The process for building the WBS is as follows:
1. Break the project into lower level “chunks of work”
First determine the large chunks of work that must be completed for the entire project to be completed. At this point, it does not matter how you define the large chunks of work. It is only important that all the work is identified at the end of the process. For instance, a traditional breakdown might be ‘planning / analysis / design / construct / test / implement’, which lays out the project in a high-level timeline. The breakdown could also be by deliverable – for instance ‘online application / data warehouse / data marts / user query tools’. It could also be by some functional breakout such as ‘extract data / load data / report on information’. You can break down the work into whatever structure makes sense for your project.
2. Evaluate each lower element of the WBS
Check each lower level component to see if it meets the following two criteria.
- Is the effort required to complete the work component less than your estimating threshold? A rule of thumb is 80 hours or less, but this could be higher or lower based on the size of the project.
- Do you (or someone on your team) understand the detailed work required to complete this work component? The detailed work components on the WBS will ultimately be moved to the schedule. You don’t want to have activities on your schedule that no one on your team fully understands.
If you understand the detailed work required to complete the component and if the estimated level of effort is smaller than the estimating threshold you do not need to break the component down further.
3. Continue to break down each component as needed
Work components that require more effort than the estimating threshold, or work components that you do not fully understand should be broken down further.
It is important to understand that as you break the work down, you are ultimately going to create activities. The detailed activities from the WBS are what get carried forward to the schedule.
This process of breaking the work components into a lower level set of components should continue until all of the work components are represented as granularly as necessary to ensure that no activities have estimated effort larger than the estimating threshold, and that you understand the work. This takes you to levels 3, 4, 5 etc. Rarely would you need to break the work down greater than five levels.
There is one exception to this process. If your project is large, it is likely that you may not know enough to be able to break all of the work down to this discreet level. If you cannot break the work down into small enough components you may still be okay as long as those higher-level components do not need to be worked on any time soon. In that case, you can leave the work components at the higher level until you get closer to execution (three months), at which time you will know enough to be able to break the work down at a more granular level.
Since you have not sequenced the work yet, you may not know whether the work needs to be done sooner or later. Nevertheless, if you do not know enough to break down the work lower that the estimating threshold, leave it at that level until the work is sequenced later. At that point you will know if you have a problem. If the work needs to be done relatively soon, you will need to figure out how to break the larger components into a lower level of detail so that you can assign the work to a team member. If the work ends up being executed in the more distant future, the components can be left at the higher level for now.