For those of you who missed it, earned value made the news last week (June 4, 2008), including front-page articles in the Washington Post. What happened? Lockheed-Martin was called out for failing to live up to 19 of the 32 criteria of the Cost-Schedule Control Systems Criteria (also known as "CS-squared").
Who cares?
We should! As project managers, this has implications that stretch far, far beyond Lockheed and their failings. The U.S. Government has been pushing a lot recently (in the form of Office of Management and Budget directives) for more and better earned value practice, and now, with the spotlight firmly fixed on earned value, the situation is likely to do little else but intensify. It's instructional to look at what the government found when they dug into Lockheed's practices on a massive weapons program.
The findings of the report (released by the non-profit "Project on Government Oversight") included:
You can read the full compliance report here. (PDF)
As you look at these, you might think, "Who cares? Another big government contractor gets slapped on the wrist for bad record-keeping." But they're not being chided for bad recordkeeping. They're being chided for bad earned value practice. As you look through the list here, you don't see signs of Machiavellian intent; you see signs of neglect. Let's walk through each bulleted item and see how you could avoid that with your earned value recordkeeping.
EVMS documentation? Most of it can/should/could be maintained in the project management software being deployed. Every package, from Primavera to Artemis to MS Project has an earned value component. Selecting a few and sticking to them should have minimized the degree of confusion (or vagueness) in regards to the reporting component.
Roles and responsibilities. Who is responsible for reporting on task completion? The task leads. Who is responsible for assessing the validity of the task leads' reports? The project manager. The assumption is, however, that task leads have been assigned to each work package, and the project manager is receiving regular reporting data from them at prescribed intervals.
Management reserve. While this is a bit of a discrepancy with current Project Management Institute terminology, the government's concern is clear. Reserves are designed to be applied when work within scope is not performed within scope. It should not be used to hide information. If anything, it should be used to highlight where additional funds had to be expended so that we know the real cost for future work. We should be reporting any time we use reserves and justifying their use against the original plan.
Work authorization and change control. According to the ANSI standard for earned value, work authorization is a documented trail of information on work authorized to individual, responsible organizations. It's not supposed to be another administrative layer, but it's supposed to ensure that the work we do is blessed and approved before we start doing it. Similar rules apply for change.
Cost and schedule integration. This largely ties to the 32 criteria's call for a time-phased budget baseline. Again, if we have work packages (from the plan), have a plan for authorizing the work that they entail, have the roles assigned to the work, and have a network schedule, this should (in theory) be part and parcel of the day-to-day of project management.
Inappropriate earned value techniques. I read this one with great interest, as it was a major component of the Earned Value Professional's exam (offered by the Association for the Advancement of Cost Engineering-International). On the EVP exam, there are quite a few questions about how to claim for work complete. Notably, all of them say that work has to be done before we can claim some or all credit for it. Beyond that, the CS2 criteria essentially say that it doesn't have to be one specific approach or another, but it does have to be consistent and according to a plan. That's the real heart of doing it right.
Using the data. On the last element of the government report, Lockheed was taken to task for failing to know the processes (such as critical path) and knowing how to put them to work. I have had the honor of working with Lockheed in the past, and know that they have personnel who are phenomenally adept at this kind of practice. In this instance, however, it's one thing to know what you have. It's another thing to put it to work to raise red flags and warn the client about what's around the bend. More than anything else, this is a sign that the culture needs to be ready to accept information as presented. We need to be adult enough to handle bad news.
What can we learn from the DCMA report and the Lockheed experience?
For one, we, as project managers, have value. Project managers with the insight and skills to know what earned value practice should look like and how it should be applied just saw their stock go up. The value of a PM who understands the 32 criteria and who knows how to apply earned value rose dramatically.
More than likely, however, many of the Lockheed personnel involved in this episode have that knowledge. The other thing that this should highlight for all project managers is the need to enforce and reinforce best practice on things like earned value, even when it seems to impede progress a bit and when it creates more of an administrative burden. Better to take the heat from management for demanding anal-retentive recordkeeping than to suffer the wrath of the Defense Contract Management Agency and the Project on Government Oversight.
Carl Pritchard, PMP, EVP is a certified earned value professional and a member of the ProjectConnections.com board of directors. He is a principal of Pritchard Management Associates. He welcomes comments at carl@carlpritchard.com