Just finished reading “Best Business: The Agile PMO – Leading the Effective, Value Driven, Project Management Office, a practical guide” by Michael Nir. Bit lightweight really. The most amusing bit is the way he lists how PMO’s often destroy value:
- Focused on tools
- Focused on processes
- Focused on standardisation
- Focused on managements needs (for reports, policing, the appearance of being in control, …)
… and so on. His key point is that a PMO needs to be focused on maximising the value generated by the organisation it is serving. Yeap. I buy that. Everybody in the IT department needs this focus. The issue is that… well… the book is a bit short on answers about how to do this.
Lets start from basics. Do we need a PMO? IT management should also be primarily concerned with maximising value delivered. Four questions that projects by themselves struggle to answer may help with this:
- Do we consistently work only on the most valuable projects (and not too many of them)?
- Do we consistently ensure bottleneck resources are allocated to projects in the best interest of the company?
- Do we consistently identify bottleneck resources and what can be done to increase their capacity?
- Are we effective in transferring learning from one project to another?
The questions are not whether IT management do all these things but more whether they know that these things are happening properly (“work on the system, not in the system”).
I’m guessing most IT management teams would answer “Don’t know” to these four questions.
I suggest that responsibility for these questions can not be delegated to a permanent side organisation (a PMO) since these issues are all too difficult to solve without management’s direct involvement – which is why projects struggle by themselves.
I can see a case for having a temporary change organisation charged with helping the organisation adopt practices which help with the above. Some examples might be: Kanban (helps identify and manage bottlenecks), T-shaped individuals (increases capacity at bottleneck), Cadenced resource scheduling (helps with allocation of scarce resource), Cost-of-delay (identifies the most urgent requirements – good for all types of prioritization discussions), Retrospectives (good for capturing learning) etc.
None of these practices are magic bullets and so, in my vision, it goes on in endless waves: management review the 4 questions, decide which one(s) most impede value delivery, identify some practices to embed in the organisation which might help, set up a temporary change programme to drive this embedding and, after a while, the programme is over and it’s time to review the 4 questions again.