Why do so many PMO's fail? It's not because they are bad evil functions or run by ineffective people. To truly answer this question we have to understand why the PMO was started in the first place. As smaller organizations grow and expand, they are typically faced with priorty confusion (not everything can be the #1 priority) and resource constraints (just not enough people to do all the work). In addition, these companies are typically fiscally responsible, so just hiring more people to do the work doesn't necessarily solve the problem.
As their companies grow and the more demand is created, the more interdependecies between projects are created. In addition, becasue of the increased number of projects, it's now much harder to assess impact when scope, schedule, and cost are being traded off. So, what's the cry for help that can be heard from near and far? "Get me one of those PMO's."
Not so fast...
The problem isn't that you don't have a PMO; the problem is that you need certain capabilities that a PMO brings. Implementing a PMO may help...but just because you have one, doesn't mean all your problems will go away. PMO's are needed to help bring an organization to their next stage of maturity. The mission statement and the goals of the PMO should match the capability and performance goals relative to where the organization wants to be by "x" date. For example, the organization may be failing projects 50% of the time. Their goal may be to fail only 25% of the time or, said another way, succeed 75% of time in a 1 year time period.
But once the organization gets there and meets this goal, then what happens to the PMO?
Well, if there is anything that I've learned from friends Stacy Goff and Lee Lambert, it's that PMO's are not meant to live forever. By definition, a good PMO, once it acheives it's goals, should be shut down. PMO's aren't meant to be operational functions, they are meant to acheive goals based on clearly defined success criteria. If the PMO isn't shutdown, it at least, needs to be re-evaluated for what that next set of capability and performance goals are. If this doesn't happen, then your PMO could begin to travel down the path of role confusion. Other functions will begin to question its value and purpose, looking at it as a process function used to make their lives more complex. "All the PMO does is create templates that aren't relevant to how we work and then yell at us when we don't use them," can be a common response heard when a PMO has reached its life expectancy.
So, what do we do about this problem?
Well, I think there's a couple of recommendations that should be considered. First, truly evaluate what organizational problems you are trying to solve. Then ask yourself, do I really need a new function established in the organization (PMO) or do I just need to get better and more efficient at certain things? Your answer may not be hiring a new head of PMO, but hiring a contractor/consultant to just implement certain pieces and processes. Pay the consultant by deliverable and by effectiveness. Each deliverable should be either it's own SOW or it should be understood that only one deliverable is worked on and paid for at a time. This way, you don't run into variable overload where too much is being implemented at once and now it's hard to tell what's really working and what's not. This approach also allows you time to determine who the champion of the PMO is going to be. Every PMO needs a champion within the organization. A strong leader willing to support and enforce the function. It may take time to evalute who this champion is going to be. I highly recommend NOT moving forward with establishing a PMO unless this person has been established.
The other recommendation is that, if you decide you do want to hire a head of PMO and build this function into the organization, then start thinking immediately, what you are going to do with this person and this function once those goals and organizational objectives are met. Don't wait until the end, but begin those conversations at the beginning. Sure you won't have all the answers you need just yet. But planning out the next goals for your PMO function will manage your expectations and everyone else involved. In addition, this approach will allow your PMO to be at the forefront of continuously improving your organizational efficiency, instead of lagging behind and becoming just an overblown group creating process for process sake.
PMO's aren't bad... they just need more thought regarding how, when, and IF they should be implemented.