How detailed do you/would you go? RRS feed

  • General discussion

  • Guys,

    I'm trying to decide between two different approaches for a new project. This is perhaps more of a PM vs Project question but I figured that how one uses Project to track the two different approaches made it relevant...

    I have a software development task, here's what I know at the moment:

    • There will be 6 'blocks' of development - for which I have approximate timings (but not scope yet!)
    • There will be 6 corresponding blocks of testing and feedback dev.
    • At the moment this will be resourced with 2x developers, the exact split of tasks isn't known yet
    • This will run from around now till about November.
    • When they have scopes devs create detailed WBSs going down to 1/4 hour tasks. A two week block might have say 50-60 tasks.

    I'm tossing up two options:

    1. 1. Have the entire WBS in project, all the little 0.25 hour tasks and everything. Track it, giving me the most level of detail.
    2. 2. Have just the 6 blocks without breaking them down - meaning I may have a task of say 2 weeks, one of say 4 weeks etc.

    I can see pros and cons of both, at the moment I'm erring towards the second approach - wondering if my project plan needs to be as detailed as approach 1. (bearing in mind a different system to record timings from the developers is in place and could be interrogated if one desires (TFS).

    I'm curious as to what approach you take in your plans, thanks for reading :)

    Monday, June 8, 2015 2:20 PM

All replies

  • Rob-R,

    I vote for option 2 every time. Unless your whole project is very short, (e.g. 2 days total), breaking tasks down to .25 hours is in my mind, ridiculous. For a "normal" project of several weeks, months or years, that fine a detail will only result in someone spending all their time tracking the minutia.

    But here's another approach. For long projects, (e.g. a few years), you might consider only planning out the first 6 months in detail, and by detail I mean tasks that are no less than a couple weeks duration, or month long tasks, depending on what gives the best visibility. Tasks beyond the 6 month window are held simply as planning packages but complete with estimated work and cost values (budget). As the program nears the end of the first 6 month window (e.g. at 4 or 5 months), expand the next planning packages with detail for another 6 months. Follow that process throughout the program.

    My thoughts.


    Monday, June 8, 2015 4:06 PM
  • Thanks John, I was steering towards that way myself too. I discussed it with a colleague who seemed to agree as well.

    As regards your other approach, we sort of do that with a rolling maintenance project, whereby the sum of the project is probably a few years but each item has an approximate cost and importance, when we can we group a few of them together to make a project - it sounds similar to what you're saying. This seems to work well for this kind of activity.

    Monday, June 8, 2015 4:10 PM
  • Rob-R,

    You're welcome and thanks for the feedback. We used the planning package approach at the company where I worked. Our projects were engineering development so they had many unknown unknowns along the path to project completion. By using the planning package approach we were able to keep an overall budget per contact but yet had flexibility to adjust the details as the development progressed.


    Monday, June 8, 2015 6:18 PM