It turns out government projects are a lot like love affairs -- and not because of their potential to end up in scandal and/or heartbreak. Check out this show-stopping quote from Roger Baker, the Department of Veterans Affairs' new CIO, included in a Federal News Radio item about the status of IT projects at the VA:
... Although it's counter-intuitive to me, it turns out to be much harder to stop an ongoing project than to start a new one.
See what I mean?
According to the article, the VA in July temporarily halted 45 IT projects that were either behind schedule or over budget. Then last week, the agency decided to kill a dozen of those projects. Even now, said Baker, VA staffers have "a lot of work to do" to put those projects on ice. The agency is using a new vetting process called the Program Management Accountability System (PMAS). Baker said he hopes to use the system to manage all VA projects within the next year.
A presentation given by Baker in June and available on Slideshare tapped schedule and cost overruns, decrease in software quality between releases and inadequate skills for project completion as key problems discovered in the 280-plus programs the VA reviewed. His recommendations included an incremental development approach, customer tests of functionality and criteria for success jointly developed by customers, the program and any vendors.
Under PMAS guidelines, projects are halted following the third missed customer delivery milestone. Following this, several steps must occur, including re-assessment of the need for a project, the program approach and all associated service contracts.
Not surprisingly, in testifying before the Senate Budget Committee's Task Force on Government Performance last week, Baker put the emphasis on projects that were resumed with "an increased probability of success." He said:
... After all, the goal of the Program Management Accountability process isn't to stop projects, it's to reliably complete the systems and projects that our customers require.
Project overruns certainly aren't exclusive to U.S. government agencies. A recent Information Age article referenced a report produced by a UK think tank called The Taxpayers' Alliance listing a number of notable government IT project failures, including the National Programme for IT (NPfIT) which at a cost of 10.4 billion (approximately U.S. $17 billion) has exceeded original budget estimates by 450 percent.
Interestingly, the VA identifies inadequate skills for project completion as a common problem, which suggests that third-party contractors may be brought in to fill any skills gaps. The Taxpayers' Alliance report, in contrast, mentions reliance on contractors as an issue that can not only create added cost, but ultimately reduce internal knowledge of IT systems.
Both the VA and The Taxpayers' Alliance knock a "big bang" approach to government IT projects as a common problem. The big-bang approach has seemingly wrecked a number of state government IT transformation projects, including a 10-year, $2 billion deal between Northrop Grumman and Virginia and sweeping contracts with IBM In Texas and Indiana.
Of course, the public sector isn't immune to project failure. Among common project-management problems I cited in a post from last spring:
I also offered some advice, gleaned from a Baseline article. Among the tips (which make sense for both private and public organizations):