High-Risk List: Government IT Projects Fail ‘Too Frequently’

The new federal chief information officer, Tony Scott, has his work cut out for him. The Government Accountability Office today is adding information technology acquisition to its high-profile list of “high-risk” federal programs.

Despite a raft of reforms over the course of the Obama administration, “federal IT investments too frequently fail or incur cost overruns and schedule slippages while contributing little to mission-related outcomes,” the 2015 update to GAO’s High-Risk List states.

The government has wasted billions on botched IT projects that fail to deliver promised – or any – functionality and have been mothballed. Even more programs are still on the books, but remain at risk of falling behind. And hundreds of watchdog recommendations for improving the state of federal IT acquisitions have gone unaddressed.

The full list of problematic, risky and troubled programs was unveiled at a Capitol Hill press conference Wednesday. Later, Comptroller General Gene Dodaro is slated to testify before the Senate Homeland Security and Governmental Affairs Committee.

The elephant in the room is HealthCare.gov, which imploded upon launch in the fall of 2013, laying bare many of the longstanding problems with federal IT acquisitions.

The site was developed “without effective planning or oversight practices and many of its issues were attributable to changing requirements that were exacerbated by oversight gaps,” according to GAO’s latest report.

Despite a mostly glitch-free relaunch last fall – open enrollment closes Feb. 15 – GAO is maintaining close watch on the site. Auditors said they will review the Centers for Medicare and Medicaid Services’ implementation of system development best practices in a report to be published early this year.

HealthCare.gov is far from the only federal IT program under scrutiny. GAO’s report also cited:

• A Census Bureau initiative that will serve as the “backbone” of the agency’s IT architecture as it ramps up for the 2020 census, much of which for the first time will be done via the Internet. GAO said the system needs probing now “in order to avoid repeating the mistakes” of the 2010 census, “in which the bureau had to abandon its plans for the use of handheld data collection devices, due in part to fundamental weaknesses in its implementation of key IT management practices.”

• The on-again, off-again efforts between the departments of Defense and Veteran Affairs to create a joint electronic health record system that would seamlessly follow military service members through their transition to VA health care. The two agencies backed away from that plan two years after it was first announced in 2011 and instead committed to developing two different systems that could “talk” to each other.

• The Pentagon’s massive new accounting system – known as DEAMS – which “has experienced significant delays and cost increases,” according to GAO. As of March 2012, deployment of the system had already been pushed back by two years and planned cost estimates of $1 billion had nearly doubled.

Poorly performing projects “have often used a ‘big bang’ approach – that is, projects are broadly scoped and aim to deliver functionality several years after initiation,” the report stated.

This has too often resulted in overdue, ineffective projects that fail to keep up with the rapid pace of technological innovation, according to GAO. It’s also out of step with official administration policy. Since 2010, the Office of Management and Budget has called on agencies to “chunk up” large-scale projects and deliver functionality in bite-sized – or at least six-month – increments.  

Yet, only about a quarter of agencies selected for review by GAO have actually followed through on the guidance.

Complicated, large-scale projects have also suffered from a lack of leadership and oversight.

Face-to-face TechStat sessions between OMB and agency officials focused on turning around failing IT investments or altogether pulling the plug on them have seen some successes, but remain drastically underused.

The government’s PortfolioStat initiative – which requires agencies to conduct annual reviews of their IT spending – could save the government nearly $6 billion if fully implemented, GAO reckons. But its full promise of cost savings remains to be seen.

Meanwhile, recommendations from GAO for shoring up IT procurement continue piling up.

Over the past five years, auditors have made a total of 747 recommendations related to IT acquisition. Yet as of January 2015, only about 23 percent of them have been “fully implemented” by agencies, GAO noted.

In the latest version of the High-Risk List, GAO doubled down, calling on agencies to implement at least 80 percent of its recommendations for improving IT acquisitions in the next four years and ensuring that at least 80 percent of the government’s major acquisitions deliver functionality within the budget year.

The watchdog also said agencies should set as a goal achieving no less than 80 percent of the more than $11 billion in potential cost savings estimated by GAO from expanding the PortfolioStat process and shuttering government data centers.

“Given the federal government’s continued experience with failed and troubled IT projects, coupled with the fact that OMB initiatives to help address such problems have not been fully implemented, the government will likely continue to produce disappointing results and will miss opportunities to improve IT management, reduce costs and improve services to the public, unless needed actions are taken,” the report concluded.

(Image via Olivier Le Moal/Shutterstock.com)