Most large enterprise IT projects miss deadlines, go over budget and fail to make users happy, according to Standish Group research.
That was certainly true of Healthcare.gov, which stumbled badly after its Oct. 1 launch.
President Barack Obama says there's "no excuse" for the problems that have afflicted the website that people are supposed to use to shop for health insurance under the Affordable Care Act (ACA). Jeffrey Zients, a former management consultant appointed by the administration to fix Healthcare.gov, is promising to have all the remedial work done by the end of this month. But "it'll take a lot of work," he said. "There are a lot of problems that need to be addressed."
The troubled rollout shouldn't come as a surprise -- the early success rate for large, complex IT systems is very low.
And this particular initiative was extremely complex, experts say.
Healthcare.gov isn't just a website -- it has multiple interactions and interdependencies with data stored at other agencies and private-sector companies, said Lev Lesokhin, senior vice president at Cast, a software analysis and measurement firm. It's likely the site has 500,000 lines of code, he added.
At the request of Computerworld, Standish Group analyzed the outcomes of multimillion-dollar initiatives in its database of some 50,000 government and commercial projects.
Of 3,555 projects from 2003 to 2012 that had labor costs of at least $10 million, only 6.4% were successful. The Standish data showed that 52% of the large projects were "challenged," meaning they were over budget, behind schedule or didn't meet user expectations. The remaining 41.4% were failures, meaning they had to be abandoned or were started anew from scratch.
The Healthcare.gov developers "didn't have a chance in hell," said Jim Johnson, founder and chairman of Standish. "There was no way they were going to get this right -- they only had a 6% chance."
Johnson said the ACA project is fixable, and the rollout problems aren't "life-threatening at this point."
The Healthcare.gov mess follows well-publicized demises of other major government IT projects. Late last year, for instance, the U.S. Air Force said it was halting an ERP project that had already cost $1 billion. Similar scenarios played out when the FBI abandoned a $170 million virtual case initiative, a U.S. Census Bureau automation project suffered big cost overruns, and an Orange County, Calif., tax system modernization project was found to be "fatally flawed."
Johnson said Healthcare.gov may have been doomed by an attempt to pull off the "most dangerous" of software project feats: a "big bang" release, in which an entire site goes live at once.
Other factors could have included an excess of changes during development, too much bureaucracy, or slow-moving government and contractor designers and developers, he added.
Contractors involved in the project blamed inadequate testing of the site's many integrated components. Indeed, the 55 contractors had but two weeks to conduct end-to-end testing of Healthcare.gov prior to launch.
The administration initially awarded contracts worth about $93 million for the project, but costs have since soared well beyond that.
Grant Gross of the IDG News Service contributed to this story.
This version of this story was originally published in Computerworld's print edition. It was adapted from an article that appeared earlier on Computerworld.com.