I will admit that I am not the smartest person on the face of the earth. Almost everything I have learned about IT governance models and project and portfolio management (PPM) I learned by doing it wrong.
For example, in my first CIO role, I recognized the need to have some type of IT decision-making process that extended beyond IT. I convinced the CEO to convene an IT steering committee composed of a subset of the executive team. The CEO acceded to my request and I spent six months slogging through the agony of trying to get eight people who had never collaborated on a single decision to come together and agree on IT priorities. My evidence that I am not the smartest person in the world? That fact that it took me six months to recognize that this IT governance model was not working and never would.
As a result of this and other abysmal failures, I feel like I can now provide you some lessons learned on effective IT governance and PPM.
Let's start with governance. My failures taught me that I should design my IT governance model around two dimensions -- the organisation's governance model and the role of IT in the organisation.
As a starting point, I should match how we make IT decisions to how the organisation makes decisions. If the company makes decisions in a centralised way, I should seriously consider using a centralised IT governance model. If the company makes decisions in a decentralised way, I should consider a decentralised or federated IT governance model.
I then adjust this depending on the role of IT in the organisation. If IT is a strategic enabler (even if just for specific initiatives), I adjust my IT governance model toward centralisation. If IT is a tactical enabler, I adjust my governance towards decentralisation or federation.
Let's return to my first big governance failure. The company operated like a holding company and each department made independent decisions. By trying to impose a centralised IT decision-making process, I was going counter to the culture. In addition, IT was a tactical enabler except for some major, strategic initiatives. After I realised my mistake, I revamped my IT governance process to match the decentralised nature of the organisation. For each of our strategic initiatives, I formed a project steering committee composed of the disparate functions that had a role in the initiative.
I remembered these two dimensions of IT governance when I landed at another company. The organisational decision process was highly centralised and so I implemented, very successfully, a centralised IT governance process.
Now let's take on project and portfolio management. Over the years, I have tried -- and failed at -- the most complex and elegant PPM processes on the face of the earth. I am now ashamed to admit it, but I once evaluated and balanced my project portfolio using a combination of six dimensions with weights that varied by the season, the winning percentage for the Boston Red Sox and the NYMEX crude oil bids. OK, so I did not use this exact weighting system, but it was a system so complex that it took much longer to explain than to just make a decision.
After failing at such PPM models, I took a different approach that has served me well. Using a model I have described before, I construct and manage my portfolio on the dimensions of purpose (is this a differentiating or a parity project?), business value and risk (with risk being affected by uncertainty and complexity).
The most important lesson I have learned in managing project portfolios is that rarely can one rank projects based on a calculation. It is simply too hard to turn considerations like "improved flexibility" or "time to market" into a number that I can drop into a calculation. After all, what is the numerical value of a project that is a strategic enabler? Since I am not the smartest person in the world, I have not been able to figure this one out.
Perhaps the most important lesson I have learned from my numerous failings is that we IT types tend to overcomplicate just about everything, and this is not a good thing. A pragmatic approach to governance and PPM turns us into real business partners, not IT bookkeepers and ideologues.