Unless you are a hopeless geek like me, this question is of no interest unless it helps you structure your own PMO, right here, right now. Well, attending the ppsosig event in September could take you a long way towards your answer. There is a session on Day 1 that, taking the comprehensive P3O framework as a starting point, runs through the more stable configurations of ‘P’ Offices in various types and sizes of organisations. At the very least you will come away with an understanding of how your peers are facing the same challenge, what works, and what is a long shot, but worth a try.
Why is structure important? On a pragmatic level it is the most visible aspect as far as the rest of the organisation is concerned: how many staff, what roles, what levels or roles, what authority, how do they relate to other areas, etc. Any business case to establish or re-energise a PMO must be absolutely clear on the proposed structure. After all, others in your organisation may not be too clear on what you do or its value, but they can certainly read an organisation chart.
On a deeper level, structure is important because it dictates how you choose to deliver the services you offer, and how you propose to evolve your PMO to continue improving and being relevant.
Let’s look at the Pyramids. Relax; this is not the usual boring aside on how project management has been around for a while. It is way more boring than that. Egypt is not the only place in antiquity to produce pyramids. There were older ones in Mesopotamia, there were stranger ones in South America, and even an inverted one near Xian, where the emperor who built the Great Wall of China was buried with all his terracotta warriors. The diagram below shows a ziggurat, a pleasant ancient type, one of which was the Hanging Gardens of Babylon. Another was the Tower of Babel, but we won’t dwell on that.
Fig. 1: Functional areas within a P3O model
Form should follow function, and when it comes to PMO that means that the best structures are those that facilitate the delivery of the services you choose to provide. It’s no use denying that the ground on which your PMO structure is built is made up of those old admin and support facilities that were initially provided by PSOs to single big projects. You may no longer do that, but that is the historical foundation. It is important not to forget it, even if you outsource that work, because mundane as it is, it provides real value to your internal customers. Somebody has to do it. Your mission is to make it clear who does it and ensure it gets done. You don’t have to do it yourself, and there are real advantages to leaving all that ‘admin’ label behind.
The first level of structure that needs to be built for our ziggurat is made up of the delivery or support functions and services. P3O recognises that helping programmes and projects succeed directly is the first step. In some organisations, a pretty flat pyramid that ends here is the most appropriate solution. The next stage, building on the previous one is the COE – the Centre of Excellence functions and services. At this point your PMO is setting the standards, possibly training programme and project managers, and certainly setting the pace for what is best practice in all aspects of change management. The final level, which requires the previous two, as well as a certain level of organisational maturity is the strategic planning or portfolio support functions and services.
The ziggurat givesyou a clue as to what you need to have in order to tackle the next stage, but it doesn’t give you an organisational chart. At this point you need to consider whether your PMO is centralised or decentralised, whether it is permanent or temporary, whether it is physical or virtual. So how do you decide that?
The key is to ensure that your PMO fits your organisation, not some external theoretical model. Hands in gloves, and all that. So you need to consider what are the types of structure that your own organisation actually favours. For instance, if ‘decentralised’ seems to feel right to you it may be because your organisation has many geographical locations and other departments are already distributed geographically. Or it could mean that every executive or board member likes to control all aspects of delivery in their patch. Your PMO has to match the business goals and the ethos of your organisation. You may have found the perfect PMO described in some journal, but tails never wag dogs: you wouldn’t get away with it and you shouldn’t.
This is why events like the ppsosig Autumn Conference are so valuable. You can ask other PMO practitioners how they are dealing with the question of structure, what has worked for them, whether your respective organisations are similar in some way, what they hope to try next, etc. You could even keep in touch. And every one there is focused on PMO, no other roles – a small and select band of fellow travellers.
Consider the next question, once you have a broad org chart – how big should it be? If you look at the P3O book (pp. 58-59) you are told that a PMO could be as big as one person, on average between 5 and 10, and easily up to 100+ in some cases. So how do you know? Well, for programme offices you could use a rule of thumb related to the monetary value of the programme (3% to 5%), or another rule related to the team size, where a PMO would be 10% of a 30 strong team or 3% ofa 100 strong team. Asking others PMO practitioners what they have done and why is even better.
A more accurate method (same source as above) for larger multi-programme offices asks you to list your functions or services and then estimate about how many hours per week it takes to deliver each, broken down by the different skills and levels involved. Keep in mind that to use this method you need to be sure about your services, i.e., why you offer what you offer and how, which ones are essential, etc.
Also keep in mind that you must take into account the impact of the maturity of the organisation you serve, before settling on a menu of functions.
Given that you may not get a bigger budget (safe bet?), you should also consider upgrading the skills you have and re-aligning existing resources and structures to cope with actual requirements.
In fact, you should consider attending the ppsosig Autumn Event and we can continue chatting with those who’ve ‘got the T-shirt’ over a nice drink.
Lain Burgos-Lovece is a member of the managing committee of the ppsosig, the UK’s programme and project office specialist interest group. Lain is also Director of Serissa which helps clients to get the best out of the control environment for their projects & programmes, often through corporate centres of excellence known variously as PMO, PSO, PPSO, etc
One of the most eloquent summaries I have encountered recently on matters PMO and displays clarity of thought and excellent grasp of the subject. Looking forward to gaining similar insights at the conference
Twitter: itgEvangelist
A very thoughtful article that shows great insight into the numerous dimensions and challenges of establishing a PMO. I am certain many folks would benefit greatly from this post.
That being said, I think it underscores a major flaw in eneterprises today – our function-centric nature. This function-centric nature demands we rely on org charts to deliver our services and get our work done.
I would much rather enterprises were process-centric. I have written a number of posts recently in regard to this subject. Governance is the processes and relationships that lead to reasoned decision-makiing. Given our lack of end-to-end processes, we must rely on relationships (org charts) to get the work done.
Reference is made to process in this post, but as usual, setting up the PMO begins with structure. This is the most pragrmatic approach, given our function-centric nature. If asked to set up a PMO, I would simply begin with business objectives and the processes required to meet them. Then and only then, would I look at the opportunities and constraints of my current organizational constructs.
So I will continue to advocate and foster turning our enterprises on their ears, in the hopes of becoming process-centric – which would make org charts near meaningless.
Steve Romero, IT Governance Evangelist
http://community.ca.com/blogs/theitgovernanceevangelist/
Thank you Yohan and I hope to see you there. Steve, I hope to see you there, too, and let me say first of all that I support your drive for process-centric enterprises. I agree that many enterprises today are still too function-centric, but luckily, practices like business process re-engineering and total quality management have been mainstream for nearly two decades now. I think process-centric won the argument a long time ago.
To clarify, then: my brief post was only about the specific question of how to structure a PMO team and how the P3O framework can help. There is much more to establishing or re-energising a PMO, and I’d love to trade experiences about that some time.
My post applies whether the organisation is function or process centric. For example, if the enterprise is structured as end-to-end processes with the external suppliers at one end and the external customers at the other end, the PMO (an internal process) can be described in terms of its inputs, outputs, resources and constraints – in other words as a process. However, the skills (a type of resource) that make up one part of that process definition come packaged as people. People need to play defined roles for the process to succeed. They can’t ‘just be themselves’ and hang around, and they are not robots to be told ‘just do it’. The P3O framework defines a number of roles that have been found to be best practice, such as P3O Sponsor, Head of P3O, Portfolio Analyst, etc.. To be able to work as a process team the people playing those roles need to organise themselves. Their org chart simply illustrates their process-centric setup.