FRIEDLAND, Circuit Judge

Allan J. Graf (argued) and Dean Francis Pace, Carlsmith Ball LLP, Los Angeles, CA, for Plaintiff–Appellant.

Kimberly A. Dunne (argued), Sean A. Commons, and Brent L. Nichols, Sidley Austin LLP, Los Angeles, CA, Alan Charles Raul, Sidley Austin LLP, Washington, D.C., for Defendant–Appellee.

OPINION

FRIEDLAND, Circuit Judge:

Steven Mateski appeals the dismissal of his False Claims Act suit against Raytheon Co., in which he alleged fraud in the performance of a Government contract. Mateski argues that the district court erred in holding that his Complaint was based upon prior public disclosures and was thus precluded by the public disclosure bar of the False Claims Act. We agree. Mateski's Complaint alleges fraud that is different in kind and degree from the previously disclosed information about Raytheon's problems in performing on the contract at issue. We therefore reverse and remand for further proceedings.

I.

Between 1994 and 2002, the National Oceanic and Atmospheric Administration, Department of Defense, and NASA contracted with various companies to design and build the National Polar–Orbiting Operational Environmental Satellite System (“NPOESS”), a system for collecting meteorological, oceanographic, environmental, and climatic data. Raytheon entered a contract to design and build a Visible Infrared Imaging Radiometer Suite (“VIIRS”) sensor, which would be part of NPOESS. Eventually, the Government agencies awarded a satellite integration contract to a prime contractor, Northrop Grumman, and incorporated previously awarded NPOESS contracts, including Raytheon's VIIRS contract, as subcontracts to the prime contract.

The NPOESS project incurred many delays and cost overruns. Beginning at least as early as 2003, VIIRS began to attract public attention as a source of these problems. For example, a 2004 Government Accountability Office (“GAO”) report stated, “At present, the program office considers the three critical sensors—VIIRS, CMIS, and CrIS—to be key program risks because of technical challenges that each is facing.” U.S. Gov't Accountability Off., GAO–04–1054, Polar–Orbiting Envtl. Satellites: Information on Program Cost and Schedule Changes 18 (Sept. 2004). A 2005 GAO statement explained that “VIIRS sensor development issues were attributed, in part, to [Raytheon's] inadequate project management.” U.S. Gov't Accountability Off., GAO–06–249T, Polar–Orbiting Operational Envtl. Satellites: Tech. Problems, Cost Increases, & Schedule Delays Trigger Need for Difficult Trade-off Decisions 18 (Nov. 2005). A report from the Office of Inspector General at the U.S. Department of Commerce noted that “[i]nadequate oversight, in effect, postponed the critical evaluations and decisions needed to replan the program's faltering elements and contain the cost and schedule overruns. Time and money were thus wasted as the problems with NPOESS continued unchecked.” U.S. Dep't of Commerce, OIG–177794–6–0001, Nat'l Oceanic & Atmospheric Admin.: Poor Mgmt. Oversight & Ineffective Incentives Leave NPOESS Program Well Over

Budget & Behind Schedule 12 (May 2006). A slew of news articles also reported cost overruns and schedule delays with NPOESS, including in Raytheon's work on VIIRS.

Steven Mateski, an engineer who worked at Raytheon from 1997 to 2006, was assigned to work on VIIRS beginning in 2005. Mateski filed a complaint in June 2006 in federal district court alleging that Raytheon had violated the False Claims Act (“FCA”), 31 U.S.C. §§ 3729–3733, by failing to comply with numerous contractual requirements in the development of VIIRS, fraudulently covering up areas of noncompliance, and improperly billing the Government for erroneous and incomplete work.

Six years after Mateski filed his initial complaint, the United States declined to exercise its right under the FCA to intervene in Mateski's suit. Mateski then filed a fourth amended complaint (“Complaint”). The Complaint alleges that “[f]rom 2002 to at least 2012, Defendant [Raytheon] has knowingly submitted false NPOESS VIIRS claims for payment, whereby the United States Government has been induced to pay money that it would not have paid if Defendant [Raytheon] had disclosed the true defective nonconformances with the NPOESS VIIRS specifications and requirements.” As did the original complaint, this Complaint makes numerous specific allegations, including: creation of false waivers; improper (and forged) signoffs certifying work performed; failure to rectify issues relating to electrostatic discharge; cross contamination of flight and non-flight quality materials; and use of prohibited materials such as tin plating.

Raytheon moved to dismiss for lack of subject matter jurisdiction. It argued that the suit was barred by § 3730(e)(4)(A) of the FCA, also known as the “public disclosure bar,” which at the time of filing provided: “No court shall have jurisdiction over an action under this section based upon the public disclosure of allegations or transactions.” 31 U.S.C. § 3730(e)(4)(A) (1986); United States ex rel. Hartpence v. Kinetic Concepts, Inc., 792 F.3d 1121, 1127 (9th Cir.2015) (en banc). The district court granted Raytheon's motion to dismiss, explaining, “[I]t was publicly known that there was rampant mismanagement, deviations from protocol, and other problems with VIIRS.... [W]hile the public disclosures d[id] not discuss the problems on the VIIRS program in the level of detail that Mateski does in his [Complaint], the allegations are nonetheless the same for the purposes of 31 U.S.C. § 3730(e)(4)(A).”

II.

We review de novo a district court's dismissal for lack of subject matter jurisdiction and its interpretation of the False Claims Act. United States ex rel. Hartpence v. Kinetic Concepts, 792 F.3d 1121, 1126 (9th Cir.2015) (en banc). We review for clear error a district court's findings of fact that underlie its decisions on subject matter jurisdiction. Id. at 1126–27. Whether a particular disclosure triggers the public disclosure bar is a mixed question of law and fact that we review de novo. United States ex rel. Found. Aiding The Elderly v. Horizon W., Inc., 265 F.3d 1011, 1013 (9th Cir.), amended on denial of reh'g, 275 F.3d 1189 (9th Cir.2001); see also United States v. Alcan Elec. & Eng'g, Inc., 197 F.3d 1014, 1017 (9th Cir.1999). The plaintiff “bears the burden of establishing subject matter jurisdiction by a preponderance of the evidence.” Alcan, 197 F.3d at 1018.

III.

The FCA prohibits “knowingly present[ing], or caus[ing] to be presented, a false or fraudulent claim for payment or approval; [or] knowingly mak[ing], us [ing], or caus[ing] to be made or used, a false record or statement material to a false or fraudulent claim” to the federal government. 31 U.S.C. § 3729(a)(1)(A), (B). The FCA allows private individuals, referred to as “relators,” to bring suit on the Government's behalf against entities that have violated the Act's prohibitions. 31 U.S.C. § 3730(b)(1); see also United States ex rel. Hartpence v. Kinetic Concepts, 792 F.3d 1121, 1123 (9th Cir.2015) (en banc). Such suits are commonly called qui tam suits. Vt. Agency of Nat. Res. v. United States ex rel. Stevens, 529 U.S. 765, 768, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000) (“Originally enacted in 1863, the False Claims Act ... is the most frequently used of a handful of extant laws creating a form of civil action known as qui tam.”); see Kinetic Concepts, 792 F.3d at 1123.

The FCA's public disclosure bar deprives federal courts of subject matter jurisdiction when a relator alleges fraud that has already been publicly disclosed, unless the relator qualifies as an “original source.” Kinetic Concepts, 792 F.3d at 1123 (quoting 31 U.S.C. § 3730(e)(4)). Under the public disclosure bar:

No court shall have jurisdiction over an action under this section based upon the public disclosure of allegations or transactions ... unless the action is brought by the Attorney General or the person bringing the action is an original source of the information.

31 U.S.C. § 3730(e)(4)(A) (1986) (emphases added). 6 The public disclosure bar is intended to encourage suits by whistle-blowers with genuinely valuable information, while discouraging litigation by plaintiffs who have no significant information of their own to contribute. See Graham Cty. Soil & Water Conservation Dist. v. United States ex rel. Wilson, 559 U.S. 280, 294–95, 130 S.Ct. 1396, 176 L.Ed.2d 225 (2010). A prior version of the FCA had attempted to achieve a similar goal by banning “qui tam actions based on information already in the Government's possession.” Id. at 294, 130 S.Ct. 1396. Congress decided that provision “thwarted a significant number of potentially valuable claims. Rather than simply repeal [it], however, Congress replaced it [in 1986] with the public disclosure bar in an effort to strike a balance between encouraging private persons to root out fraud and stifling parasitic lawsuits.” Id. at 294–95, 130 S.Ct. 1396.

“The public disclosure bar is triggered if three things are true: (1) the disclosure at issue occurred through one of the channels specified in the statute; (2) the disclosure was ‘public’; and (3) the relator's action is ‘based upon’ the allegations or transactions publicly disclosed.” Malhotra v. Steinberg, 770 F.3d 853, 858 (9th Cir.2014) (quoting 31 U.S.C. § 3730(e)(4)(A) (1986)). It is undisputed that the first two elements of this test are satisfied in this case. The statements Raytheon relies upon in invoking the bar occurred through the channels specified in the statute—news media, congressional hearings, and GAO reports—all of which were public. See United States ex rel. Found. Aiding the Elderly v. Horizon W., Inc., 265 F.3d 1011, 1014 (9th Cir.), amended on denial of reh'g, 275 F.3d 1189 (9th Cir.2001) (explaining that “[p]ublic disclosure can occur in one of only three categories of public fora: (1) in a ‘criminal, civil, or administrative hearing;’ (2) in a ‘congressional, administrative, or Government Accounting Office report, hearing, audit, or investigation;’ or (3) in the ‘news media.’ ”) (quoting A–1 Ambulance Serv., Inc. v. California, 202 F.3d 1238, 1243 (9th Cir.2000)).

The dispute in this case is thus about whether Mateski's action is “ ‘based upon’ the allegations or transactions publicly disclosed.” Malhotra, 770 F.3d at 858 (quoting 31 U.S.C. § 3730(e)(4)(A) (1986)). This depends on: (A) whether the publicly available information about Raytheon's work on VIIRS contained an “allegation or transaction” of fraud; and, if so, (B) whether Mateski's Complaint was “based upon” said “allegation or transaction.” See United States ex rel. Zizic v. Q2Administrators, LLC, 728 F.3d 228, 235 (3d Cir.2013). We address these issues in turn.

A.

The False Claims Act's public disclosure bar uses the terms “allegations” and “transactions” without defining either term. 31 U.S.C. § 3730(e)(4)(A). Courts have interpreted “allegation” to refer to a direct claim of fraud, and “transaction” to refer to facts from which fraud can be inferred. See, e.g., Zizic, 728 F.3d at 235–36 (“An allegation of fraud is an explicit accusation of wrongdoing. A transaction warranting an inference of fraud is one that is composed of a misrepresented state of facts plus the actual state of facts.”) (citation omitted); United States ex rel. Springfield Terminal Ry. Co. v. Quinn, 14 F.3d 645, 653–54 (D.C.Cir.1994) (defining “allegation” as a direct claim of fraud and “transaction” as a combination of facts from which “readers or listeners may infer” fraud).

For purposes of the public disclosure bar, we have held that “[t]he substance of the disclosure ... need not contain an explicit ‘allegation’ of fraud, so long as the material elements of the allegedly fraudulent ‘transaction’ are disclosed in the public domain.” Found. Aiding, 265 F.3d at 1014; see also A–1 Ambulance Serv., 202 F.3d at 1243. We have explained:

[I]f X + Y = Z, Z represents the allegation of fraud and X and Y represent its essential elements. In order to disclose the fraudulent transaction publicly, the combination of X and Y must be revealed, from which readers or listeners may infer Z, i.e., the conclusion that fraud has been committed.

Found. Aiding, 265 F.3d at 1015 (alteration in original) (quoting Springfield Terminal, 14 F.3d at 654). We have further explained that, “in a fraud case, X and Y inevitably stand for but two elements: ‘a misrepresented state of facts and a true state of facts.’ ” Id. (quoting Springfield Terminal, 14 F.3d at 654). “[I]n order to invoke the jurisdictional bar, a defendant must show ‘that the transaction ... [is] one in which a set of misrepresented facts has been submitted to the government.’ ” Id. at 1016–17 (first alteration in original) (quoting United States ex rel. Dunleavy v. Cty. of Del., 123 F.3d 734, 741 (3d Cir.1997), abrogated on other grounds by Graham Cty., 559 U.S. 280, 130 S.Ct. 1396 (2010)).

Applying the foregoing principles here, we find no allegation of fraud because the publicly disclosed information does not contain “an explicit accusation of wrongdoing.” Zizic, 728 F.3d at 236. Although the public reports described delays and incompetence, none explicitly asserted deception by Raytheon.

Whether the public reports described a “transaction” within the meaning of the public disclosure bar is less clear. Many of the public documents Raytheon supplied to the district court described management and engineering problems with the VIIRS project—thus presenting a “true” state of facts. A November 2005 GAO statement is the most specific public statement Raytheon identifies in this respect. It provides:

Problems involving multiple levels of management—including subcontractor, contractor, program office, and executive leadership—have played a role in bringing the NPOESS program to its current state. As noted earlier, VIIRS sensor development issues were attributed, in part, to the subcontractor's inadequate project management. Specifically, after a series of technical problems, internal review teams sent by the prime contractor and the program office found that the VIIRS subcontractor had deviated from a number of contract, management, and policy directives set out by the main office and that both management and process engineering were inadequate. Neither the contractor nor the program office recognized the underlying problems in time to fix them. After these issues were identified, the subcontractor's management team was replaced. Further, in January 2005, the NPOESS Executive Committee (Excom) called for an independent review of the VIIRS problems. This independent review, delivered in August 2005, reported that the program management office did not have the technical system engineering support it needed to effectively manage the contractor, among other things.

U.S. Gov't Accountability Off., GAO–06–249T, Polar–Orbiting Operational Envtl. Satellites: Tech. Problems, Cost Increases, & Schedule Delays Trigger Need for Difficult Trade-off Decisions 18 (Nov. 2005) (emphasis added).

While we are therefore satisfied that these reports describe a “true” state of facts, whether we can glean a “misrepresented” state of facts is a closer question. Reading between the lines of the publicly disclosed information, we possibly could infer that Raytheon falsely represented to the Government that all was well with VIIRS. Potentially supporting this inference is the fact that, even as the problems with VIIRS continued, the public statements suggest that Raytheon continued to be paid for its work. For example, an Office of the Inspector General report triggered by cost overruns noted that “VIIRS itself was 12 percent behind schedule and approximately 30 percent over budget. Nevertheless, the contractor received 92 percent of available award fees.” U.S. Dep't of Commerce, OIG–177794–6–0001, Nat'l Oceanic & Atmospheric Admin.: Poor Mgmt. Oversight & Ineffective Incentives Leave NPOESS Program Well Over Budget & Behind Schedule iii (May 2006). The Inspector General's report concluded that “[t]ime and money were ... wasted as the problems with NPOESS continued unchecked.” Id. Applying our X+Y=Z equation, arguably [Raytheon's contract deviations] + [the Government's continued payments to Raytheon] = the plausible inference that [Raytheon was continuing to bill the Federal Government and represent that all was well despite its failure to follow contract and policy directives]. On the other hand, the information in the public reports may come closer to suggesting breach of contract than fraud. Cf. Cafasso, United States ex rel. v. Gen. Dynamics C4 Sys., Inc., 637 F.3d 1047, 1057–58 (9th Cir.2011) (explaining that generally “breach of contract claims are not the same as fraudulent conduct claims, and the normal run of contractual disputes are not cognizable under the [FCA,]” and that “unsavory conduct is not, without more, actionable under the FCA.”) (alteration in original) (quoting United States ex rel. Wilson v. Kellogg Brown & Root, Inc., 525 F.3d 370, 383 (4th Cir.2008)). Ultimately, we believe it is arguable that a “transaction” was publicly disclosed, but we decline to definitively answer that question because doing so is unnecessary to the resolution of this case. Even assuming that the public reports disclosed a “transaction,” the public disclosure bar does not deprive the district court of jurisdiction—because, as we next explain, when we examine the public statements most likely to constitute a transaction and compare them to Mateski's Complaint, we find that the Complaint is not “based upon” those statements.

B.

Under our case law, for a relator's allegations to be “based upon” a prior public disclosure, “the publicly disclosed facts need not be identical with, but only substantially similar to, the relator's allegations.” United States ex rel. Meyer v. Horizon Health Corp., 565 F.3d 1195, 1199 (9th Cir.2009), overruled on other grounds by Kinetic Concepts, 792 F.3d at 1128 n. 6; see also Malhotra, 770 F.3d at 858 (“[T]he phrase ‘based upon’ in § 3730(e)(4)(A) means ‘substantially similar to,’ not ‘derived from.’ ”) (quoting Meyer, 565 F.3d at 1199).

1.

Although “substantially similar” is a phrase that appears frequently in our FCA decisions, we have never had occasion to articulate an approach to evaluating whether two sets of allegations are similar enough to qualify as “substantially similar” under the public disclosure bar. Our prior cases fall at the far ends of the similarity spectrum—on one end are cases finding substantial similarity because the allegations in the qui tam complaint were virtually identical to prior public disclosures, and on the other end are cases finding no similarity because the allegations in the qui tam complaint were completely different from prior disclosures. Mateski's case falls between these two extremes. If considered at a high level of generality, Mateski's Complaint and the public reports both discuss problems with VIIRS. If considered at a more granular level, the allegations in Mateski's Complaint discuss specific issues found nowhere in the publicly disclosed information. This case therefore requires us to at least partially fill the gap left by our prior decisions.

United States v. Alcan Electrical & Engineering, Inc., 197 F.3d 1014 (9th Cir.1999), is an example of a case on the virtually-identical end of the spectrum. In Alcan, we explained that an FCA complaint was substantially similar to an earlier publicly filed complaint because there was no difference in the substance of the fraud alleged in the two. The only difference was that the earlier complaint described the defendant in a fairly identifiable way but without actually using the defendant's name, whereas the later complaint added the defendant's name specifically. Id. at 1018–19.

On the other end of the spectrum, we have found no substantial similarity when the prior public statements disclosed nothing about the fraud alleged in the later qui tam complaint. In Foundation Aiding, for example, we concluded that an earlier complaint alleged “a very different problem—asbestos contamination—than the one alleged” in the relator's complaint about receiving Medicaid and Medicare payments for care not actually provided. 265 F.3d at 1013, 1016. Thus, “the allegations contained [in the earlier complaint] completely failed to disclose anything remotely similar to the fraud alleged” in the complaint at issue. Id. at 1016.

In our prior cases, we sometimes have asked whether the Government was on notice to investigate the fraud before the relator filed his complaint—which is another way of thinking about substantial similarity. But our cases discussing notice to the Government also fall at the same far ends of the similarity spectrum, so they likewise do not clarify the level of similarity required. In Alcan, for example, we noted that the Tenth Circuit in United States ex rel. Fine v. Sandia Corp., 70 F.3d 568 (10th Cir.1995), had found that the “the prior public disclosures contained enough information to enable the government to pursue an investigation against [the defendant].” 197 F.3d at 1019. We explained that “the instant case is similar to Sandia, in that the government, as regulator and owner, presumably would have ready access to documents identifying [the wrongdoers].” Id. We then concluded that “[t]his ready access makes it highly likely that the government could easily identify the [wrongdoers] at issue.” Id. In contrast, in Foundation Aiding, we observed that “it is impossible to say that the evidence and information in the possession of the United States at the time the False Claims Act suit was brought was sufficient to enable it adequately to investigate the case and to make a decision whether to prosecute.” 265 F.3d at 1016 (alteration omitted) (quoting United States ex rel. Joseph v. Cannon, 642 F.2d 1373, 1377 (D.C.Cir.1981)).

Because the allegations in Alcan were virtually identical to prior public statements, and the allegations in Foundation Aiding were completely different from prior public statements, these conclusions were inevitable. These cases' alternative articulation of the “substantially similar” inquiry—asking whether the Government was on notice—therefore leaves the same gap identified above.

Raytheon has pointed to nothing in this circuit's case law to help guide us in filling that gap. Raytheon relies upon Wang v. FMC Corp., 975 F.2d 1412 (9th Cir.1992), overruled on other grounds by Kinetic

Concepts, 792 F.3d 1121, to support its argument that the public disclosures were “substantially similar” to Mateski's allegations. Wang, however, is inapposite because we never addressed whether Wang's complaint was substantially similar to allegations or transactions that were previously disclosed. Rather, in resolving the case based upon whether Wang qualified as an original source, we specifically explained that “the necessary premise of the [district] court's ruling is that Wang's allegation of fraud [about a military vehicle] had been publicly disclosed before Wang brought his suit. Wang has never disputed, and his arguments appear to accept, that his allegation had been publicly disclosed.” Id. at 1417 (emphasis added). We touched briefly on the similarity of the prior disclosure but did so while emphasizing that the issue was not contested:

It is true that Wang's allegation about [a military vehicle] is supported by a few factual assertions never before publicly disclosed; but “fairly characterized” the allegation repeats what the public already knows: that serious problems existed with the [vehicle's] transmission. The district court characterized Wang's allegation and most of his information as a rehash of what already had been publicly disclosed. Wang does not dispute this characterization, and it finds support in the record.

Id. (emphasis added) (citation omitted) (quoting United States ex rel. Dick v. Long Island Lighting Co., 912 F.2d 13, 14 (2d Cir.1990)).

Mateski's case falls between the poles created by our prior precedent, because whether his Complaint is substantially similar to prior public reports depends on the level of generality at which the comparison is made. This case therefore requires us to address for the first time whether we should approach the substantial similarity question at a high or low level of generality, and accordingly whether a complaint that is similar only at a high level of generality triggers the public disclosure bar.

2.

The Seventh Circuit appears to be the only circuit to have focused on this level-of-generality question, developing its response over the course of three key cases. In United States ex rel. Baltazar v. Warden, 635 F.3d 866 (7th Cir.2011), the relator, a chiropractor, had filed suit alleging that her former employer had submitted fraudulent bills to the Medicare and Medicaid programs by adding services that had not been performed and by “upcoding.” Id. at 866–67. Applying the public disclosure bar, the district court dismissed the complaint based on the existence of an earlier public report that alleged that “57% of chiropractors' claims (in a sample of 400) were for services not covered by the Medicare program, and another 16% were for covered services that had been miscoded.” Id. at 867. The Seventh Circuit reversed, holding that the district court had erred when it concluded that the report “establishe[d] such prevalent fraud” “that it [is] unnecessary to give private relators a piece of the action in order to locate wrongdoers.” Id. The court explained that the relator's suit “supplied vital facts that were not in the public domain” because it alleged “that [the defendant] not only was submitting false claims but also was submitting them knowing them to be false, and thus was committing fraud.” Id. at 869. The court concluded that “[the relator's] suit is ‘based on’ those defendant-specific facts, not on the public information that false or mistaken claims are common.” Id. (emphasis added).

A year later, in United States ex rel. Goldberg v. Rush University Medical Center, 680 F.3d 933 (7th Cir.2012), the Seventh Circuit again reversed a district court's dismissal pursuant to the public disclosure bar, finding no substantial similarity because the earlier public disclosure failed to identify the precise type of hospital-billing deceit alleged in the later qui tam complaint. There, a GAO report and several public audits had indicated that teaching hospitals were often receiving double compensation for procedures performed by residents. The Seventh Circuit explained that Medicare pays teaching hospitals for work by residents on a fee-for-service basis only when a teaching physician supervises the residents. Id. at 933–34. “Technically [, those] payments are for the services rendered by the teacher,” and the cost of educating residents is reimbursed through government grants instead of through payments for specific services. Id. Nevertheless, many teaching hospitals, according to the GAO report, were billing on a fee-for-service basis for unsupervised services that the residents performed on their own, and then also receiving the teaching grants. Id. at 934.

The Goldberg relators' complaint alleged a somewhat different billing issue—that a particular university had allowed teaching physicians to supervise multiple operations simultaneously and then had sought Medicare reimbursement for all of the procedures by certifying that they had all been supervised. Id. at 935. The relators alleged fraud on the basis that regulations permit Medicare reimbursement only when the teaching physician is “present during all critical portions of the procedure and immediately available to furnish services during the entire service or procedure,” and that the double booking of the teaching physicians at this university meant that they were not “present” and “available” as the regulations required. Id. (quoting 42 C.F.R. § 415.172(a)(1)).

The Seventh Circuit concluded in Goldberg that “[relators] allege a kind of deceit that the GAO report does not attribute to any teaching hospital,” and that the public disclosure bar therefore did not apply. Id. at 936. Importantly, the court explained, “Unless we understand the ‘unsupervised services' conclusion of the GAO report ... at the highest level of generality—as covering all ways that supervision could be missing or inadequate—the allegations of these relators are not ‘substantially similar.’ ” Id. Referencing Baltazar, the court concluded that “boosting the level of generality in order to wipe out qui tam suits that rest on genuinely new and material information is not sound.” Id.

Most recently, in Leveski v. ITT Educational Services, Inc., 719 F.3d 818 (7th Cir.2013), the Seventh Circuit examined whether a qui tam complaint was substantially similar to a prior complaint that had been publicly filed. “To be sure,” the court explained, “Leveski's case looks similar to the [earlier] Graves case at first blush. The relators in both cases are former employees of ITT—and even held the same job title. The relators in both cases also allege that ITT violated the incentive compensation provision of the [Higher Education Act].” Id. at 832. That, however, was where the similarities ended. Id. The court explained that “[t]he details of how ITT allegedly violated the [Act] are quite different in Leveski's case than they were in Graves. Unlike the Graves relators, who alleged a more rudimentary scheme by ITT to violate the [Act's] incentive compensation provision, Leveski alleges a more sophisticated, second-generation method of violating the [Act].” Id. Addressing the question of “whether Leveski's allegations are different enough from the Graves allegations to bring her suit outside the public disclosure bar,” the Seventh Circuit held:

A review of our recent case law leads us to the conclusion that they are different enough. Indeed, Leveski's allegations against ITT are only similar to the Graves allegations when viewed at the highest level of generality. But in the last few years, we have indicated on more than one occasion that viewing FCA claims “at the highest level of generality ... in order to wipe out qui tam suits that rest on genuinely new and material information is not sound.”

Id. at 831 (alteration in original) (emphasis added) (quoting Goldberg, 680 F.3d at 936).

We find the reasoning of these cases persuasive, and we believe that the Seventh Circuit's approach effectuates the purpose of the public disclosure bar by “strik[ing] a balance between encouraging private persons to root out fraud and stifling parasitic lawsuits.” Schindler Elevator Corp. v. United States ex rel. Kirk, 563 U.S. 401, 413, 131 S.Ct. 1885, 179 L.Ed.2d 825 (2011) (quoting Graham Cty., 559 U.S. at 295, 130 S.Ct. 1396). Allowing a public document describing “problems”—or even some generalized fraud in a massive project or across a swath of an industry—to bar all FCA suits identifying specific instances of fraud in that project or industry would deprive the Government of information that could lead to recovery of misspent Government funds and prevention of further fraud. We believe that adopting the Seventh Circuit's approach therefore brings us closer to “the golden mean between adequate incentives for whistle-blowing insiders with genuinely valuable information and discouragement of opportunistic plaintiffs who have no significant information to contribute of their own.” Graham Cty., 559 U.S. at 294, 130 S.Ct. 1396 (quoting Springfield Terminal, 14 F.3d at 649).

3.

Raytheon argues that a subset of qui tam cases, which employ the phrase “quick trigger” to describe the substantial similarity inquiry, counsel against our adoption of the Seventh Circuit's approach. Raytheon's argument fails.

Specifically, Raytheon points to our footnote in Hagood v. Sonoma County Water Agency, 81 F.3d 1465 (9th Cir.1996), stating that “[t]he original source test is often treated as the focus of the jurisdictional inquiry, with courts treating the ‘based upon public disclosure’ step as a ‘quick trigger to get to the more exacting original source inquiry.’ ” Id. at 1476 n. 18 (quoting Cooper v. Blue Cross & Blue Shield of Fla., Inc., 19 F.3d 562, 568 n. 10 (11th Cir.1994)). Raytheon suggests that this means the public disclosure bar should turn on the “original source” inquiry, and that “based upon” should be only a superficial inquiry at a high level of generality. Our decision in Hagood did not, however, rely upon (or even analyze) this “quick trigger” notion because the footnote followed our conclusion that the prior public “filings may be enough to constitute public disclosure of the” fraud alleged in the qui tam complaint at issue. Id. at 1475. Rather than analyze substantial similarity, we skipped directly to finding that Hagood clearly qualified as an original source. As such, Hagood could avoid the public disclosure bar even if the allegations were substantially similar. The “quick trigger” language Raytheon emphasizes was therefore not an operative concept in Hagood at all, let alone one that requires viewing complaints at only a high level of generality. Raytheon emphasizes that other circuits also have incorporated the phrase “quick trigger,” but the cases that include this language have not used it to avoid a full substantial similarity analysis. In United States ex rel. Boothe v. Sun Healthcare Group, Inc., 496 F.3d 1169 (10th Cir.2007), for example, despite including the quick trigger language, the court conducted “[a] side-by-side comparison of the first three allegations of Ms. Boothe's complaint with those contained in prior qui tam actions,” to find that “the fraudulent schemes alleged are materially identical.” Id. at 1174. In United States ex rel. Osheroff v. Humana Inc., 776 F.3d 805 (11th Cir.2015), the Eleventh Circuit also recently quoted the “quick trigger” language, stating, “We have described the [based upon] test as a ‘[ ] quick trigger to get to the more exacting original source inquiry.’ ” Id. at 814 (quoting Cooper, 19 F.3d at 568 n. 10). Osheroff ultimately held, however, that the “significant overlap between [relator's] allegations and the public disclosures [wa]s sufficient to show that the disclosed information form[ed] the basis of this lawsuit and [wa]s substantially similar to the allegations in the complaint.” Id. Whatever work the “quick trigger” characterization may have done in Osheroff, nothing in the court's analysis suggested that the “quick trigger” concept overrides the need to analyze with some specificity the similarity between the allegations in a qui tam complaint and prior public disclosures. Raytheon's reliance on this language is thus to no avail.

IV.

Heeding the Seventh Circuit's warning against reading qui tam complaints at only the “highest level of generality,” Leveski v. ITT Educ. Servs., Inc., 719 F.3d 818, 831 (7th Cir.2013), we now reverse the district court's dismissal of this case because Mateski's Complaint alleges fraud that is different in kind and in degree from the previously disclosed information about VIIRS. See Hagood v. Sonoma Cty. Water Agency, 81 F.3d 1465, 1475 (9th Cir.1996) (examining whether, “fairly characterized,” the allegations in a relator's complaint “repeat[ ] what the public already knows”) (quoting Wang v. FMC Corp., 975 F.2d 1412, 1417 (9th Cir.1992), overruled on other grounds by United States ex rel. Hartpence v. Kinetic Concepts, Inc., 792 F.3d 1121 (9th Cir.2015) (en banc)). Although prior public reports had described general problems with Raytheon's work on VIIRS, none provided specific examples or the level of detail offered by Mateski.

A few examples from Mateski's lengthy Complaint suffice to demonstrate that his allegations are vastly more precise than the prior public reports about the problems with VIIRS. For instance, Mateski alleges numerous particular false waivers of VIIRS specifications and requirements. He also describes false and inappropriate signoffs and certifications in violation of the Program Quality Requirements, including “obvious forged signoffs” by Raytheon VIIRS operators. Mateski further details Raytheon's alleged substitution of “reduced Special Test Requirements ... in lieu of specified testing,” which he claims “compromise[d] the NPOESS/VIIRS Unit/System integrity and mission assurance.”

With respect to materials used in the VIIRS project, Mateski alleges the “use of Prohibited Materials (pure Tin), use of Prohibited Metallic materials known to cause corrosion ... when used together, use of Debris shedding locking fasteners (locking Heli–Coils), [and] use of Prohibited Materials and processes selected (Electro-deposited Nickel plating).” Mateski draws particular attention to problems with the J7 Power Connector, which he claims “[wa]s wired with forbidden (‘D & E’) materials of pure Tin plated wire.” He further alleges that “Raytheon ... falsely stated ... that the pure Tin plated wire would be acceptable for flight use despite the failure to pot the J7 Power Connector.”

Mateski also alleges numerous problems related to electrostatic discharge (“ESD”), asserting, for example, that Raytheon failed to maintain ESD protection of VIIRS flight hardware; and that certain cables were constructed using “hot plastics,” which are “ESD unapproved materials ... capable of building and storing excessive electrical charges.”

In contrast to these specific allegations, the prior public reports presented by Raytheon merely allege general problems involving mismanagement, technical difficulties, and noncompliance with contract and policy directives. Indeed, Raytheon's own description of how these disclosures characterized the “problems” in general terms—“numerous, major, profound, serious, severe, significant, systemic, worsening, costing billions of dollars, and resulting in decreased functionality”—supports reading the prior disclosures as revealing only very generalized problems with VIIRS.

Even if, as Raytheon argues, the prior public reports provided “enough information to ... pursue an investigation” into some fraud, and even though the Government did in fact undertake some investigation of VIIRS, the prior reports could not have alerted the Government to the specific areas of fraud alleged by Mateski. The practical consequence of adopting the Seventh Circuit's approach to defining substantial similarity is to allow relators who provide the Government with genuinely new and material information of fraud to move forward with their qui tam suits. Mateski is such a relator.

V.

Finally, Raytheon suggests that even if Mateski's Complaint contains some new allegations, the public disclosure bar applies because the Complaint is at least “partly based upon” the prior public reports about VIIRS. We disagree.

It is true that numerous cases from other circuits have indicated that, if a qui tam action is even “partly based upon” publicly disclosed allegations or transactions, it is “based upon' ” those allegations or transactions for purposes of the public disclosure bar. See, e.g., United States ex rel. Heath v. Wis. Bell, Inc., 760 F.3d 688, 691 (7th Cir.2014); Glaser v. Wound Care Consultants, Inc., 570 F.3d 907, 920–21 (7th Cir.2009); United States ex rel. Poteet v. Medtronic, Inc., 552 F.3d 503, 514 (6th Cir.2009).

Even were we to adopt this “partly based upon” concept, it would not help Raytheon. To the extent these cases utilize this “partly based upon” concept to reach their holdings at all, we understand them to still require that some of the relator's allegations be substantially similar to prior public disclosures. In Poteet, for example, the Sixth Circuit explained that “[d]espite the presence of one major allegation that was not made in the [earlier] complaint ... the primary focus of Poteet's complaint is the same ... illegal kickback scheme [that was] described in [the earlier] complaint.” 552 F.3d at 514. There was no question in Poteet that some of the allegations in the complaint that the Sixth Circuit held were precluded by the public disclosure bar were substantially similar to those already disclosed. 552 F.3d at 515 n. 7 (“[A]t least some parts of Poteet's complaint are clearly ‘based upon’ the [earlier] complaint.”).

Likewise, in Glaser, the Seventh Circuit held that it was “true that Glaser's complaint add[ed] a few allegations” not previously public, but that was not enough to avoid the public disclosure bar “because the allegations in Glaser's complaint (or most of them ) were substantially similar to publicly disclosed allegations.” 570 F.3d at 920–21 (emphasis added).

Because, as we have explained above, none of Mateski's allegations are “substantially similar” to the prior public reports when viewed at the appropriate level of generality, the “partly based upon” cases are of no assistance to Raytheon.

VI.

Mateski's allegations differ in both degree and kind from the very general previously disclosed information about problems with VIIRS. As such, if his allegations prove to be true, Mateski will undoubtedly have been one of those “whistle-blowing insiders with genuinely valuable information,” rather than an “opportunistic plaintiff[ ] who ha[s] no significant information to contribute.” Graham Cty. Soil & Water Conservation Dist. v. United States ex rel. Wilson, 559 U.S. 280, 294, 130 S.Ct. 1396, 176 L.Ed.2d 225 (2010) (quoting United States ex rel. Springfield Terminal Ry. Co. v. Quinn, 14 F.3d 645, 649 (D.C.Cir.1994)).

For the foregoing reasons, we REVERSE and REMAND.