I am creating a work breakdown structure for a project I am on, but my supervisor seems to think it is taking too long. He says I should be able to create a usable document in somewhere between 2-8 hours. However, I've been working on the WBS for several days now and my work only goes as fast as the conversations with the project team members. I feel like I'm being productive but, based on his feedback, I am concerned that I am going into too much detail. How can I get a better sense of the right level of detail, and produce a WBS that provides an understanding of the work without taking too much time to produce?
A good WBS is created in tandem with the people who are accountable for the work packages. A poor WBS is created in isolation. You can create a poor WBS in 2 hours. (I can do it in 1 hour for half the cost, and 10% of the quality).
Up to some limit, the more time you spend on the WBS, the more effective it will be in planning and predicting project closure.
That said, your boss has expressed an unreasonable request. If my boss asked for that, I'd submit three WBS within 5 hours - an optimistic, a pessimistic and a PERT estimate. I'd promise only that the results were within +/- 100% of the most likely value, and I would include a cyclical activity for reification of the resulting project plan narrowing those confidence intervals.
I would also start to plan a campaign to educate my manager on the true value of a WBS - if you develop it fast and cheap it will be a very poor tool for predicting the future. If I couldn't come to agreement with my manager, I'd start circulating my resume.
The complexity of your product the project is creating dictates how the complexity of your WBS. Can you imagine what the WBS of a fighter aircraft looks like? However, one person would not be responsible for building the entire WBS; it would be created by SMEs of the very components being built that create the aircraft when all put together.
Your OP reads as if you are the only one creating it and are slowly and painstakingly interviewing those SMEs as necessary. I cannot opine on an acceptable duration since that varies across projects, industries, etc., but it seems your chosen process is not the most efficient, if my assumption is right.
Once you break down the first couple of levels of your product, you need to charge others to continue decomposition in their respective areas.
Regarding how low you go, there's no one answer fits all solution. You break it down as far as you need to go so you can manage the work and no further. Some would put rules down such as you break down the product and its tasks to the point that its planned duration does not exceed a certain period of time. Those kinds of rules might help you in this case.
That depends on a few different factors. What level of detail the business expects, the complexity of the project, and how fast you yourself do it.
For the first factor, your best bet, I'd say, is to look at a previous WBS from a previous (ideally successfully-completed) project, ideally around the same level of complexity as this project, and see how much detail it went into. While there's always going to be some subjectivity in questions such as 'How much detail should I go into?', precedent is usually a good place to start. If possible, you could also try to find out how long it took to make that previous WBS, to get a benchmark for time as well as detail.
Second factor only really matters if you can't find a precedent that's of a comparable complexity. In which case, you'll have to estimate the relative difference in complexity between this project and your precedent, and adjust your benchmark accordingly.
As for how fast you yourself do it, I'm assuming from your question that this is one of your first ones, so it's probably going to take longer. Don't be too distraught if you end up going over by a bit.
this is not a question about project management but about comunication between you and your supervisor. he may not have defined good enough what he expects... he might be looking for high level tasks just to get the management approval before moving on. maybe someone already made some work that you should rely on instead of inventing the wheel. tell your supervisor what you did and what took you so much time and ask him where he thinks you have not been efficient....and if he can guide you the first time... this is if you believe he is a good guy. he might just enjoy abusing you or teasing you...
Ask your supervisor which level of the WBS he want you to operate on. You can definitely spend weeks in this task depending on the level you go.
Just keep in mind that you may not need to go to the very low level of details. If you can isolate nodes that can be assigned to a resource and is not longer than 40h duration you can start work. The further breakdown can happen later, as long as it doesn't compromise the duration and effort.
the quickest and by far the best way to create a 'first-cut' WBS is to get the right people in a room (or online) and jointly develop the WBS. I have done this on a major aircraft project and developed the first-cut WBS in a single afternoon. Populating it is then a different story - a WBS is not just a picture - it is a description of the scope of the project. Populating the WBS properly will take time - weeks at least. Forgot to add - we have a post that describes how to create a WBS - hope this helps. http://www.pmis-consulting.com/how-to-create-a-work-breakdown-structure-wbs/