It’s easier to get a progressive bill past a Senate filibuster than it is to kill a failed government IT project. At least, that’s the lesson of the last several years. The only time failed information technology projects are killed, according to federal Chief Information Officer Vivek Kundra, is “when there are headlines” that embarrass the government.
That could change today. The Obama administration has announced the second phase of its multi-year effort to combat this problem, immediately halting $20 billion worth of federal IT projects and commencing detailed reviews of $15 billion worth of additional struggling projects. If these reviews succeed, they could eventually trim billions from the federal government’s $80 billion a year IT budget.
Monday’s announcement is the logical next step after last year’s launch of a federal website known as the “IT Dashboard,” which allows anyone with a web browser to track each federal IT project and identify those that are past-due, over-budget, or otherwise underperforming.
Government IT officials have begun using this diagnostic tool to target projects that need to be eliminated. The Department of Veterans Affairs, for example, used the dashboard to identify $200 million in delinquent projects, many of which will be cut entirely. The dashboard also rates each project’s performance on a scale from 1-10, which enables reporters and lay-users to quickly identify failing projects and generates the headlines that have motivated action in the past.
The IT Dashboard helped the administration determine that financial management projects—IT initiatives that track and manage agencies’ accounts and finances—are particularly troubled. These have been an “area of chronic problems,” according to Daniel Werfel, controller of federal financial management in the Office of Management and Budget. “We really have a poor track record,” he said.
Agencies time and time again contracted for bloated solutions laden with expensive features that require a significant amount of time to design, rather than commissioning simple, streamlined software that would satisfy mission-critical needs. The government, as a result, would pay more than necessary for financial management systems, and it would often not receive a completed product until months or years after new technology rendered that product obsolete.
The administration will combat these slow-moving behemoths by halting $20 billion worth of financial management projects in their tracks. As of today, no new contracts may be awarded with respect to any of these projects until a board of IT experts convened by the Office of Management and Budget determines that the agency has a plan to quickly and efficiently bring the project to completion. The primary goal of this effort will be to strip out superfluous features that do little more than delay IT projects and drive up their costs.
Today’s announcement demonstrates that the federal government has finally figured out how to build good IT solutions. Consider Microsoft Word, the popular word processing software that is the backbone of most American offices. Modern versions of Word are laden with features such as a spell checker, a thesaurus, and the ability to insert complicated tables and charts, but such features were foreign to earlier word processors that were little more than electronic typewriters.
It would have been foolish, if not impossible, for designers of the first word processors to attempt to create today’s Word from scratch. IT development must be a dialogue between developers—who offer the latest version of an IT solution to users—and the users themselves—who, in using the product, discover unanticipated needs that can be addressed in a later version.
The IT Dashboard has also enabled Kundra and his team to identify 30 additional troubled projects, which together are expected to cost the government approximately $15 billion. As part of today’s just-announced reforms, these projects will be run through an intensive review process known as TechStat, where Kundra will sit down with each project’s management team to determine whether the project can be brought back on track—or whether it must be killed entirely. The administration hopes to use such reviews to avoid another disaster like a Department of Defense human resources project that cost a billion dollars over 12 years, only to fail to deliver a working product.
Of course, the administration’s most difficult challenges await them in the future. It is easier to identify broken IT projects than it is to fix them, and the administration will have to conduct its reviews amid both intense lobbying from IT vendors and equally intense anger from agencies that don’t want OMB to second-guess their decisions. Such pressure will only raise the stakes of the administration’s reforms, since any failures will be used to defend the status quo.
But today’s announcement brings hope that the government will end its failed experiment in trying to walk before it crawls. If nothing changes, federal IT projects will continue to bleed billions of taxpayer dollars every year. The Obama administration deserves praise for taking steps to staunch the bleeding.
Ian Millhiser is a Policy Analyst with the Doing What Works project at the Center for American Progresss.
For more information, see: