Big government IT projects are doomed

Breaking government IT projects down into small chunks is the best way to guarantee success, according to the chief executive of Fujitsu Services.

Richard Christou explained that "big bang" projects, using bespoke software and driven by political aspirations, are much more risky.

"If you drive projects purely from political aspirations you set unreal targets," he warned.

Speaking at an event organised by vendor body Intellect, Christou insisted that the government should be wary of large-scale projects.

"You don't hear so much about disasters in financial services because their systems grow up over a number of years," he explained. "One of the things we should avoid is too many big bang projects."

Earlier this month Fujitsu Services - formerly ICL - was criticised by MPs following a National Audit Office report into the problems surrounding the Libra computer system for magistrates' courts, which is likely to cost double what was expected.

Fujitsu Services was also involved in the failed Pathway benefits card project.

But Christou maintained that the approach being taken to procuring new systems for the NHS is the right one, with much more consultation and preparation.

"What would have been a big mistake was if this was bundled into one package," he said.

"It is too large for any one of us to take on sensibly, and that would have been a disaster and would not have delivered.

"A bit of humility, and taking it in bite-sized chunks, is more likely to deliver."

Christou suggested that the government funding of IT projects has matured. "I think we have learnt a lot of lessons over the last few years, some of which were painful," he admitted.

"Private finance initiatives [PFIs] are not something you can shoe-horn all of these contracts into. I don't believe that big bang bespoke software projects are appropriate to PFIs.

"The current policy on public/private partnerships says that there are a variety of ways to engage, and many different models and not one size fits all. We've got a much better idea of how risk should be allocated."