Working on a dashboard of sorts, that essentially is a variant of 'earned v. burned'.
Goal: GIve executive management a general view into whether our overall portfolio is on track.
Calculate total of all project days elapsed, as of "today" / total of all project durations (yields a %)
Calculate total of all project actual hours worked / total of all project allocated hours (yields a %)
If % ~match, in aggregate all projects have the amount of work expended expected within the time elapsed
What does the audience think of this method and have you used other methods?
I can get the project elapsed days and durations for the first calculation.
For the second calculation, I can get actual hours. But, I'm not sure how to easily get or calculate the number of hours *allocated* (not assigned nor ETC) for all our *active* projects.
Ideas from the experts?
It is hard to say if your proposed solution is the best one for you without knowing more about your projects and organization. E.g. if majority of your project work hours are delivered in the end of the project then the percentage of actuals/allocated would not necessarily tell you the truth of being on track.
Anyway, looking at percentage in big picture might result in overseeing issues on small critical projects. Being 30 days late in 6 months project is much but if total project duration of all your projects is 6000 days then these 30 days will never be noticed.
I would recommend using status reporting and let each project manager make status (green, yellow, red) on their projects every month. On portfolio level, you can focus on red or maybe yellow ones for critical projects.
Hope this helps you.
Yes, thanks Devor for your reply! Allow me to expand upon the present and the goal.
Currently, we indeed use a PM set status indicator or Red, Yellow, Green. It is one of 3 status elements we consider. The other two are: Schedule, whereby the active major phase milestones end dates of Define, Develop or Deploy are measured against the most recent baseline schedule and a 10% or 5% variance produces a Red or Yellow Schedule status indicator respectively; Budget, whereby if projected spend becomes 10% or 5% over planned budget, a Red or Yellow status indicator for Budget is flagged. When we report overall project status, we use "worst color wins" logic. i.e., if any one of the indicators is Red, the project is flagged Red.
The go-forward goal from our senior leadership is to create an empirical, time-valued earned vs. burned chart. This is requested on each project as well as across the full portfolio (for a 50,000 foot view). The “burned” value we are considering is how much time has elapsed (burned) for the project. The “earned” value we are considering is how much actual work has been booked/performed on the project.
For instance, consider the following hypothetical project:
We have 100 physical servers we need to virtualize
Configuring, virtualizing and testing each server takes 8 hours on average
A resource has been identified to perform the virtualization who will be 100% dedicated to the project
Therefore, that resource can virtualize 1 server per day
Duration: The project will take 100 business days to complete
1 server per day
5 servers per week
20 weeks to complete (assuming no vacations or holidays)
Effort: The project will take 800 hours of effort
8 hours per day
40 hours per week
800 hours for the full project (40 hrs per week x 20 weeks)
Let's say that due to support calls on the resource, after 2 weeks have elapsed, only 6 servers have been virtualized
This is because between support calls, the resource has only booked 48 hours to the project in the 2 week period
Example Scenario Earned vs. Burned Calculation
Earned Value (expressed as a %) = 48 hours actual work / 800 hours planned work = 6%
Burned Value (expressed as a %) = 2 weeks elapsed / 20 weeks planned schedule = 10%
Earned vs. Burned = 6% / 10% = 60% (or 0.6)
Indicates a 40% deficit in Earned Value at this point in time (should be 100% or 1.0 or greater)
Are you familiar with this method?
The challenge is that while our PMs are generally laying out their project tasks from a duration standpoint, with task start and end dates sequenced to create an entire project start and end date, the PMs do Not typically attempt to obtain accurate ETCs from resources when planning. Therefore, when time is booked to a task, there isn’t really a yardstick to measure progress against.
However, what we are evolving toward is a high-level Resource Allocation view, whereby a resource would be X% allocated for Y days or weeks. By doing the math on [the % of the resource’s time allocated] times [the amount of days or weeks of the project] the result is the number of hours expected to be booked to their work on the project. This can be divided up evenly over the project phases, front loaded, back loaded etc, but to start, we're aiming to linearly spread the allocation. With the notion that, in aggregate, the project tasks would normalize the work as relatively flat across the project (and portfolio) so that at least a mid-level view of Actual v Allocated math could be performed. This would lead to the Earned Value element of the Earned vs. Burned calculation.
I hope that I’ve managed to communicate this reasonably clearly. In any case, I’m very interested in your and others input on how you do E vs. B calculations per project and across the portfolio, even if they are of only moderate accuracy. My sr. management is eager for a quantitative way to at least detect trends in progress as we evolve our processes.