GAO Releases Annual Bid Protest Report to Congress for FY 2021, Identifies New “Top Ground” for Sustains

On November 16, 2021, the U.S. Government Accountability Office (GAO) released its Annual Report on Bid Protests for Fiscal Year 2021. While GAO received slightly fewer protests in FY2021 than in the year prior, the overall protest “Effectiveness Rate”—meaning the percentage of cases in which the protester received relief, such as voluntary corrective action or a GAO sustain—remained relatively constant, at 48% (the rate has ranged from 44% to 51% in each of the past five years).

GAO’s Annual Report also provides a helpful summary of the most common grounds for sustained protests in the prior year. In FY2021, those grounds were as follows: (1) unreasonable technical evaluation; (2) flawed discussions; (3) unreasonable cost or price evaluation; and (4) unequal treatment. The inclusion of “flawed discussions” on the list is notable—it is the first time in recent history that discussions-based protest arguments have proven so successful. Though firm conclusions are difficult to draw based upon a single year’s data, this may indicate that GAO is taking a closer look—and holding agencies to a higher standard—at the propriety and fairness of discussions.  

The chart below shows the top sustain grounds by year. As seen below, flawed technical evaluations and flawed cost/price evaluations continue to represent some of the most consistently successful grounds for sustains, meaning would-be protesters should consider whether they have a credible basis to make such arguments when weighing an award challenge. 

Most Prevalent Sustain Grounds By Year
	2021	2020	2019	2018	2017	2016	2015	2014	2013
Flawed technical evaluation	X	X	X	X	X	X		X
Flawed cost/price evaluation	X	X		X	X	X	X		X
Flawed past performance evaluation		X			X	X	X
Flawed selection decision			X	X		X		X
Inadequate documentation			X		X		X		X
Unequal Treatment	X		X					X	X
Failure to follow evaluation criteria							X	X	X
Flawed solicitation		X
Flawed discussions	X