Back from the brink
Andrew Ball, head of IT Performance Audit at the Audit Commission, was one of the speakers at last week’s BPUG Congress. He spoke during a session in the strategic project and programme management strand about how to cope when projects go wrong.
The main thrust of his argument was trying to avoid projects going wrong in the first place by planning for failure. While that sounds a bit negative – after all, no one starts a project predicting it is going to be a disaster – it is actually all about constructive management of risk.
You can plan for failure through risk management, either at the very outset or as soon as you see the warning signs which Andrew identified as:
- Slippage
- A conspiracy of silence: no one speaking about project problems
- A big project board
- Lack of clarity of roles
- Organisational dissent, as that means people go off and develop their own solutions instead of using the product of the project
- Critical resource dependencies, where all the skill is in one or two key people
- An elusive solution that no one has really defined yet
- Lack of resources (money and people).
These are all risks that can be tackled head-on and taking action when you see the warning signs could stop a project failure.
Andrew also made a good point about the role that experts have to play in project failure. He warned that we should be wary of experts who cannot see the risks that others can. Subject matter experts, whatever their discipline, can become conditionally desensitised to risks in their domain. Just because something can be done does not mean it is easy to do. There is also the difference between a gamble and a risk to be considered: your expert may have the skills to do x, y and z but no one has the skill to throw six sixes with six dice. Luck may be on your side, but that is a gamble rather than a calculated risk. So beware experts who tell you the project will get lucky and it will all work out – other people may be better at clarifying the risks involved than those closest to it.