Portfolio Management For Government Agencies

There are a number of levels that must be dealt with in order to align assets. The number of levels varies depending on the size of the organization. Let's take a simplified look.
At the heart of all operations is a strategic plan. Government agencies are required to have a StratPlan, and it should outline mission, values, and goals for the agency. Many times, though, the strat plan is very broad and unfocused. This isn't due to any fault of the agency -- many times Congress mucks around with the mission of agencies to the point where they may actually be supporting conflicting goals.
From the StratPlan, you should be able to discern a goals map. A goals map is major and minor goals and their relationship with each other.
Mapping into this is your Critical Success Factors, or CSFs. A Critical Success Factor is a thing that your organization needs to do well in order to be the best in the world. Hopefully you will have a smaller number of CSFs than goals, or it gets ugly. Remember to KISS.
The final business (or management) layer is a business process model, which is a more technical picture of exactly what your business does. Once again, you can spend years on this and work at a very fine level of detail, or weeks and get a SWAG. It works both ways.
At this point, you have process identification for all of your business activities and an explanation of how they support your strategic plan. Some things will probably not support your strategic plan, but if you're like many organizations, the StratPlan and main goals are so fuzzy that it's a close call. Listen up! This is the reason for this exercise. While having a mission and core values statement is critical for the success of your organization, the simple fact is these things are not written with the level of precision needed to run a large agency. Let everybody understand what's up? Sure. But give some guidance on how to pick between two competing projects who have something to do with your mission? Not at all. Black and white is easy: what we're talking about here is working in shades of gray. Working with ten people is easy: what we're talking about here is running an organization of thousands.
So that is the last pure management step -- defining exactly what the business is supposed to be doing. There is one critical cross-over step between management and IT, however, and that is process valuation.
There are many methods for process valuation -- I prefer Quality Function Deployment, or QFD. QFD is a brutal process -- it can easily take days or weeks with a large number of processes -- but it gives you a decimal rating for each of your business processes. For instance, handle customer concerns might rate a .875 and maintain databases might rate a .342. Does that mean one is more important than the other? Not at all. Both are critical to the success of the organization. What it does mean, however, is that these two processes are "more or less" assigned to organizational resources. This is an important point: at the end of the day, everybody is number one. Everybody has the most important project or program. If you don't have a way of prioritizing, measuring, and allocating resources, you're working blind.
At this point you have a Rosetta Stone -- a definition and rating of each of the things you are supposed to be doing. This provides the framework for the real activity of portfolio management: deciding on new projects and allocating resources to existing ones. From here you can view your IT resources in terms of a balance sheet, with actual versus planned expenditures (and the word "planned" means what you've agreed they should be, not the budgeted amount, track whether costs are going out of bounds from your predetermined support levels, and understand which activities must be kept in-house and which can be farmed out.
Combined with a tool like my invention, the MAT, you can start implementing a training and organizational improvement effort, targeted exactly at the parts you need to tune up -- those activities that are mission critical and have poorly performing processes. And I'm not using the terms "mission critical" and "poorly performing" in the usual BS way -- I mean something that you can measure and agree on.
It used to be that decision-making at the board level was more art than science. Sure, we used all sorts of projections, analysis, and expert opinions, but at the end of the day there was something very human and variable about the process. It also used to be that programming was completely an art -- programmers each created a masterpiece every time they worked.
Those days aren't over, but times have changed. On both of these fronts, software development and portfolio management, the work is becoming more defined and more like engineering than art. We're probably going to see some backlash against the new stuff, but organizations that embrace a more defined method of portfolio management will succeed far and beyond the others. That means that it's a done deal.
Note that I haven't even started talking about tools. Don't buy the tool instead of taking responsibility! You know who you are -- there are readers out there who have the budget authority to buy a canned portfolio management solution but don't have organizational buy-in to actually do portfolio management. Unless you want to be known as the manager who generates useless reports and nobody listens to, it's probably a bad idea. Your mileage may vary, but that's my advice. At least if you want to buy the whiz-bang solution take some time to plan how you are going to get organizational acceptance -- once again back to the MAT.
Gating, balance sheets, tracking -- there are bunch of other topics here. Perhaps we'll cover them in a later post.