimately, the company prevailed with a unanimous ruling by a three-judge panel concluding that the Army had indeed failed to conduct a mandated analysis of the availability of a commercial solution to meet its needs.
The Army’s technology challenges are not unique. Arizona agencies are in the same position with aging computer systems. However, in Arizona there isn’t a formal procurement policy to encourage agencies to explore commercial products over custom-built systems.
If you are familiar with some of the state’s IT efforts, then you know that agencies have a mixed adoption rate in terms of preferring COTS solutions. I can think of at least three system projects that have been in development for over three years with costs above $50 million. Will these efforts result in modernized systems? I think so. Did the agencies give any serious weight to evaluating platforms already performing in the market for all or even a portion of the project? I don’t know.
And while Arizona state law requires agencies to conduct a third-party analysis of technology purchases valued at over $5M, there isn’t a requirement that the agency evaluate commercial products over custom-built products. Certainly, there is not a legal standard that a company could use to protest an agency’s failure to fully explore a commercial solution.
With a long list of outdated computer systems ripe for replacement, is it time for Arizona to consider a similar policy as the federal government? At a minimum, it would be beneficial to have a policy that requires agencies to fully investigate whether commercial items could be modified or integrated to meet an agency need. Indeed, it is very expensive to have a government agency build functionality when it can purchase the core functionality from the private sector and integrate other applications. An added bonus would be to make the agency’s analysis easily accessible to the public to review and require a justification if an agency still insists that only a custom build solution can be purchased.
Given that the qualifications, references and pricing are favorable, the procurement system should fully leverage existing and proven commercial solutions. To exclude them from consideration in favor of a custom software engineering effort without a firm-fixed price or outcomes-based scope doesn’t benefit anyone, especially the taxpayer.