IT directors and project managers have rejected a recommendation from analyst firm Gartner that they should scrap up to a quarter of their IT projects to improve success ratios.
Last month Gartner said companies need to be more ruthless with IT projects, claiming that two-thirds of them fail to achieve their intended result. Any IT project without a realistic payback should be stopped within 60 days, says John Mahoney, managing vice-president at Gartner. This would mean scrapping about one in four IT projects.
Seasoned IT professionals are not impressed with the advice, however. They argue that business directors need to take greater responsibility for failing IT projects, rather than passing the buck to IT managers.
"This completely misses the target," says IT industry expert Colin Beveridge. He says only three types of project can be described as pure IT projects: data re-engineering, infrastructure renewal and infrastructure rationalisation. "Everything else is a business investment. IT managers are getting the blame for poor business investment, when the real focus should be unrealistic business expectations of technology."
Miles Shepherd, chairman of the Association for Project Management, agrees that scrapping failing IT projects would usually be a business decision, rather than a matter simply for the IT department. He also questions whether IT decision-makers have the power to scrap an IT project, even if they wanted to.
"Many IT managers would not have the clout to pull a project," he says. "That would be a business decision, although IT managers might well take a view on it."
The payback on IT projects can be difficult to quantify, says Shepherd, making it hard to meet Gartner's specifications. However, Shepherd agrees that where a project is failing, it is better to scrap it early than to let it run on.
Philip King, support service manager at Three Rivers District Council in Hertfordshire, says any sensible project manager should scrap a project if its cost and timescale threaten to balloon out of control. "But a bit of forward planning should avoid that situation," he says. "At Three Rivers, we have a three-year IT strategy that takes into account our capacity to implement systems. Something always crawls out of the woodwork, but we include some contingency planning to cover that."
Chip Steinmetz, chief technical officer at Lastminute. com, believes a high level of abandoned projects would suggest that IT managers were being reckless with company investments. "The idea of dipping a toe in the water is fine, but not on major IT projects," he says.
However, some experts agree with Gartner that swift action is needed if things start to go wrong. David Rippon, professor of IT infrastructure management at Buckinghamshire Chilterns University College, says IT departments facing failing projects need to take action to restore their credibility. "The earlier they identify issues with projects, the less it costs to resolve the situation," he says.
David Roberts, chief executive of the Corporate IT Forum, says Gartner's recommendations reflect an industry-wide focus on cost-cutting. "That is top of the agenda at the moment," he says.
Roberts says greater attention to the planning of IT projects would prevent so many having to be scrapped after implementation, but he points out that this can be difficult in the current climate, where IT managers are under increasing pressure. "Spending more time on planning is the obvious thing to do, but there is significant pressure to make projects take less time, be more reliable and cost less," he says.
The overall result of these pressures will be to speed up the move towards greater standardisation of software products and increased interoperability to reduce the amount of custom building needed in IT projects, Roberts adds.
Philip Virgo, strategic advisor to the Institute for the Management of Information Systems, says the emphasis on catching failing projects within 60 days is a sign of the way most IT projects now start as pilot schemes that are then scaled up.
"If people have not broken their projects up into manageable chunks, they have not learned the lessons of the past 20 years," he says. "In that context, Gartner is absolutely right and is simply reflecting current practice."
Virgo says, in the present economic climate, companies are focusing on clear economic returns from their IT investment.
"The rules of cost justification and project management are back," he says. "Most company boards know the easiest ways for them to cut costs is to take an axe to IT projects, because IT is now one of the biggest-spending areas."
He adds that many IT projects are failing because of a lack of project management skills in the IT workforce.
"A lot of people have no training in the original disciplines of organisational methods and operational research, which tell you if a job is feasible," he says.
Ray Scotter, IT director at credit card company Capital One, says scrapping projects after 60 days is far too late in the development process.
"We need more rigour in planning projects in the first case," he says. "We have had to make a difficult decision to scrap projects, but not in many cases. Scrapping a project would be by far the exception, rather than the rule."
Tim Jennings, research director at analystfirm Butler Group, says the concept of portfolio management is becoming increasingly important.
"There is a new breed of [software] products that enables organisations to work from the top down," he says. "You can start with very high-level business objectives and break those down into the projects that need to be supported and the resource availability to do that.
"It means you have the big picture, because it is important not to plan individual projects in isolation."
Simon Mingay, vice-president at Gartner, responds to criticism of its project management advice by saying that while planning is always important, it can never be a substitute for project roll-out.
"As you start to deliver a project, you learn far more about the environment and the complexity of your architecture than during the requirement and justification phase," he says. "More importantly, you start to learn the reality of the promised benefits."
Mingay believes it is vital to assess IT projects on a continual basis. "The reality is that justification is more of an art than a science and some benefits do prove unachievable," he says.
"It is in the first 60 days that this is most likely to come out, but realistic assessment has to go on throughout the whole project. It is the role of IT managers to raise any issues of which they are aware and to take a project back into the governance process if appropriate."
Gartner has certainly highlighted an important issue in when to call time on a troubled IT project. Its point about axing ailing IT projects sooner rather than later is a fair one.
But amid the continued economic downturn, IT managers are under considerable pressure to make projects work. Pulling the plug may not go down well with the board after so much time and effort has been invested.
IT managers must wish project management was as clear-cut as Gartner implies.