The BBC has a story concerning a report on IT projects claiming that most projects run over budget with one in six going over by a staggering 200%. Sadly to any developer this isn’t news.

Trust academics to research what we all know - the management of IT projects is usually rubbish! The report suggests this is down to “black swan blindness”, ignoring low probability but high-impact risks. I think they have a point in that IT projects do tend to have a lot of potential show-stopping risks. The reason for this is that most projects are doing something new, that’s why the project exists, and new means risk. It’s rare for IT to be like building a house which is so repeatedly mundane it can be pre-fabricated off-site. But risk can be managed, so the question is really why do so many IT projects struggle to manage it? From my own perspective on smaller, private company projects I would suggest the following reasons:-

• Most projects may not even have identified ‘managers’ and even when they do they are largely untrained, usually senior developers.
• Pressures to ‘get started’ reduce the time spent in the setting up and planning stages.
• It’s hard to really know how a project is faring, so problems are often not identified and tackled early enough.
• Hubris with too much ambition for too little experience.

None of these issues are unique to IT, but if they happen in a development environment they can be catastrophic. So what can be done? Recognising the need for experience and reigning in ambitions would help immensely. Modern methodologies tend to break development down into a series of smaller projects which enables a stronger grip of things. And being diligent (preferably unambiguous and automated) on measuring progress helps ensure problems don’t magically appear at the project’s end. As an analogy, project management for IT is rather like flying a very old, twitchy biplane. You have to keep tight hold of the controls making continuous adjustments. Tricky, but doable.