Seven in 10 government IT projects have failed, according to the chief information officer of the Department for Work and Pensions.
Joe Harley called for projects to be completed at a lower cost to the taxpayer, and said the government wanted to reduce the number of project failures to just one in 10.
Speaking at the Government UK IT Summit this week, Harley said: "Today, only 30 percent of government IT projects and programmes are successful. We want 90 percent by 2010/11. We want to achieve a 20 percent overall reduction on IT spend in government, including reducing the total cost of a government laptop by 40 percent [in the same timescale]."
Harley said that the criteria for success of a project included whether it was delivered on time, to cost, and to the quality promised. While private sector IT projects had a similar failure rate, government IT projects needed to be more efficient both in terms of cost and delivery, Harley said.
"The government spends £14bn per year on IT in the UK. It's not sustainable as a government to continue to spend at these levels. We need to up the quality while reducing the spend," Harley added.
One government project that has been heavily criticised in terms of missed deadlines and inflated costs is the troubled NHS National Programme for IT (NPfIT), which is overseen by Connecting for Health (CfH).
Andy Burn, head of information management and technology planning for CfH, said that, while the project had achieved some successes, taken as a whole, it had failed so far. "The programme still has three wheels still on. It's not in hand in some respects, but it is in others. At a local level, progress has been made over the years. At an organisational level, less [progress has been made]. The challenge is joining up services — we've been struggling with that for quite some time."
Burn added that it would take a lot of work to put NPfIT back on track. "Inevitably, with the size of the programme, we're bound to be up against the wall [for the next year]. For the next decade, not for the next year."
Talkback
Sir,
It is not before time for senior government IT officials to wake up to the failure rate of public IT projects (“Seven in 10 government IT projects fail”, 17th May).
An enormous amount of money has been spent on IT projects, such as the NPfIT, that are more likely to be deemed failures than successes. The majority of problems facing public sector IT projects stem from a complete failure to plan properly prior to the implementation process. Frustratingly, it seems that failure is all too often accepted as pair for the course, when in fact it can be easily avoided. Rigorous testing prior to and during any software project can highlight problems before they become critical. Added to this, by thoroughly defining the objectives of the project at the outset, schemes can be run successfully and with fewer time delays and budget concerns.
Software development should be viewed as a managed business process, which can be successfully regulated like any other project within the business.
I very much hope that Joe Hawley’s comments lead to alert people to this issue so that the simple processes which would rectify this situation are put in place.
Yours sincerely,
Steve Gedney
MD, Borland UK
This post has been removed by a moderator.
Dear Sir,
Many of your points are valid -- and worryingly Government seems to be moving away from rigorous planning for large-scale IT projects. The National Identity Register seems to have had no rigorous planning or testing at all, for example, with none on the horizon.
However, purely in the interests of clarity, I'd like to point out to other readers that Borland does have a commercial interest in espousing software testing, as it is an application lifecycle management company.
Also in the interests of transparency, we checked the email address linked to this ZDNet UK member, and found that it seems to be related to a PR company.
Best wishes,
Tom Espiner