On December 12, 2025, GAO released its Bid Protest Annual Report to Congress for Fiscal Year 2025, which provides bid protest statistics and other information regarding GAO’s protest system.
- The number of protest filings decreased by 6% from FY24. After a 22% increase between FY22 and FY23 to surpass 2,000 cases, the number of cases has steadily fallen over the last two years. This year’s decline in protest filings is likely due to the overhaul of procurement practices that took place at the start of 2025 — an overhaul which significantly reduced the overall number of contract awards. However, the number of protest filings in FY25 (1688) is still higher than the number of filings in FY22 (1658).
- GAO’s sustain rate was 14%. This sustain rate is consistent with the sustain rates of 13%-16% in recent fiscal years (with the exception of FY23, where a large number of protests challenging the CIO-SP4 GWAC procurement led to an unusually-high sustain rate of 31%).
- The effectiveness rate remained high, at 52%. A significant number of protests filed at GAO do not result in a decision on the merits because agencies voluntarily take corrective action before a decision on the merits is reached. As a result, the more indicative statistic for favorable outcomes in a bid protest is the “effectiveness rate,” which measures the percentage of all protests filed in which the protester obtains “some form of relief from the agency . . . either as a result of voluntary agency corrective action or [GAO] sustaining the protest.” The FY25 rate of 52% matches the FY24 rate, and provides evidence that most protesters are using the protest process to raise legitimate claims and concerns.
- The number of hearings remained low, at only 3 cases. Last year, we observed that GAO held a hearing in only 1 case, which suggested that an upward trend in hearings during FY23 may have been an aberration. This year’s report corroborates that conclusion, revealing that GAO held hearings in just 3 FY25 cases (i.e., 0.5% of all protests filed).
- The number of merit decisions remained low, consistent with FY24 but marking a departure from earlier years. GAO issued just 380 decisions resolving protests on the merits in FY25. That total is consistent with FY24, but marks a significant decrease from earlier in the decade: For example, 455 of the 1658 cases filed in FY22 were decided on the merits, and similarly, in FY21, 581 of the 1897 cases filed were decided on the merits. This decrease in merit decisions, combined with the continued high effectiveness rate, suggests that more cases are being resolved through voluntary corrective action by the agency.
- One federal agency declined to fully implement GAO’s recommendations. GAO also reported that, in a rare occurrence, one federal agency declined to implement GAO’s recommendations. In ATP Gov, LLC, B-422938; B-422938.2, Dec. 12, 2024, 2024 CPD ¶ 306, GAO found that the agency had announced a certification requirement in a solicitation, but then failed to apply that requirement when awarding the contract. GAO recommended that, in addition to reimbursing the protester for its protest costs, the agency either reevaluate proposals in accordance with the solicitation, or amend the solicitation. The agency, however, took the position that — because there had not been an automatic stay during the pendency of the protest — stopping performance would be unacceptable. The agency ultimately advised GAO that it intended to reimburse the protester for the costs of filing and pursuing its protest, but would not follow the rest of GAO’s recommendation. GAO notified Congress of these events in a June 2, 2025 letter, and recommended that Congress enact a private bill directing the agency to reimburse the protester for proposal preparation costs. Congress has, to date, not weighed in on GAO’s recommendation.
GAO’s annual bid protest report continues to provide useful information regarding GAO’s protest system, and indicates that the protest process continues to be an effective mechanism for oversight of the federal procurement process.