Communicating With The Past

 

Bouncing off of last week’s blog (“Communicating With The Future”) and keeping with ProjectManagement.com’s theme for August, communications, all while avoiding my usual complaint about the silliness of the notion oft repeated by communications experts that anyone identifying as a “stakeholder” ought to have ready, complete, and free access to all information pertaining to your project, I’m going to address a phenomenon I’ve been seeing more and more as my career advances. This phenomenon expresses in a variety of ways, but the one thing each of the manifestations has in common is a single question that pops into my mind each time I see it: Are we really going to do this again?

In a 1993 edition of the National Contract Management Journal[i], David S. Christensen and Scott R. Heis published an article entitled Cost Performance Index Stability, which showed rather convincingly that a project’s cumulative Cost Performance Index (CPI = Earned Value cumulative / Actual Costs cumulative) doesn’t vary more than ten points once the project has cleared the 20% complete point. Proponents of a calculated Estimate at Completion (EAC) quickly recognized the implications of what would otherwise appear to be a fairly anodyne observation, that EACs calculated using the traditional formula of dividing the CPI into the Budget at Completion would be reliably accurate to within ten points of the project’s final costs, assuming the calculation was performed after the project had cleared its 20% complete point. Probably the most common way of calculating the project’s EAC at the time of this research was to re-estimate the costs of the remaining activities, and add this figure to the cumulative actual costs. This so-called “bottoms-up” technique had an ill-deserved reputation for better accuracy, despite the dearth of supporting evidence. Supporters of the “bottoms-up” method liked to point out that, by re-estimating the project’s remaining work, events or conditions that had been unforeseen at the time of the original baseline development could be accounted for, and folded into the EAC.

But there were problems with this approach as well, including:

• According to an estimator-certifying professional organization at the time, the most accurate of the estimate types, a Detailed Estimate, required a professional estimator using off-the-shelf software to generate. It was so detailed, in fact, that, upon completion, the estimator could simply hand a copy to the procurement specialist, and that person would know precisely which pieces of equipment or material to order, and what level of personnel to hire, for each and every activity in the baseline. Even with all this rigor, the Detailed Estimate was only reliably accurate to within 15 points of the final costs – and the bottoms-up version of the EAC was rarely done to this level of precision.
• True, re-doing the estimate for the remaining work could take into account events and circumstances not foreseen when the original baseline was assembled. But it also would accommodate wishful thinking on the part of the Control Account Managers (CAMs), if not the PM himself. Early indications of a potential overrun event tended to attract unwanted upper-level management attention. Far easier to mix in some optimism into the “bottoms-up” EAC, and count on better future performance to come in on-time, on-budget.
• What happens if there’s a difference between the Budgeted Cost of Work Remaining (BCWR) and the recently-redone cost estimate for the remaining work? Should the project proceed as if the new estimate represents the best projection? If not, why not? I mean, aren’t the “bottoms-up” advocates maintaining the superiority of the latter? And, if the re-done estimate becomes the basis for future managerial decisions, doesn’t that turn the original baseline to rubber?

When the Cost Performance Index Stability study came out, its implications for using a calculated EAC over a “bottoms-up” version included:

• The calculated version was faster,
• Easier,
• Cheaper,
• And, most important of all, more accurate.

With all of this being the case, one could be forgiven for assuming that it was the calculated version of the projects’ EACs that had become prevalent, with the “bottoms-up” version attaining the same sort of status enjoyed by proponents of the Sunken Cost[ii] fallacy. Frustratingly enough, this is not the case, as “bottoms-up” EACs are not only common, they are often required on many contracts and in multiple guidance documents. And each time I see such a requirement from one of the guidance-generating organizations in the PM field, I ask myself one question.

“Are we really going to do this again?”

[i] Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.592.2040&rep=rep1&type=pdf on August 9, 2020, 15:50 MDT.

[ii] The Sunken Cost fallacy is the argument that a certain business strategy ought not to be abandoned due to the amount of resources that have already been used in pursuing it, regardless of newer information indicating that the strategy in question was a mistake.