Small steps best on road to success

Small steps best on road to success

A more incremental approach to rolling out technology can scare IT departments and corporate boards, but the upside is there's a chance to change business processes and learn from experience along the way.

IT strategist Peter Reynolds says the Australian Department of Immigration's new approach to its controversial software project is the right one. Last week, Australian Finance Minister Lindsay Tanner took a swipe at the Department of Immigration and Citizenship's controversial A$495 million Systems for People software project.

The initiative, running $44 million over budget, was a prime example of the type of technology program the government should more tightly control as it overhauled the management of its $6 billion a year computing and communications budget, he said.

But former Commonwealth Bank of Australia chief technology officer and IT strategist Peter Reynolds says while there is much Canberra can learn from Systems for People, it's wrong to hold it out as an example of how projects should not be run.

Instead, the government should embrace Immigration's approach, which relies on less upfront planning and promotes a stream of incremental stages that are adjusted on the go.

"What they're doing in Immigration absolutely needs to be done," Reynolds says.

"With the issues they've had in the past what they're doing is the right solution to the problem, and the staged-incremental approach that's needed when you're working in an environment that's quite uncertain is the correct response.

"The challenge that the government has is to ask, 'how can we encourage that kind of approach and how do we share the learning from that?' "

Reynolds, who is a research fellow in The University of Melbourne's information systems department, believes he's in a strong position to defend the Systems for People project.

SfP, as it's known, is modelled on Commonwealth Bank's $250 million CommSee customer relationship management system. The software project, in which Reynolds played a leading risk management role, endured similar controversies.

Reynolds also sits on SfP's steering committee after Immigration chief information officer Bob Correll decided he wanted to use the CommSee approach.

"If you compare CommSee and Systems for People, they're at a very similar stage of development and, of course, CommSee went on to be quite a strong success."

But CommSee wasn't always viewed so positively, and in 2004 shareholders were unsettled by hints that CBA management wasn't satisfied with the way the project was progressing.

But such surprises can be part and parcel of adopting a more flexible approach to building large computer systems and ultimately improves the chance of success, Reynolds says.

"What we see in CommSee is a different way of running IT projects.

"Traditional projects tend to involve a lot of upfront planning with a three-year forecast of requirements, a three-year forecast of a schedule and a three to five-year forecast of benefits.

"The projects gather a lot of requirements upfront and five years later they deliver. But by that time the requirements have changed . . . so those projects have a one in three chance of success."

Reynolds' claim that two out of three large IT projects fail is backed by research from The Standish Group.

But he says many people have trouble abandoning traditional approaches to installing big, new software systems despite the statistics.

A more incremental approach to rolling out technology can scare IT departments and corporate boards because as little as 60 per cent of a project's requirements are locked down in advance, with the remaining 40 per cent worked out on the fly.

But the upside to that uncertainty is there's a chance to change business processes and learn from experience along the way.

"The advantage of a project like CommSee is that you can do the delivery of a stage and then respond to that in the next stage.

"If you'd only done planning upfront you're locked into a set of decisions. This is what I see in common between CommSee and Systems for People."

Reynolds acknowledges Immigration got its funding model wrong. Traditionally, large IT projects tend to incur most of their costs about two-thirds of the way through. In the case of SfP, the department banked on making its biggest investment upfront with funding then dropping off over the five years of the project's life.

The best model is consistent spending in each year of an initiative, which ensures consistent development and progress, he says.

Immigration's Correll has acknowledged the department misjudged its financing model, but he also said increasing contract labour costs in Canberra contributed to SfP's budgetary woes.

The Australian Taxation Office and Centrelink have reported similar contractor price pressures.

Reynolds points to the success of CommSee and says the missteps shouldn't mean the government turns its back on SfP.

"Commonwealth Bank went through a lot of issues when it tried to push through this new approach. People weren't used to it, and in particular other senior business executives were used to that feeling of control, or having these upfront schedules. That's where the government has a large challenge.

"They need to be working with the Department of Immigration to help them implement this new approach, which is the right approach. It's what we've seen work in CommSee."

Matching IT skills with business savvy

Former Commonwealth Bank of Australia chief technology officer Peter Reynolds says he always liked the idea of a career in academia, but to get there he took a detour through the navy, Boral and CBA.

It seems a meandering course, but the University of Melbourne research fellow says there is one thread linking all the jobs.

"My focus has always been on IT strategy, how you get things done in big organisations, how you can use IT to change the business.

"The common theme in my career really has been large-scale, IT-based business change."

Now he's witnessing academia first hand, Reynolds says universities need to adjust the focus of their technology courses and the way they're marketed to put more emphasis on the skills needed to successfully manage change.

"The debate needs to be broadened. It needs to be broadened from what are the J2EE or .Net skills that are required, to look at the full range of skills you need to drive value from IT for Australian companies.

"When that debate gets broader I think people will see that this is an exciting place to be and that it's quite business-oriented."

However, he cautions that in debating the emphasis that university courses put on technical skills as against business acumen, it must be remembered that corporate IT bosses want both.

"Talking to chief information officers around the country, they're demanding that their people understand the technology at least to a level where they understand what the systems are, how they work, the processes they influence and what it costs to change them.

"They're not being asked just to come in as a pure business person. They're IT professionals who are business savvy, and that's a focus we haven't seen for some time."

Fairfax Business Media

Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

Tags costproject failurestrategies

Show Comments