IT project failure “it’s the people stupid”

I have long contemplated why IT projects and in particular, software projects fail so regularly and seem so hard to do compared to other project based disciplines. There has to be a reason why a bridge or factory (and physical IT infrastructure for that matter) can be built per project plan and be successful but software applications so often fail to live up to expectations. There are a great many lists and case studies that document software project failures and post-mortems on the likely causes and all seem valid and potentially useful. But they are still unsatisfactory in explaining why writing good software applications is so hard.

It struck me, that the answer is people, and no it’s not the project team, they exist in every project, so do the project sponsors etc. No, it is the fact that application software is dependent on the interaction of individual humans for its use and ultimate success (and arguably its existence). As humans we are a fickle and diverse lot and each of us experiences and interprets our surroundings in our own unique way, this makes designing software a unique challenge.

While we also experience buildings and bridges individually they have to conform to the physical constraints and characteristics of the materials they are made from, their intended purpose and location. It appears likely that our ‘experience’ of software is somewhat different, it is possible to divorce people from a building, you still have a successful building, software without its intended users is a failure.

The implications for software project management is that users must be the central feature of the development, while this sounds logical it is not always easy to engage users and in the case of consumer software (such as Apples IOS mobile operating system) it is even considered undesirable to engage users.

It is significant that modern development methodologies such as Agile emphasise and centralise the involvement of users throughout its iterative approach, there is room for refinement and better ways of doing software development that shift the emphasis from process, project specifications, project risk etc to one centralised on users. The bottom line is, if users fail to engage with the application it has failed.

David Gwillim
Exploring the value of IT to organisations
email: david.gwillim@optusnet.com.au
blog: http://www.businessitvalue.blogspot.com/