The US government yesterday announced sweeping reviews of major government IT projects, especially those designated as high-risk.
According to these two memos (here and here), Federal Chief Information Officer (CIO) Vivek Kundra has been ordered by the Office of Management and Budget (OMB) to:
" ... undertake detailed reviews of the highest-risk IT projects across the Federal Government. Agencies will be required to present improvement plans to the CIO for projects that are behind schedule or over budget.Where serious problems are identified and cannot be corrected, further actions should be taken, including potential adjustments to Fiscal Year 2012 agency budgets..."
Translation: OMB intends to cancel projects that don't improve quickly.
In addition, executive departments and agencies are to refrain:
"... from awarding new task orders or contracts for financial system modernization projects - an area of persistent problems - pending review and approval by OMB...."
As noted in a blog post by outgoing OMB Director Peter Orszag:
"Financial system modernizations projects in the federal government have become too large and complex. By setting the scope of projects too broadly rather than focusing on essential business needs, federal agencies are incurring substantial cost overruns and lengthy delays in planned deployments. Compounding this problem, projects persistently fall short of planned results once deployed. For instance, the Department of Veterans Affairs (VA) has invested over $300 million in two financial system projects over the past 10 years. The first project ended in failure and no operational capability has been realized with the second."
"Across the government, there are approximately 30 financial systems projects that are affected by this policy. The total cost expended on these projects is anticipated to be $20 billion over the life of these projects, with an approximate annual spend of $3 billion."
Finally, OMB will develop recommendation within 120 days on improving IT procurement and management practices. These recommendations are aimed at addressing:
"... the root causes of problems plaguing Federal IT projects by strengthening existing policies and procedures where appropriate, eliminating outdated and cumbersome rules, and focusing on proven best practices from inside and outside the Federal Government."
The recommendations are outlined in this memo, including requiring programs to be completed within 18 to 24 months, with task orders written to reflect 90 to 120-day deliverables.
I heartily applaud the desire to rein in out-of-control government IT projects - it is long, long overdue.
I wish them luck, and hope they'll read Phil Rosenzweig's brilliant book, "The Halo Effect ... and Eight Other Business Delusions That Deceive Managers," before they codify their best IT practices. I think they'll find that their proposed best practices have side effects that may not necessarily improve IT acquisition and management when used in a governmental context.
There are two good stories on the announced government IT initiative here and here.
Robert N. Charette is a Contributing Editor to IEEE Spectrum and an acknowledged international authority on information technology and systems risk management. A self-described “risk ecologist,” he is interested in the intersections of business, political, technological, and societal risks. Charette is an award-winning author of multiple books and numerous articles on the subjects of risk management, project and program management, innovation, and entrepreneurship. A Life Senior Member of the IEEE, Charette was a recipient of the IEEE Computer Society’s Golden Core Award in 2008.