June 16, 2013
Posted by on
I was reflecting on the news during the week that a second head has rolled in the Novopay debacle. After Education Secretary Lesley Longstone, Deputy Secretary Anne Jackson has also fallen on her sword. There has been some interesting commentary regarding the extent government should be involved in IT.
Government policy changes as often as election cycles, sometimes even more frequently with change of ministers. It seems quite a common consensus around the developed world that government is gradually stepping away from providing the IT services and replacing with suppliers. I have worked most of my career in the supplier space. Still I wonder if sometimes it provides the best outcome for the governments. There are good reasons to outsource to experts. Are there equally good reasons not to?
In my experience, those of us working in IT have an inflated sense of self worth. We tend to forget that IT is almost without exception an enabler of capability. No one does IT because they need to deliver IT. It is therefore a support activity to help core deliverables of an organisation. It is argued that this activity should be outsourced to experts to deliver. Expertise in IT does not equal expertise in delivering the business outcomes for the organisation. This distinction is not always made.
Quality in any IT service delivery is relative to the the context. You can provide a quality service only when you build up knowledge of the outcomes the organisation is trying to deliver. A good understanding of the constraints are also needed for success. These are constantly evolving. In a usual scenario a number of suppliers will provide a range of services to an organisation. Typically none will have visibility of the end outcomes desired. There are so many instances of projects delivered to specifications that do not deliver business benefits, it is frightening.
The pressure from bean counters to reduce head count has resulted in false accounting. People are very resourceful and learn to manipulate systems to get what they need. By reducing IT head count, the cost is passed on to specific projects. When suppliers are engaged in those projects, government is paying for the actual cost to deliver the project + contingency applied by the supplier + supplier’s profit margin. Even if you retain the same vendor, you are not always guaranteed retention of intellectual property. In reality the cost to deliver the same outcome is no less, and is likely to be significantly more.
I do not see government contracting model encouraging sharing of risks and rewards. Contracting method encourages offloading risks totally to the other party. This results in most contracts being administered in an adversarial way. Where deliverables are provided for a fixed cost, suppliers will always try and deliver minimum that will pass acceptance, rather than trying to maximise outcomes. Where many individual contractors are used, I have seen instances where they have tried to stretch out the delivery in subtle ways.
The role of the government is to deliver a service to the taxpayers. All that is required to deliver the service in an effective and efficient manner should be part of the role of government. A government department has its own legal teams and human resource staff. Why should IT be any different. This is even more critical considering the long list of failed IT projects. Disagree? Let’s discuss.
April 14, 2011
Posted by on
There is a lot of discussion about what the best methodology is to adopt for an organisation. As soon as the discussion moves in that direction, it moves away from realities of any Software Development Life Cycle (SDLC). Projects simply do not mushroom out of thin air. Something causes a realisation in the organisation that moves them to mobilize and start a project. The project than needs monitoring and control to bring the desired transformation in the organisation. Read more of this post