3 January 2006

Are IT projects or IT systems more important?

By Andrew Clifford

In IT we are obsessed with projects. Focusing your effort on the systems themselves, not just on projects, can deliver significant tangible benefits.

I gave up my regular job just over a year ago. Since then I have been researching and developing a new approach to IT management that really gets to grips with many of the areas of IT that remain so difficult. I have become so convinced of this new approach that I have jointly set up a company, Metrici, to take it further.

Over the coming weeks, I want to share this new approach with you. I want to explain the rationale behind the approach, how it adds value, and how and where to use it.

To start, I want to ask a question. Which is more important: IT projects, or IT systems?

IT is constantly changing, and projects are our vehicle for that change. It is natural that we want to manage and control projects well. The problem is that we focus so much on projects that we forget that they are only the means to an end.

Consider the basics of business IT.

The product of IT is automation of business activities. But we have become so obsessed with projects that we treat projects as if they were the product of IT.

Some would disagree and say that we do not run pure IT projects any more, we run business change projects. Sometimes this is only a pretence. I have seen major IT-led projects pretending to be business change projects, but completely ignorant of the underlying business that they claim to be changing.

Many IT projects truly are part of broader business change. But the purpose of the IT portion of the project is to support a different pattern of business activity, which requires different IT systems. The project value is still delivered through IT systems.

Treating the project as the product of IT leads to significant problems. Systems have to go live because of project pressures, even though they are not complete, untested, full of bugs, unusable, and fail to comply to even basic standards. This may meet the project's objectives, but in business terms it is pointless. The only purpose of an IT project is to deliver a working, valuable system.

We have even worse long-term problems. Through time, systems become so complicated and error prone that we hardly dare touch them. We have difficulties keeping up with current standards, and many systems run on unsupported technology. Most systems have poor documentation. We have risk areas like security and disaster recovery. These are all problems with the systems themselves, not with how we manage projects.

I am not attacking project management. But we can gain massive benefits if we can overcome these problems, and to do this we need to emphasise the management of systems much more. This shift of emphasis is not easy though. Next week I will describe some of the barriers we face and how to overcome them.