This article is an excerpt from The Art of Tendering: A Global Due Diligence Guide, which is available for purchase.
In its March 2006 decision in United Paradigm Corporation, the U.S. Government Accountability Office (GAO) ordered the Department of the Air Force to conduct a re-evaluation after determining that the initial evaluation of prior experience was unreasonable. The dispute stemmed from a Request for Proposals (RFP) for fuel management services at the Wright-Patterson Air Force Base in Ohio. The evaluation turned on the assessment of past performance and price, which were weighed almost equally in the assessment of “best value.”
The GAO determined that the Air Force’s evaluation of prior performance was based on the following undisclosed “relevance” rating:
The Air Force arrived at the performance confidence rating for each offeror using the following methodology, which was not disclosed to offerors in the RFP. After rating the relevance of each contract identified by the offeror in its proposal, the Air Force averaged these ratings to arrive at an overall “relevance” rating. It then integrated this rating with an overall past performance rating, also obtained by averaging the point scores on the performance questionnaires furnished by the offeror’s references. This process yielded an overall performance confidence rating. Thus, for example, an average relevance rating of “highly relevant” combined with an average past performance rating of “exceptional/high confidence” yielded an overall performance confidence rating of “exceptional/high confidence”; an average relevance rating of “relevant” combined with an average past performance rating of “exceptional/high confidence” yielded an overall performance confidence rating of “very good/significant confidence”; an average relevance rating of “somewhat relevant” combined with an average past performance rating of “exceptional/high confidence” yielded an overall performance confidence rating of “satisfactory/confidence”; and an average relevance rating of “not relevant” combined with an average past performance rating of “exceptional/high confidence” yielded an overall performance confidence rating of “neutral/unknown confidence.”
The GAO found this was unreasonable because it penalized those with relevant experience by averaging it against their less-relevant experience, an approach that was both non-transparent and irrational:
We think that the agency’s approach to evaluating past performance was unreasonable because it had the effect of penalizing offerors with relevant experience such as UPC and the other offeror noted above for their non-relevant experience. For example, using the agency’s methodology, an offeror that had performed four highly relevant contracts well would have received a higher performance confidence rating than an offeror that had performed four highly relevant and four somewhat or not relevant contracts equally well. Such a result is, in our view, clearly irrational.
The agency’s methodology is further unreasonable in that it gave equal weight in the calculation of offerors’ past performance ratings to highly relevant and non-relevant performance.
The GAO therefore ordered a re-evaluation of past performance. In the event that the re-evaluation changed the initial outcome, the GAO ordered that the originally awarded contract be terminated and re-awarded based on the new evaluation ranking.
This case illustrates that purchasing institutions should be clear in their disclosure of evaluation procedures, including those that assess past performance. Relying on undisclosed criteria can create significant legal exposures, particularly when those undisclosed factors have the effect of skewing the evaluation’s outcome.
