Project Management Central

Please login or join to subscribe to this thread

Topics: Agile, PMO, Risk Management
How to manage an infinite number of projects?

Okay, maybe not infinite, but at least many: For example, if you are running a sales program that should scale to a growing number of parallel customer projects, how would you structure the project management/PMO efficiently? There should be a number when it doesn't pay of to have a full-time project manager per project, when individual risks start showing a repeating pattern, when kick-offs should be organized centrally, when you ask all projects to use the same live cycles to allow a common reporting. It looks like the more projects you have the more you have to standardize. But is that suitable if you still want each project to acts self-organised and autonomous?
Sort By:

You can view it from another perspective- treat each "parallel customer project" as the PROJECT and your "sales programme" as a PROGRAMME. so now you are in charge of a PROGRAMME or you are a Programme Manager. Any risk that cannot be contained in the Project level, escalate to the Programme level.

But there will be instances where you will be handling multiple projects independent of each other

You have to apply the same principle that is applied in physic: simplify. To simplify, one of the strategies is what you named: standarisation. I am working into a hugh organization where we have the accountability for lot of projects that becomes from different areas. So, I am facing what you stated each day.

Mirko -

Assuming there is a high degree of similarity across these projects, at some point, you cross the threshold from project to operational process and manage the initiatives that way using normal process control methods and a detailed "playbook" for delivering these initiatives.


I'll give you a real life example and how we've dealt with it sometimes for better or for worse:

When you buy a large airplane, it's a lot like buying a car. Some features are standard, some you can select out of the catalog, and some are custom for each customer and sometimes individual plane. Depending on how much is new and/or unique, the level of complexity and risk goes up. With multiple new orders per month and each customer purchasing several units, there are hundreds of potential projects ranging from easy, to very challenging.

The first thing we need to do in order to use our limited PM resources effectively is to rank them based on risk. (Don't get too fancy or you'll spend all your resources ranking things.) The highest risk items get a PM assigned by default. Lower risk items often get assigned to someone merely to keep an eye on but not spend vast effort managing. Very low risk items we might poke at once in a while to see if there's anything to note.

When addressing the 100s of potential projects in the PMO teams, we pick a set of key milestones in the development of each and manage them in a large matrix to ensure they are on track. A red/yellow/green stoplight at each milestone makes a good visual cue. The individual PMs will at a minimum check on how things are going at each milestone and rank the current risk at that stage. The most complex ones are getting much more personal attention. Based on the overall risk and the progress to plan, each plane is also ranked on the overall program risk, again with red/yellow/or green.

When reviewing the hundreds of potential projects from a program level, we focus on the "red" or high risk planes. "Yellow" or medium risk planes, we ensure they are being managed appropriately but don't get as much leadership attention. The "green" or low risk airplanes, we don't address with the program leadership unless there is a specific question. In this way we know we are keeping an eye on everything, but focusing our resources where it will do the most good.

Hope that helps!

Thanks Mirko for bringing up this discussion of managing multiple projects. Here are my thoughts.

Increase visibility of multiple projects by planning and handling them with a multi-project management software with collaboration tool.

Clearly define project's goals, plans, roles and responsibilities will minimize inconsistencies in execution, thus producing consistent deliverables.

Utilize phase estimating (a two-estimate approach) whereby a detailed (micro) estimate is made for the immediate stage and an overview (macro) estimate is made for the remaining stages. The accuracy of estimate is progressively being refined with fewer fluctuations of requirements as the project progresses.

Mirko, you need Portfolio mgmt and categories to ensure visibility.

I have run more than 400 projects / year, and identified approx 50 of them as key projects, using a simple XLS and monthly reporting to understand what we have, who is in charge and where we stand with these 50. There were PMs who ran 3-4 of them. If you standardize in a 2nd step, you may add more criteria to the XLS.

The remainder was managed by 10 teams in a backlog manner, all of them were below 100 person days.

Ah ja, and do not try to manage resources in the beginning, focus on work.

Thank you all for the valuable input food for thought. I like Keith's risk approach, Sergio's simplification, and, Pang I'll look into some multi-project tools.

Definitely need a bigger project team as the number of projects rise. But there is eventually diminishing return as that number grows too (per PM).

Please login or join to reply

Content ID:

"One of the symptoms of approaching nervous breakdown is the belief that one's work is terribly important."

- Bertrand Russell