chevron-down Created with Sketch Beta.


The Procurement Lawyer Archives

Palantir USG, Inc. v. United States Adds Teeth to Government Preference for Buying Commercial

Johana A Reed


  • Discusses impact of Palantir USG, Inc. v. United States on the government's procurement of commercial items
  • Outlines the history of the dispute between Palantir and the Army over the Army's determination that Palantir's software was a noncommercial item
  • Describes the Federal Circuit's holding and how the holding reinforces the government's preference for buying commercial items
Palantir USG, Inc. v. United States Adds Teeth to Government Preference for Buying Commercial
SanyaSM via Getty Images

Jump to:

The First Gulf War highlighted a major issue with government procurement of commercial products, especially with latest information technology items. Motorola had radios that met the needs of the government; however, it was unwilling to sell them to the federal government on the commercial terms being offered pursuant to the Federal Acquisition Regulation (FAR). Ultimately, the Japanese government purchased the radios and provided them as part of their contribution to the First Gulf War. As a result, Congress passed the Federal Acquisition Streamlining Act (FASA) in 1994 FASA required federal agencies to procure commercially available products to the maximum extent practicable.Another key requirement of FASA was for market research to determine whether commercial products could satisfy the agency’s needs, with or without minor modifications, and if a commercial item was not available, to determine if nondevelopmental items other than commercial items would meet the agency’s needs.

As a result of the passage of FASA, FAR Part 12 (Acquisition of Commercial Items) was revised to more closely reflect commercial practices and terms and conditions. Many clauses that were otherwise required to be flowed down to contractors were eliminated or made optional.Moreover, FAR Part 12 began to make a distinction between commercial items and commercial off-the-shelf (COTS) items. In making this distinction, certain clauses could be applicable to commercial items, but not to COTS items. Congress has continued its commitment to purchasing commercial items in the three most recent National Defense Authorization Acts (NDAAs).

Although Congress has repeatedly made its preference for commercial contracting clear, Palantir USG, Inc. v. United States may be the first protest case at the U.S. Court of Appeals for the Federal Circuit (CAFC) brought by the federal government. The plaintiff, Palantir USG, filed a pre-award protest at the Court of Federal Claims (COFC) challenging the Army’s solicitation for the Distributed Common Ground System — Army Increment 2 (DCGS-A2) after losing a pre-award protest at the Government Accountability Office (GAO) for the same solicitation. Judge Horn granted a permanent injunction against the Army because she found that the Army had failed to comply with FASA. The government appealed to the CAFC, which affirmed the COFC’s decision.

The Distributed Common Ground System (DCGS) is comprised of “Army, Air Force, Navy, and Marine Corps ground processing systems that can share information across the Joint Force.” At the time, the DCGS-A1 combined all of the Army’s intelligence hardware and software capabilities into one system. The system included many software programs — commercial, open source, and government — that were integrated to allow the different products and components to communicate and operate seamlessly.The DCGS-A1 was operational and deployed worldwide; however, its “data architecture is over 10 years old and is based upon technology that is nearing obsolescence, with no growth margin.”

As a result, in 2014, the Army began the process of purchasing a modernized, modular data management architecture (DMA) to perform Army intelligence analysis capabilities. The performance work statement (PWS) for this new system, the DCGS-A2, stated that the requirements included the “development of new data architecture, standards based enhanced visualization and analytical tools, cloud computing and ‘big data’ analytic capabilities; cyber analytics and data integration, visualization capabilities, Cyber Operations, Interoperability, Counter Intelligence/ HUMINT, Weather, GEOINT, Geospatial Engineering and Sensor Management.” Additional requirements included interoperability and compatibility with the “Joint command system infrastructure and mission applications,” and the DMA would “serve as the architecture foundation and the heart with which the rest of the capabilities will depend on to function. The [DMA] development is therefore the focus of the first task order executed under the [DCGS-A2] contract.”

Palantir USG had developed a DMA for the CIA that it had begun marketing in 2009. The product, the “Palantir Gotham” platform, was being marketed to private and government customers. Moreover, this product had been previously purchased by the government and was listed on Palantir USG’s General Services Administration (GSA) Schedule. The Gotham platform enables “agencies to integrate, visualize, and analyze large amounts of data from different sources that reside in different databases in different formats.”

The Army followed the requirements of FASA by “charter[ing] an independent Data Integration, Visualization and Analytics (‘DIVA’) Market Study.” This study, dated July 2014, was conducted by the MITRE Corporation and looked at various potential solutions to achieve the government’s requirements. It looked at a cloud infrastructure platform provider a turn-key, and a hybrid approach. Ultimately, the study determined that the hybrid approach was the best solution.

Next, the Army issued three requests for information (RFIs) in August 2014, December 2014, and May 2015, requesting information regarding a full software development effort. Palantir USG responded to each RFI stating that it had a commercial solution that could meet the Army’s requirements and that it had provided similar systems to other government agencies. In July 2015, the Army Materiel Systems Analysis Activity issued a trade space analysis report, “which identified and evaluated technical functionality, cost, usability, schedule risk, and technical risk for DCGS-A2.” The report analyzed three options: COTS, government off-the-shelf (GOTS), and hybrid. This report concluded that a hybrid COTS-development approach was best. Finally, on July 13, 2015, the Army issued a market research report that concluded exactly the opposite: “the [DCGS-A2] development effort cannot be procured as a commercial product.”

Despite Palantir USG’s responses to the three RFIs and the two studies favoring a hybrid commercial approach, the Army issued a draft PWS that “defined the requirements for DCGS-A2 to include ‘development of new data architecture’ and completion of the ‘design, development, integration and test.’” Once again, Palantir USG responded that it had a commercial data management platform and that a full-scale development effort was not necessary. Two days after Palantir USG submitted its response on the draft PWS, October 21, 2015, the senior procurement executive signed a determinations and findings (D&F) for

“Award of a Single Source Indefinite-Delivery Indefinite-Quantity (IDIQ) Single Award Contract Exceeding $103M” for DCGS-A2. The [D&F] noted that DCGS-A2 “is heavily focused on design and development of a new data management architecture by a contractor as the systems integrator,” and “[d]evelopment of the data integration layer is pivotal and complicated by multiple interfaces and interoperability requirements with external intelligence systems.” The [D&F] concluded that:
[I]ssuing a single award IDIQ contract will mitigate many of the risks identified herein and is in the best interest of the Government. Due to the complex developmental efforts this work entails, further competition atthe task order level would interrupt development, ultimately increase price, and cause schedule slippages.
. . .
[A] single-source task or delivery order contract estimated to exceed $103 million for [DCGS-A2] Engineering Manufacturing and Development contract is authorized because the task or delivery orders expected under the contract are so integrally related that only a single source can reasonably perform the work.

Based on this D&F, on October 21, 2015, the Army issued the solicitation for the award of a single IDIQ contract for DCGS-A2, “with the simultaneous issuance of a cost-reimbursement type task order.” The solicitation “sought a single contractor to be the system data architect, developer, and integrator of DCGS-A2. The solicitation also required a software capability demonstration, which the Army contemplated ‘could include a Government Furnished Information (GFI), [COTS], [GOTS], or Open Source product(s).’”

Shortly after the solicitation was issued, Palantir USG filed a pre-award protest at the GAO, which denied the protest in May 2016. A month later, on June 30, 2016, Palantir USG filed another pre-award bid protest at the COFC. The day after Palantir USG filed its protest at the COFC, the Army issued a determination of noncommercial item. In this determination, the Army justified why it did not believe that this was a commercial item after the market research.

Count one of Palantir USG’s complaint at the COFC alleged that the Army violated 10 U.S.C. § 2377 and 48 C.F.R. §§ 10.002 and 11.002 by “refusing to solicit the data management platform as a commercial item.”Count two alleged that the Army violated FASA by refusing to solicit the entire program as a commercial product; and count three alleged that the Army violated § 2377(c) by “failing to determine whether its needs could be met by ommercial items.” The parties filed cross-motions on the administrative record in the COFC. The COFC “granted judgment in Palantir’s favor, concluding that the Army failed to determine whether commercial items meet or could be modified to meet the agency’s needs and that, by failing to do so, the Army acted in an arbitrary and capricious manner in violation of 10 U.S.C. § 2377.” The COFC also found that “the Army’s actions caused Palantir to suffer a ‘non-trivial competitive injury which can be addressed by judicial relief,’ [and] permanently enjoined the Army from issuing a contract award under the protested solicitation.”

The Army appealed the COFC decision to the CAFC. The CAFC reviews the COFC’s ruling on cross-motions for judgment de novo. The standard of review at the CAFC is the same as at the COFC. “The Army’s procurement decision must be set aside only if it is ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law,’ or ‘without observance of procedure required by law.’” The government raised two issues on appeal: “(1) whether the trial court went beyond the statutory and regulatory language of FASA and its implementing regulations and imposed heightened obligations; and (2) whether the trial court wrongly discarded the presumption of regularity and substituted its judgment in determining that the Army acted arbitrarily and capriciously and in violation of 10 U.S.C. § 2377.”

Ultimately, the CAFC concluded that although the court gives deference to the Army’s procurement, “the Army’s procurement actions in this case were arbitrary and capricious and in violation of § 2377.” The CAFC based its decision on the overwhelming evidence that a commercial solution existed. Not only had Palantir USG submitted responses to each RFI and the draft PWS, but two of the three studies also had indicated that a hybrid commercial solution was available. And the one study that found otherwise did so without any support for its conclusions. Finally, there were three operational needs statements from other Department of Defense agencies requesting Palantir USG’s data management platform.

The CAFC rejected the Army’s contention and market research report that stated that there were no commercial offerings regarding data fusion, cyber intelligence support, and interoperability with the DCGS integrated backbone. Once again, the court found that the Palantir Gotham platform may provide data fusion capability and that there was ample evidence that the Gotham platform was interoperable with the DCGS integrated backbone. The court also found that there was evidence that there may be a commercial solution available for the cyber intelligence support.

After discussing and recognizing that FASA did not have a specific requirement regarding what or how much documentation must be included in an agency’s record regarding the market survey, the CAFC determined that if a court orders an agency to provide support for its decision, it must do so: “On this record, we conclude that the Army did not rationally use its market research results to determine whether there are available commercial items that: ‘(A) meet the agency’s requirements; (B) could be modified to meet the agency’s requirements; or (C) could meet the agency’s requirements if those requirements were modified to a reasonable extent.’”

The court also rejected the government’s second ground for challenging the COFC’s decision, finding that the Army’s ignoring of the extensive evidence that there was in fact a commercial solution was arbitrary and capricious. This finding rebutted the presumption of regularity.

Lastly, the CAFC stated: “To be clear, we are not suggesting that the Army must choose Palantir as the awardee. We simply affirm that the Army must satisfy the requirements of 10 U.S.C. § 2377.” The court also awarded costs to Palantir USG.

The lessons to be learned from this case are as follows: First, if you work hard enough and ask enough organizations, you can eventually find an organization that will support your position. Second, once you find that organization, it must document why it has agreed with your position. Third, if a court asks you to provide documentation to support your position, you must provide the documentation; it cannot be simply conclusory statements without explanations. And finally, even if you find an organization that supports your position, you cannot ignore all the other evidence you have that does not support your position.