Projects within Projects?

J

jrdouce

0
I was just wondering if people kept sub-projects within projects? For example, my main project is "Deliver Fund Raising Data Mart", but there are phases and subject area that are easier to track. Right now I have two sub projects, "Research Fund Raising" and "Draft Data Model". Each of those has many NAs, and the "Deliver..." project has nothing but sub-projects(for now at least).

It seems obvious, that if a project has no NA's, it's not a project, but Deliver the Datamart is the real project. The sub's are just there to group NAs into logical collections of related activities for ease of tracking.

Any thoughts?
 
I find that too much hierarchy is a hindrance. Occasionally, I will put something with a longer horizon on a 20K (focus area) or 30K (1-2 year horizon) list, but mostly I tend to just start a new project that flows out of the old one. I just gave a midterm exam in a physics course with 85 students. The project "write midterm exam", when finished, gave rise to "write solution to midterm." I also had a project "handle special cases" running in parallel: 7 students with some particular circumstance that need to be dealt with. Suppose I had one big 4-month project "Teach Physics 171." Then I would feel a need to organize in one big place stuff that does not need to be reviewed at the same time. For example, over four months I am giving 2 1-1/2 hour lectures every week. I generally prepare them the night before, taking about 3 hours. I also review the overall organization of each topic (2-6 lectures) before starting it. This really has nothing to do with contacting the coaches who proctored exams on my behalf for student athletes who were traveling when the exam was given. Of course, many of these projects are coupled. I need to get the exams from the coaches so I can give them from the TA for grading, and the solution to the exam also needs to go to the TA. But when I've done all my projects related to the midterm, the only things left will be:

@mac post midterm solution to website
@computer email secretary midterm solution for copying
@waiting WF TA to return midterms; then review grades

which are not themselves projects. At that point, I will act appropriately on students whose grades are cause for concern, another project. Now most of the projects do have "171" somewhere in their title, so I could find them all if I had to, but in fact I don't need to do so.

I think the benefit of this approach is lots of cheap wins and closure, combined with better focus on more narrowly drawn outcomes. I'm not saying that "Build copy of Empire State Building in North Dakota" isn't a Project with a capital P, but maybe the first project in a series is really "Investigate feasibility of building copy of Empire State Building in North Dakota." But YMMV.
 
jrdouce;56579 said:
I was just wondering if people kept sub-projects within projects? For example, my main project is "Deliver Fund Raising Data Mart", but there are phases and subject area that are easier to track. Right now I have two sub projects, "Research Fund Raising" and "Draft Data Model". Each of those has many NAs, and the "Deliver..." project has nothing but sub-projects(for now at least).

It seems obvious, that if a project has no NA's, it's not a project, but Deliver the Datamart is the real project. The sub's are just there to group NAs into logical collections of related activities for ease of tracking.

Any thoughts?
Click to expand...

First, just because a project has no NA, does not mean it is not a project. It is just not an ACTIVE project. Therefore, it does not belong on the active project list. But it does belong in the someday/maybe, the someday, or the maybe, or the future, or however you break that down. This varies from forum member to forum member. How you organize that really also depends on your system. That is a whole different subject in and of itself.

Some projects really are too large and complex to keep as one. Additionally, I find that separating a project into smaller projects, makes the master project more doable, particularly the more complex ones. I love finishing projects and I can actually see myself moving toward completing the master project. However, visually I keep a straight simple active project list, whereby I don't separate this list into projects and subprojects. But I do have it organized as such in my project support material.

Since I have only just begun linking my horizons of focus to one another, I have not assigned my actual projects to any 20k, 30k or even 40k level. Therefore, all my projects, no matter how large, are, on paper, 10k level.
 
In my experience too much hierarchy is a bad thing. However, in this case I believe "Build Data Mart" should be a 20K objective with multiple supporting (though likely sequential) projects such as:

R&D Fund Raising
Finalize Fund Raising Data Mart Requirements
Identify & document transactional data sources
Develop Data Model
Write Data Integration programs (First build and scheduled updates)
Write data mart reports
Testing (Integration, System & User Acceptance)
Deploy Fund Raising Data Mart
Monitor and Report ROI

Depending on your scope, this is a minimum of a three month project and could be as much as three years or more...

I would also recommend a more formal project management methodology than is available within GTD. Unless your project is a very small data mart you'll need to understand task dependencies and resource constraints in order to finish the project in a timely manner...

Getting the granularity right can be difficult, but I find that it's best when doing this kind of project to map higher level project tasks on the team project plan (the formal project plan Gannt chart) to GTD type projects. This works best when your formal gantt-type project plans are planned at the appropriate granularity. Most formal project tasks should take from 1-4 weeks to complete and these map nicely to GTD projects.

I do this in the GTD add-in as follows:

20 K Objective: Deploy Fund Raising Data Mart (target date 12/31/08)
10 K Project/GANNT level Task: Develop Data Model (target date 04/01/08)
Next Action: Call bob re: what facts he needs in the fact table.

This assumes a reasonably modest data mart is being developed. If you're doing a "Double Data Warehouse with Cheese" then that becomes a 30-40 K goal and Develop Data Model probably is a 20K objective (or maybe even a 30K goal itself).

Hope this helps.
 
Sounds like area of focus

Jrdouce,
I would have to say that it sounds like the “Deliver Fund ….” is more of an area of focus (at least for the moment). The 2 others are definitely projects as they have current NAs. Eventually when you get closer to the “delivery date”, you can move down the “Deliver Fund…” to your projects list.

Scott STEPHEN
O&A Coaching
Efficacité et productivité sans stress
http://www.oeta.fr
 
You must log in or register to reply here.
Top