Rants
Questions
Soapbox
Best Practices
Apply today for a FREE subscription to CIO Magazine!
Fri, Oct 30, 2009 14:41 EDT

|
Posted by: Jim Vaughan in Best Practices Topic: IT Organization ManagementBlog: The IT Project Management Blog
Current Rating: |
Most project managers and project management organizations tend to shy away from the use of earned value management. If you study PMI’s Project Management Body of Knowledge you will find that one of the dimensions of earned value is defined in terms of cost. Some of the terms which are used are “estimated costs,” “actual budget” and “actual cost.” When we think of these terms we think of dollars (or pounds, or franks, etc.). Yet the fundamentals of this tool discuss value. Ask people what they value in their lives and they probably will not reference money (bar a few). I would say the same is true if you asked them what they valued on their project. To use earned value management requires us to start thinking in terms of value and not money.
Since I live in the world of IT and software development most of the “cost” for a product is the salaries of the people that work on the project. Project managers typically do not have access to salary information to measure costs. They might be able to put together an average dollar amount per hour, but even that is not necessary. What we really “value” on our projects is not money but people. More specifically we value people’s time.
The key to performing earned value on our projects is using time spent as our measurement of “cost.” It is time that we truly value as our projects are resource limited. This becomes a much easier task for a project manager as they merely need to gather estimates of the time it will take to create project deliverables. Each deliverable on the project has a time estimate “value” attached to it. From the project plan each deliverable will also have a date attached to it. Using this information, project managers can measure the project against the two dimensions of value over time.
The second piece of making this work requires a tool to track the estimates against the actual time spent. In other words, you will need a time tracking system. One mistake that many organizations make here is that they attempt to collect too much information. For this method of earned value management you need only track time against the deliverable. You do not need to know the specifics of what work is being performed such as planning, designing, coding, etc. This will greatly reduce the complexity for people to measure their own time. The reduction of complexity will increase the accuracy of the data that is collected and what I have found is that you do not need to measure your project in excruciating detail. The more time you spend on creating these details the more “value” you are expending by requiring additional time to be spent on your measurements. This is the primary reason that many government project costs so much money. Government contracts require that data be kept at these very detailed levels.
To all you project managers out there, Keep It Simple, Stupid. Remember that when you are using earned value management you must think in terms of value and not cost.
Here Here!
When I learned to use EV, long ago at the Software Engineering Institute, This was how I was taught to do it. The value part of the equation is budgeted effort.
Jim V is absolutely right! It is a simple and effective way to manage software work.
Try it!
ask
Thanks for the post and discussion!
Where I Disagree
First, let me say I'm not an EVM fundamentalist! I've written before about it's shortcomings. That said, I have been a practitioner of it.
Progress on deliverables is what drives EVM. Estimate/actual costs put the value delivered into a consistent measure, and time is already factored in.
I think it's a straw man to say that EVM practitioners equate cost to value. You're not doing EVM if that's the case.
I value going to the gym every morning when it takes me 15 minutes to get there and $1 in gasoline, but I'd value it less if it took an hour and $4 in gasoline. If I planned on 15min/$1 and in reality experienced 1hr/$4 I'm in trouble (and a lousy estimator).
The fact is, properly-done EVM considers time, cost, scope, and quality. When there's sign off that something is "done" then the value is earned. If it's not done on time or in a way that meets requirements, it's not "done".
Where I Agree
It's very important to track costs and time at a level where it adds the most value.
I err towards less detail, because every additional charge code you give the team means sacrificed productivity for the sake of a little extra reporting. Half the time, that level of detail is worthless anyway. The customer and sponsor want to know how the project is doing, not how much time Tony spent on documentation this week.
As you pointed out, once tracking becomes arduous it simultaneously becomes less reliable.
Josh Nankivel
pmStudent.com