19 PMO ORGANIZATION MODELS EVERY PORTFOLIO IS UNIQUE Before making any decision on the type of PMO you will establish, we recommend that you do some homework on what your project portfolio currently looks like. You can collect information on all active projects, or just on those that the PMO might be asked to help. This research will help you determine the risks that three out of the four models of PMOs represent. Consider Figure 19.1. This figure shows one example of how to catego- rize current projects. In this example, we have plotted three different project types (development, package integration, and maintenance) into the model as data points. We used this IT terminology because in most organizations, IT owns the majority of projects. We also separated programs or projects by small, medium, and large. Can you already draw some conclusions from this data about what chal- lenges this PMO faces? This "scatter-gram" statistical approach shows some of the bias in this organization. For example, only 5 out of 27 projects shown are not IT. Usually, this means that there is a lot of animosity toward IT, which will be blamed for projects being late and for resource conflicts. It is also a sign that the part of the organization gathering this data is missing some key projects taking place outside of the IT area. There are more projects 283
284 Advanced Project Portfolio Management and the PMO Enterprise D P/I M/E 1 Functional Units D P/I M/E IT D P/I Large Medium Small 2,3 4 5 6,7,8 9,10 18,19 20,21 22,23 11,12 13 24 25,26 M/E 14,15 16,17 27 D = Development Projects; P/I = Package Integration Projects; M/E = Maintenance or Enhancement Projects Figure 19.1 Scatter Diagram Showing Breakdown of Enterprise Projects in this organization. It is a red flag that there might be "leakage" - IT is losing ground to outsourcing. Another strange characteristic of this data is the total absence of small projects. Of course, they must exist within the organization, but the organi- zation is not accounting for them. This suggests an organization embattled in resource conflicts. The big, long, complex projects keep losing resources to the small, quick, "show results now" projects. Medium and large projects are constantly delayed, but no one in the functional units is willing to do things differently. Why should they? It would only benefit IT, and the func- tional units do not view IT as their partner. This is where the PMO can have a great impact. One of the first jobs of the PMO will be to transfer ownership of some of these projects to func- tional or executive areas, assuming that the projects are worthwhile. This scatter-gram should be created with several other factors included. For example, plot the projects using level of complexity of each project. If you need to show fast results, are you likely to accomplish this on large, complex projects or on smaller, simpler projects? Some organizations insist on measurable results within six