After the shaky rollout of the Obamacare website, the collapse of some state healthcare exchanges, chronic problems with data management at the Veterans Administration, allegedly lost e-mails at the IRS and the Air Force’s $1 billion expenditure in a failed attempt to implement a service-wide resource planning system, some might jump to the conclusion that government just cannot do IT projects well. However, the real story is much less bleak. In fact, there are a number of important lessons that can be learned from successful government IT projects.
1. There is a significant difference between lowest price and best value. There has been a growing trend in government at all levels to equate lowest price with best value. The Department of Defense has made something of a fetish in its application of the Lowest Price, Technically Acceptable (LPTA) standard in software and support contract awards. As long as the bidder meets the minimum qualifications, the competition is a price shoot out. Often, higher price but more qualified IT companies will no-bid a competition.
Recently, the State of Michigan awarded nearly $90 million in contracts to build the state’s new enterprise resource planning (ERP) system to the higher bidder. It did so in the face of criticism because the winning company was judged to provide best value, to have a greater potential to deliver long-term savings and to have a proven track record of successfully implementing such programs at the threshold state and local level.
2. A successful IT system is about more than just hardware and connectors. It is a mistake to think that implementing a major IT system, whether for records keeping, ERP, tax collection or intelligence collection and processing, is simply a matter of installing computers and providing networking services. Complex tasks such as these generally require a lot of high-end professional services to support system definition, program implementation and the inevitably tweaking. There needs to be a balance between low-cost activities and high-value but higher-cost services in most large IT projects.
The General Services Administration (GSA) applied this lesson in its new IT contract vehicle known as One Acquisition Solution for Integrated Services (OASIS). GSA has made an art of creating large, multi-contractor, open-ended contract vehicles that any government department or agency can access for needed resources and services. OASIS is an “indefinite delivery, indefinite quantity” (IDIQ) contract that provides a range of program management, management consulting, logistics, engineering, scientific and financial management services, and IT solutions through a single contract, saving customers time and money.
3. The government has got to know its limitations. A few years ago it was common for government organizations to turn the entire responsibility for defining, organizing, managing and implementing a major technology project over to a contractor, known as a Large System Integrator (LSI). Of course, when the pendulum swings too far in one direction it inevitably swings back the other way. After experiencing a number of major program failures as well as unexpected cost overruns, government agencies decided that they needed to be in charge of such projects, acting as the general contractor and overall program integrator, and hire private companies to do different parts of the work. The problem is that in many cases, government doesn’t have the resident management expertise or familiarity with the fast-changing world of IT to play the role of LSI.
An example of knowing one’s limitations is the way the Navy handled the contracting for its Next Generation Enterprise Network (NextGen), the successor to the Navy-Marine Corps Intranet (NMCI). NMCI (and soon NextGen) provides a full range of network and computing services for 400,000 computers and 800,000 users. The Navy recognized its limitations by awarding the current NMCI contractor, Hewlett-Packard, the two major component contracts for NextGen. This was a smart move. Both NMCI and the new NextGen are extremely complex undertakings with hundreds of thousands of parts. It is going to be hard enough to ensure oversight of the entire program, not to mention integration of hardware, software, processes, protocols and information. By giving the major program elements to HP, the Navy reduced its risk of major problems down the road. It also gave itself the breathing room, five years at a minimum, to develop the in-house management expertise to take on the role of LSI.
Find Archived Articles: