Attention: open in a new window. PDFPrintE-mail

User Rating: / 0
PoorBest 

"How many tasks should this project have?" project management students tend to ask me in class at Boston University. "How much detail should a project planhave?" clients often ask.

The usual mistake project managers (PM) make is tolay out too many tasks; subdividing the major achievements into smaller and smaller subtasks untilthe work breakdown structure (WBS) is a "to do" list of one-hour chores. It's easy to get caught up in the idea that a project plan should detail everything everybody is going to do on the project. This springs from the misconceived logic that a project manager's job is to walk around with a checklist of 17,432 items and tick each item off as people complete them. This view is usually linked with another fallacy: that the project plan should be a step-by-step procedure for doing everything in the project in case we have todo it again. If the PM is managing the wrong things, this may be handy because we increase the odds of having to do this project again.

Project sponsors encourage these fallacies by marveling at monstrous project plans because they make it seem that the PM has thought of everything. Unfortunately, on significant cross-functional projects, there is absolutely nochance that the project manager will think of everything. The subject matter experts and specialists are the ones we must hold accountable for that.

The result of these fallacies is that PMs produce project plans with hundreds or even thousands of tasks. Many of them have durations of a few hours or a few days. Does this level of detail give us better control and lead to successful projects? In our view, a "to do" list approach does not give effective control, and it interferes with the achievement of a successful end result.

First, the laundry list approach leads to, and even encourages, micromanagement of the people working on the project. Micro-management is appropriate when you have slackers and nincompoops working for you, but few project teams are composed entirely of these losers. The majority of your project team members will not thrive under micro-management. This style tends to encourage dependency on the project manager rather than independence where people are held responsible for their results.

Second, PMs are consistently more effective when they hold people accountable for reaching measured achievements rather than completing a list of tasks. How often does it happen that people complete a list of tasks and achieve nothing? When we base our assignments and monitoring on well conceived and measurable achievements, no one loses sight of the desired end result.

Third, the laundry list approach is hard to maintain. People have to report on many tasks which decrease the odds of receiving accurate and timely status reports. The PM, with or without clerical support, has a great deal of data entry to do to input all this status data. Amid the pressure of on-going multiple projects, tracking can fall behind and may even be dropped because the amount of effort is too large. This may sound like a stupid and improbable solution, but it happens with alarming frequency even on large and important projects. The logic is, "No one is looking at all that detail anyway, so why spend all that time to catch up?"

As a general rule, we like to see the majority of assignments in a project planhave durations that are between one week and eight weeks long. Coupled withthis, we advocate weekly status reporting of hours worked percentage complete and an estimate of the hours of work remaining to complete theassignment. This combination allows the project manager to maintain good control while placing the responsibility for achievements on the team members.

Using the work breakdown structure (WBS) for cross-functional corporate projects,you have the opportunity to design a assignment and monitoring process. As part of our Achievement-driven approach, we recommend breaking work downinto "packets" of achievement for which you will hold people and teams accountable.

by Marcus Goncalves, April 2003.

BACKTOTOP