The project killers

Thursday 28 February 2002 11:31

How can you prevent IT projects ending in disaster? Computer Weekly and The Coverdale Organisation have identified the causes and can now look to a cure. Karl Schneider reports.

We have all been there - involved in an IT project that is going off the rails: deadlines are missed, budgets busted and the original objectives (if they still matter) look ever less likely to be achieved.

According to research firm Forrester, about two-thirds of major IT projects fail to deliver some of their original objectives, and more than a quarter fail altogether.

What makes the difference between a project that delivers and one that does not? What are the crucial factors, among the hundreds of possible culprits, most likely to determine whether project ends in triumph or tragedy?

If we knew which factors to focus on - which ones above all others

are most likely to cause a project to flounder if they go wrong - we would be better able to improve the miserable success rate of IT projects.

In one of the biggest pieces of research of its kind ever conducted in the UK, Computer Weekly and process consultancy The Coverdale Organisation have investigated the causes of IT project failures in the UK.

We asked more than 800 senior IT professionals to tell us which factors actually contributed to the failure of real projects in which they have been involved - not their general opinions on what makes projects fail, but their experience of live (or, more appropriately, dead or dying) projects.

What emerges quite clearly from their responses is that the most important factors are not IT-specific or technical issues. The real problems are failings in the way people work together to produce a desired end result - things like communications, leadership, clarity of purpose. According to the research the most common problems contributing to project failures are with the project management process.

The other categories of project failings, in order of descending importance, are failings concerning the project leader, stakeholders, the project team and suppliers or third parties.

Within the project management process it is most important to focus on the up-front stuff - defining good objectives, setting realistic estimates, identifying resources. Getting these bits wrong is much more likely to lead to project failure than problems further along the process.

Defining, agreeing, understanding and communicating clear, measurable project objectives is a particularly significant part of the project management process. The crucial characteristics needed in a project leader are communications skills and leadership skills.

Understanding of technology and technical competence are less critical for the project leader; more important for the project team; and particularly important for suppliers and third parties. Even so, for the project team, issues such as having clear roles and responsibilities, teamworking and motivation are more important than technical competence or formal training. The involvement and support of senior management is a key issue - without it, projects are much more likely to fail.

Issues concerning suppliers and other third parties are not among the 10 most common factors contributing to project failure, although managing them tightly and ensuring that they have appropriate technical competence are the most important things to get right in that relationship.

For new project managers, the top priority must be: communication, communication, communication.

Our respondents identified the project process as clearly the most important project stumbling-blocks. Three of the top four project killers are failings in the way the project is put together - unrealistic time and budget estimates, poorly defined or unmeasurable objectives and changing objectives during the course of the project. Three-quarters of respondents said that unrealistic estimates had been a major contributor to project failures.

Coverdale's Steve Goodman, who has had years of experience with major IT projects in blue-chip firms, says that failure to identify clearly what is involved in a particular task can lead to wildly-varying time and budget estimates. "With one project team we agreed a work breakdown structure and then asked them separately to estimate the time required for each task. For one task estimates ranged from half a day to two weeks."

Goodman also emphasises the importance of clear objectives, though he adds that all stakeholders must have the same understanding of exactly what those objectives mean - what the result will look like when it is delivered. "If we agree that we are going to deliver a famous film star with dark eyes, we all need to be clear whether we mean Cary Grant or Mickey Mouse."

Failings in the project leader came second only to problems with the project process. The research suggests that good project leaders need to be good communicators, good leaders and proactive rather than reactive. Whether they have had formal project management training, their understanding of the technology involved or their experience of similar projects are less significant.

The next time you embark on a major IT project, take a look at the culprits listed above - if you spot any of them in your project, beware.

The basis for the research
Computer Weekly and The Coverdale Organisation asked senior IT professionals (heads of IT and project/team leaders) to consider more than 60 factors that could affect the success of an IT project, divided into five categories: the project process; the project leader; the project team; stakeholders; and suppliers or other third parties. For each factor, we asked respondents to say whether it had been a major contributor, a minor contributor or not a contributor at all to the failure of real projects of which they had direct experience.

The researchers interviewed 867 people. Data processing was conducted by NSM Research, an independent market research agency.

Your opportunity to learn from the results
Computer Weekly and The Coverdale Organisation, the UK's leading process consultancy, have put together a series of workshops designed to help IT project leaders to become more successful at delivering winning projects.

The workshops combine the full results of our project management research with the latest project management best practice and hands-on experience, to hone the skills of IT project leaders.

Participants will come away with new skills and a deeper understanding of the factors that make a successful IT project, plus a greater understanding of the ingredients that make a project likely to fail.

The three-day residential event will be run at the following times and locations: