Building A PMO: Alpha & Omega of Project Management – Part I


Once you are given a mandate to build a Project Management Office (PMO), you will have taken the first of many steps to build this corporate function. But you won’t have years to develop your model, assemble a team of experts, and train the organization. Instead, you need to get started on high-priority  projects almost immediately. You need to focus on a few key elements of the PMO. Of course, you need to know the type of PMO you’re building, and you need to identify the key issues that have held back the success of your firm’s projects. But if you need to immediately launch your PMO, you need to focus on … the start and the end of your projects!

Every tool and process that your PMO puts in place can improve the success of each project. However, the processes that you put in place at the beginning and the end of each project have the ability to improve the success of EVERY project. However, even examining just  these two points in the PMO process covers a lot of ground. So, I’m going to split this into two blogs. Today, we’re going to cover the start of a project. The start of the project cycle, let’s call it the “Alpha”, is where you identify and prioritize the projects in your portfolio. Let’s examine the common steps:

  1. PORTFOLIO: A PMO is nothing without a list of projects, a project portfolio. The ability of the PMO to drive change is dependent on the quality of the portfolio. If the portfolio is filled with projects hat yield very little positive results, or that are not strongly sponsored or clearly defined, even if they are accomplished, the success may not be as great as expected. You need to either review the existing portfolio or create a new one (if one does not already exist).
  2. PROJECT IDENTIFICATION: Business should drive the identification of projects, and these projects should populate the portfolio. Of course, a business unit may not know how to identify projects, other than in terms such as “My department costs too much to operate.”  Whether projects come directly from business units and you then test their assumed value, or if you work with the business unit to define the value of the project, you need an objective method.. a common language… for defining project value and cost, which will then drive scheduling and project success.
  3. PHANTOM PROJECTS: As you transition from the old project process to a formal PMO, you will find informal and incomplete project agreements. These are “real” projects that you have not been informed of, and projects that are not ready but are starting or being completed. You need discussions with sponsors and team members to track down all of these projects, especially those without a sponsor. Expect to find projects that sponsors are not aware of, or have not signed off on. Stop these as quickly as possible, and ensure that ALL new projects have some form of Project Charter (if may be called a contract, an SOW, authorization form, or something else) that requires formal sponsor signoff.
  4. COST ESTIMATION: Cost can be difficult to accurately define, and internal projects often have misleading cost estimates. Not surprisingly, “free” elements are often ignored. The space that the sponsor’s department does not pay for, or the unused furniture sitting in a warehouse are usually thought of as free. But next year when the lease for the free space needs to be renewed or the rented furniture in the warehouse must be returned, cost structures rise. A cost-saving  project suddenly raises the cost structure. When savings are the goal, the financial assumptions must be verifiable. A PMO should have a standard set of assumptions for cost elements, and needs to challenge business assumptions when standard cost elements are missing or are unrealistically estimated.
  5. PERSONNEL ESTIMATION: Personnel are frequently part of the cost saving equation. And just as frequently, the number of positions that are reduced falls far short of the number estimated. Estimating people is very different than estimating dollars. Constraints caused by the need for “minimum organizations” often prevent the reduction of unneeded staff. If you have two workers performing the same task, and you introduce software that doubles the efficiency of that position, you should only need one worker. However, what happens when that one worker is on vacation, or out sick? Who will do the work? Instead, say you started out with five workers. Then you only need 2.5. Hmmm… can you convert a full-time worker to a part-time position, or will you need to leave staffing at 3? What if it is only one worker on each of 5 shifts or locations? Can you make any reductions at all? Understanding personnel constraints, training the firm how to estimate staff reductions accurately, and learning to put specific names or at least positions behind reductions early in the process all drive out inaccuracies.
  6. SCHEDULING: When you estimate costs, you start out with a  common understanding of money… there are 100 cents to the dollar, paper money is all the same size, nickels have different art work but are all worth 5 pennies. Too obvious to mention? Well, what are the common understandings for time estimates? How many weeks does a full-time worker work in 52 weeks? It’s not 52. A worker usually has two weeks of holidays, two weeks of sick time, and at least two weeks of vacation time. So, 52 weeks shrinks to 47. How many hours in a 40 hour week? Did you allow for lunch? What about two 15 minute breaks a day? This is required by law in many states, and it just makes sense; doesn’t every workers leave his desk for at least one cup of coffee and one bathroom break a day? That takes us from 52×40 to 47×32.5 (from 2,080 to 1,528) as a maximum of work hours. Some of your workers smoke, some have more vacation time, or have illnesses that further reduce work time. Still more time is taken up in internal meetings, training, reading emails and other time that is rarely accounted for in scheduling.
  7. HISTORICAL ACCURACY: Project history will tell you how well or poorly estimations worked in the past. Before a formal PMO is in place, historical information may be of poor quality and frequently missing. What little you find will probably show huge variations between estimates and reality for both costs and schedule. Improving accuracy in estimation both improves the value of the portfolio and client satisfaction, which in turn will improve the volume and quality of new projects. Use whatever historical information you find to add an “adjustment” to costs and schedules. If upgrades to accounting systems always tend to be delivered 30 days late and 40% over budget, add this to estimates… until the results improve.

Very broadly speaking, establishing a professional PMO creates a more reliable and effective method of delivering projects. To have the greatest impact on projects you need to do more than just establish good practices, you need to establish good REPRODUCIBLE practices. The two greatest opportunities to leverage best practices are at the start and the end of each project. At the beginning getting the estimates right, making sure that all projects in the portfolio should move forward, and ensuring that all projects reflect historical cost and schedule reality… all increase the value of your project portfolio, and kill projects that should never move forward. Our next blog will show you the opportunities you have at the end of a project to deliver value. But for today, that’s my Niccolls worth!

About these ads
This entry was posted in Best Practices, Decision Making, Delivering Services, Learning and Development, Project Management Office and tagged , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s