Much of the pixel ink spilled in advancing Project Management has to do with encouraging organizations that do not use its precepts to change their ways, recognize its utility, and thenceforth thrive. Make no mistake – I’m all for this.
To a point.
After that point, I have to ask, how much is too much? In my point/counterpoint-format debate, documented in PMNetwork (February 2008) with Dave Hillson (a.k.a., “The Risk Doctor”), David stated “…risk is an uncertain event or condition that—if it occurs—has a positive or negative effect on achievement of one or more project objectives.”[i] I quickly pointed out that such a definition excluded no part of the management process. I believed then, as I do now, that the risk management (no initial caps) crowd were (and still are) seriously overextending their epistemological purview. My understanding of this assertion is that almost every aspect of managerial decision-making can be reliably informed by performing Gaussian curve-based analyses, which I whole-heartedly reject. Generally speaking, they simply didn’t know when their preferred techniques’ efficacy started to deteriorate into irrelevance.
Sadly, they’re not alone. As GTIM Nation knows, I’m a huge advocate of Earned Value Management Systems (EVMSs), and have no problems asserting that they are an absolutely essential aspect of the Management Information Systems supporting any real project of any real size or importance. And yet even here are multiple opportunities to over-play its importance at the urging of self-appointed experts, including those who belong to organizations that issue “guidance” on EVMSs. How does one know if those pushing Earned Value have gone too far?
Probably the most obvious such indicator has to do with the development and use of the Estimate at Completion (EAC). Again, GTIM Nation knows of my opposition to the so-called “Bottoms Up” EAC, which involves re-estimating the remaining budget, adding that figure on to the cumulative actuals, and declaring victory. This method is not only wasteful in time, energy, and money, but it’s also demonstrably inaccurate when compared to its EVM-calculated counterpart, via the formula:
EAC = BAC / CPI
…where EAC is the Estimate at Completion, BAC is the Budget at Completion, and CPI is the Cost performance Index. After the Project’s 20% completion point, this formula has been shown to be consistently reliable to within 10% of the Project’s true at-completion costs[ii]. Also, since these data points are automatically produced by the EVMS, all the way down to the reporting level of the Work Breakdown Structure (WBS), this EAC is easily generated, again all the way down to the reporting level of the WBS. In other words, calculating the EAC not only alerts the PM to the existence of a cost variance problem, it will also pinpoint which part the Project is having the difficulty. Conversely, the re-estimate approach, which is essentially re-doing the Estimate to Complete (ETC), is probably not going to be reliably accurate to within 15%, and typically requires a professional estimator, with access to the most recent rate tables. It also does not take into account the Project’s performance to that point.
As wrong-headed as this approach is, it actually gets worse. At least one of the aforementioned PM “guidance” organizations has issued documents that strongly assert the utility of time-phasing this ETC, and measuring performance against it! This, dear GTIM Nation members, is the very definition of a rubber baseline. Which time-phased budget (Budgeted Cost of Work Scheduled, or BCWS for us old-timers) documents are we using, hmmm? The assertion that the re-done one should be considered more accurate, because it takes into account more recent data, dontcha know, completely discounts the fact that it does not take into account Project performance to date, which, again, has been shown to be the far more accurate predictor of future outcomes.
Then there’s the discussion of whether or not an EAC should be considered a single information point. To my thinking, the answer is absolutely yes, mainly because its multi-point, time-phased alternative is simply a derivative of comparing a plan to actual costs, a.k.a. watching the actuals go by. The valid way to interpret the EAC is the most probably final cost of a piece of scope if performance continues at its current pace. Don’t like what the calculated EAC is telling you? Then change your performance, don’t mess around with overextending Earned Value data all the way to irrelevant comparisons.
To reiterate, Earned Value is the only way to properly measure cost performance on your Project. But don’t overdo it – it makes us look as overextended as the risk managers (no initial caps).
[i] Hatfield, M., & Hillson, D. (2008). Danger ahead?: Some project managers contend there's a silver lining in risk management; others say it's called opportunity. PM Network, 22(3), 76–80.
[ii] Cost Performance Index Stability: Fact or Fiction? Journal of Parametrics 12(1):27-40, December 2011
