JOHN M. GERRARD, District Judge.
This matter is before the Court on the parties' post-trial motions. DatabaseUSA.com and Vinod Gupta (collectively, DatabaseUSA) move for renewed judgment as a matter of law (filing 507) and alternatively, for a new trial (filing 510). Both parties have also filed motions to amend the judgment (filing 473, filing 498, filing 510), and Infogroup has moved for attorney's fees (filing 495). The Court will grant those motions in part, and deny those motions in part as set forth below.
The seeds of this litigation were planted nearly five years ago when Infogroup found nine of its own seeds in a sample of DatabaseUSA's database. Since then, there have been numerous substantive disputes, various written and oral decisions by this Court, and a seven-day jury trial where the parties zealously presented their best evidence and arguments to a jury of their peers. After careful deliberation, the jury returned a verdict in favor of Infogroup on all seven of its claims: (1) False Advertising; (2) Copyright Infringement; (3) Mark Infringement; (4) Unfair Competition; (5) Breach of the 2008 Separation Agreement; (6) Breach of the 2012 Settlement Agreement; and (7) Unjust Enrichment. See generally filing 464. The jury awarded Infogroup $53,600,000.00 in damages. Filing 464.
Now, DatabaseUSA argues that the jury's verdict cannot stand as a matter of law. Filing 508 at 4. Alternatively, DatabaseUSA moves for a new trial arguing various errors by the Court—including instructional errors and improper admissions. See filing 511 at 2. Infogroup opposes both motions, arguing that the Court's instructions were correct and that the jury's verdict is sound. See filing 518 at 6; filing 520 at 1.
When considering a motion for judgment as a matter of law, a court must determine whether or not the evidence was sufficient to create an issue of fact for the jury. Lane v. Chowning, 610 F.2d 1385, 1388 (8th Cir. 1979). The Court will grant a motion for judgment as a matter of law when all the evidence points one way and is susceptible of no reasonable inferences sustaining the position of the nonmoving party. Ehrhardt v. Penn. Mut. Life Ins. Co., 21 F.3d 266, 269 (8th Cir. 1994). In considering the motion, the Court views the record in the light most favorable to the prevailing party. Wash Solutions, Inc. v. PDQ Mfg., Inc., 395 F.3d 888, 892 (8th Cir. 2005). The Court must also assume that all conflicts in the evidence were resolved in favor of the prevailing party, and the Court must assume as proved all facts that the prevailing party's evidence tended to prove. E.E.O.C. v. Kohler Co., 335 F.3d 766, 772 (8th Cir. 2003). The motion should be denied unless the Court concludes that no reasonable juror could have returned a verdict for the nonmoving party. Billingsley v. City of Omaha, 277 F.3d 990, 995 (8th Cir. 2002).
A motion for new trial is governed by Federal Rule of Civil Procedure 59.
The standard for granting a new trial is whether the verdict is against the great weight of the evidence. Butler v. French, 83 F.3d 942, 944 (8th Cir. 1996). In evaluating a motion for a new trial pursuant to Rule 59(a), the key question is whether a new trial should have been granted to avoid a miscarriage of justice. McKnight By & Through Ludwig v. Johnson Controls, Inc., 36 F.3d 1396, 1400 (8th Cir. 1994).
Before reaching the merits of the parties' arguments, the Court must briefly address three underlying contentions that form the foundation for DatabaseUSA's post-trial briefing: (1) copyright preemption; (2) exclusion of testimony; and (3) the spoliation instruction given by the Court. See filing 508 at 29-30; see also filing 511 at 14-15; 21-25. The Court will consider those arguments in turn below.
According to DatabaseUSA, two of Infogroup's state law claims—unfair competition and unjust enrichment—are preempted by Infogroup's copyright claim. That argument is based on a core principle of federal copyright law: that the Copyright Act preempts "all legal or equitable rights that are equivalent to any of the exclusive rights within the general scope of copyright . . . in works of authorship that are fixed in a tangible medium of expression and come within the subject matter of copyright." 17 U.S.C. § 301(a); Dryer v. Nat'l Football League, 814 F.3d 938, 942 (8th Cir. 2016).
In determining whether federal copyright law preempts a cause of action under state law, the Court's analysis is twofold: (1) whether the work at issue is within the subject matter of copyright as defined in §§ 102 and 103 of the Copyright Act; and (2) whether the state-law-created right is equivalent to any of the exclusive rights within the general scope of copyright as specified in § 106. Id. But section 301 preempts only those state law rights that "may be abridged by an act which, in and of itself, would infringe one of the exclusive rights provided by federal copyright law." Nat'l Car Rental Sys., Inc. v. Computer Assocs. Int'l, Inc., 991 F.2d 426, 431 (8th Cir. 1993) (quoting Computer Assocs. Int'l, Inc. v. Altai, Inc., 982 F.2d 693, 716 (2d Cir. 1992)). So, if an extra element is required to constitute a state-created cause of action— instead of or in addition to the acts of reproduction, performance, distribution or display protected by the Copyright Act—then the claim does not lie within the general scope of copyright and there is no preemption. Id. (citing 1 Nimmer on Copyright § 1.01[B], at 1-14-15)). Stated differently, the state law claims must be qualitatively different from Infogroup's copyright claim. See generally id.
To establish a claim of unfair competition, Infogroup is required to demonstrate that DatabaseUSA attempted to pass off its goods or services (i.e., its database) as Infogroup's product. See Restatement (Third) of Unfair Competition, §§ 2-4 (1995); John Markel Ford, Inc. v. Auto-Owners Ins. Co., 543 N.W.2d 173, 178 (Neb. 1996) (using the Restatement). It is wellestablished that a state law unfair competition claim that alleges the tort of passing off is not preempted because such a claim alleges an extra element of deception or misrepresentation that is not necessary for copyright infringement. Donald Frederick Evans & Assocs., Inc. v. Cont'l Homes, Inc., 785 F.2d 897, 914 (11th Cir. 1986); Warner Bros. Inc. v. Am. Broad. Cos., Inc., 720 F.2d 231, 247 (2nd Cir. 1983); Nicassio v. Viacom Int'l, Inc., 309 F.Supp.3d 381, 397 (W.D. Pa. 2018); Kitchen & Bath Concepts of Pittsburgh, LLC v. Eddy Homes, Inc., 2016 WL 7404559, *5 (W.D. Pa. 2016); Kindergartners Count, Inc. v. Demoulin, 171 F.Supp.2d 1183, 1191 (D. Kan. 2001); cf. Take-Two Interactive Software, Inc. v. Zipperer, No. 18-CV-2608, 2018 WL 4347796, at *7 (S.D.N.Y. Aug. 16, 2018).
For similar reasons, Infogroup's unjust enrichment claim is not preempted by the Copyright Act. To prove an unjust enrichment claim, Infogroup must show that DatabaseUSA obtained and benefitted from something of value that it was not entitled to. Kalkowski v. Neb. Nat'l Trails Museum Found., Inc., 862 N.W.2d 294, 301-02 (Neb. 2015) (emphasis added) (quoting Restatement (Third) of Restitution and Unjust Enrichment § 2, comment a. (2011)). That "something of value," the jury could determine, was obtained through DatabaseUSA's deception or misrepresentation. And as discussed above, the element of deception or misrepresentation is not necessary for a copyright cause of action. Accordingly, the Court concludes that neither Infogroup's unfair competition or unjust enrichment claim is preempted by its copyright claim—and as such, DatabaseUSA's Rule 50 motion will be denied on those grounds. See filing 508 at 29-31.
DatabaseUSA also claims the testimony of two witnesses should have been excluded. See filing 511 at 14. Specifically, DatabaseUSA claims that John Hofmann and Amit Khanna were allowed to testify over several foundational objections on topics upon which they were not qualified to render an opinion. See Filing 511 at 14.
First, DatabaseUSA claims that John Hofmann's testimony was erroneously admitted. In support of that contention, DatabaseUSA makes two arguments: (1) DatabaseUSA claims that Hofmann's damage model was "unreliable, illogical, inconsistent with the evidence, and otherwise flawed;" and (2) DatabaseUSA claims that because there was no way for the jury to disaggregate what portion of damages was attributable to what claims, his testimony was inherently speculative. Filing 511 at 14.
With respect to DatabaseUSA's latter contention, DatabaseUSA's specifically claims that because the jury had no way to "disaggregate damages" between Infogroup's copyright claim and the remaining causes of action, Hofmann's testimony should have been excluded. See filing 511 at 15-18. This argument is based, in large part, on various antitrust decisions. See Concord Boat Corp. v. Brunswick Corp., 207 F.3d 1039, 1057 (8th Cir. 2000); Amerinet, Inc. v. Xerox Corp., 972 F.2d 1483, 1494 (8th Cir. 1992); Farley Transp. Co. v. Santa Fe Trail Transp. Co., 786 F.2d 1342, 1352 (9th Cir. 1985). Those decisions, very generally, require a damage expert to separate lawful from unlawful conduct to ensure the expert's testimony is not speculative or based on conjecture. Concord Boat Corp., 207 F.3d at 1057 (emphasis supplied).
But in those cases, the problem is not, as DatabaseUSA suggests, that the jury was unable to parse out the amount of damage associated with each allegation of unlawful conduct. Rather, the problem is that there was "no evidence on the amount of damages attributable only to the unlawful conduct." Farley Transp. Co., 786 F.2d at 1352 (9th Cir. 1985) (emphasis in original); see also Amerinet, Inc., 972 F.2d at 1494. Stated differently, the critical flaw in the antitrust context was the lack of evidence linking the defendant's profits to the customer's own motivations (i.e., a lower price point) and the unlawful conduct of the defendant (i.e., anti-trust violations). Id. That is, there was nothing to suggest what portions of the claimed profits were attributable to unlawful conduct, and what portions of the claimed profits are attributable to lawful competition.
Here, however, Hofmann's testimony specifically separated $39.6 million of DatabaseUSA's revenue attributable to unlawful conduct—as a result of DatabaseUSA's copyright infringement, false advertising, and other misconduct—from its overall revenue of $46.6 million. Filing 479 at 37; E114. In other words, according to Hofmann, at least $7 million of DatabaseUSA's overall profits were not linked to any of DatabaseUSA's unlawful conduct, but rather, profits it obtained fair and square.
Stated more simply, DatabaseUSA's perceived issues with Hofmann's testimony are not whether the jury could determine the amount of damages attributable to DatabaseUSA's unlawful conduct. Rather, the problem is precisely what unlawful conduct (i.e., copyright infringement, false advertising, mark infringement, or some combination of that conduct) caused a specific portion of that harm. And although that might prove to be a causation issue on some of Infogroup's claims, as discussed in more detail below, that does not mean that the jury had no basis to make a reasonable and principled estimate of the amount of damage attributable to DatabaseUSA's unlawful conduct. Cf. Farley, 786 F.2d at 1352; Amerinet, Inc., 972 F.2d at 1494. Thus, Hofmann's testimony was properly admitted on those grounds.
DatabaseUSA's alternative contention, that Hofmann's damages model is flawed in various respects, does not fare any better. Specifically, DatabaseUSA takes issue with several aspects of Hofmann's testimony, including his lack of in-depth knowledge about how the "overlap" factor was calculated, the fact that a high match rate is not "statistically significant", and the fact that Hofmann failed to account for several other factors in addition to the overlap factor he used to form his overall opinion. Filing 511 at 11.
But those arguments go to weight, not admissibility. Bonner v. ISP Tech., Inc., 259 F.3d 924, 929-30 (8th Cir. 2001). Only when an expert's opinion "is so fundamentally unsupported that it can offer no assistance to the jury must such testimony be excluded." Id. That is not the case here. In fact, there is little doubt that the specialized knowledge of Hofmann—someone with over twenty years of experience in the finance industry and the sole damages expert presented by either party—would be helpful in determining the portion of DatabaseUSA's profits attributable to its unlawful conduct.
That is not to say that DatabaseUSA's attacks on Hofmann's testimony do not make some valid points, but those points were presented to the jury through a rigorous cross examination—and the jury disagreed.
Next, DatabaseUSA claims that two separate, but related, areas of the testimony of Amit Khanna—president of Local Marketing Solutions at Infogroup—were not admissible. Filing 511 at 18. First, DatabaseUSA claims there was not sufficient foundation for Khanna to opine on what Infogroup's 2011 database looked like or how it was compiled. Filing 511 at 19. Second, DatabaseUSA claims that Khanna did not possess the requisite personal knowledge to testify about the 2011 seeding. Filing 511 at 19.
In support of its former contention, DatabaseUSA points out that Khanna did not work at Infogroup at the time the 2011 database was in play. Filing 511 at 19. But that argument ignores the fact that Khanna also testified that he began working at Infogroup as a developer in 1997. Filing 476 at 249. And in that role, Khanna had significant involvement in the technical aspects of Infogroup's database. See filing 476 at 178-83. Specifically, Khanna testified that he "headed up [Infogroup's] database compilation group which was obviously responsible for compiling information and collecting all the — the data from all the sources." Filing 476 at 181. And around 2006, Khanna was involved in managing data compilation in accordance with technological advancements. Filing 476 at 181-2.
So, although it is true that Khanna was not involved in compiling the 2011 database itself, that does not mean he has no knowledge of how Infogroup generally gathered, selected, and arranged that information. And there is no evidence to suggest that the way Infogroup compiled its database meaningfully changed from 2006 until 2011. In other words, Khanna, who has over fifteen years of experience in data compilation techniques and database maintenance, demonstrated the requisite foundation to opine on how Infogroup generally arranged and compiled its 2011 database.
And more fundamentally, Khanna specifically testified that following his return to Infogroup in 2013, he reviewed and analyzed the 2011 database for purposes of obtaining a valid copyright. Filing 476 at 241. In doing so, he identified and separated the portions of Infogroup's database that were original and deleted third-party work (i.e., non-copyrightable material). Filing 476 at 2421-2. And that necessarily required Khanna to review Infogroup's 2011 database, determine how its selection and arrangement process was original, and determine what portions should be excluded. So, contrary to DatabaseUSA's assertions, Khanna did have first-hand knowledge of what Infogroup's 2011 Database looked like, how it was compiled, and how that information was originally arranged.
DatabaseUSA's second argument fares no better. Specifically, DatabaseUSA claims that Khanna had "no basis to testify as to the creation of Infogroup's 2011 `seeds' or the results of Infogroup's data `audits.'" Filing 511 at 19. But those arguments miss the mark. Indeed, Khanna testified that he routinely works closely with the audit team because it is his responsibility to first identify issues before the audit team gets involved. Filing 476 at 199. And in that role, he reviewed the 2011 seeds with the team who did the audit, and used the 2011 seed list to conduct his own audit. Filing 476 at 203-04. The purpose of reviewing that information, Khanna said, was to be able to identify if any of those seeds were present in the portion of DatabaseUSA's database purchased by Infogroup. Filing 476 at 211. Thus, there was sufficient foundation for Khanna to testify to the 2011 seeds and results of the audit. And DatabaseUSA's Rule 59 motion will be denied on those grounds.
Finally, DatabaseUSA claims that the spoliation instruction was improper. Filing 511 at 21. That argument is based on DatabaseUSA's contention that it did not intentionally destroy its database. Filing 511. DatabaseUSA raises several arguments that, it says, supports that contention. To begin, DatabaseUSA argues its 2014 database (and all previous versions) was lost due to its document retention policy—rather than the result of deliberate destruction. Filing 511 at 22. And because the information was innocently lost, DatabaseUSA claims the "information was [not] lost with the intent to prevent its use in litigation" as required under Fed. R. Civ. P. 37.
But the Court has, at this point, considered that same argument on several occasions. See filing 343, filing 400. And the Court is still not convinced that DatabaseUSA's database was deleted by accident. At the time this litigation was commenced, DatabaseUSA had copies of its business database dating back to at least February 2012. Filing 343 at 5. And that was expected because, as Gupta testified, DatabaseUSA generally retains copies of its database for at least two years. Filing 287-1 at 160-1. Yet, by the time Infogroup began setting its plan to perform a complete audit of DatabaseUSA's databases in motion, only a 2015 version of the database (and subsequent versions of that database) existed. It is difficult to imagine a scenario where the database's destruction was a result of anything other than "the intent to prevent its use in litigation." Fed. R. Civ. P. 37 (advisory committee's note to 2015 amendment).
And even if the Court were persuaded that the missing database was "accidentally" destroyed, that still would not alleviate the underlying problems of its absence. That brings the Court to DatabaseUSA's remaining arguments—each of which are equally unavailing. First, DatabaseUSA claims that the Court's reliance on Lewy v. Remington Arms Co., Inc., 836 F.2d 1104 (8th Cir. 1988) is misguided. And second, DatabaseUSA claims that Infogroup was not harmed by the absence of its entire 2014 database.
DatabaseUSA's first argument is easily disposed of, so the Court will begin there. Specifically, DatabaseUSA insinuates that the Court plucked one "isolated" quote out of a "more nuanced" decision. Filing 511 at 22. And as a result of the Court's oversight, DatabaseUSA claims that the Court completely missed the underlying premise of Lewy—instructions to the lower court to "examine whether the defendant's document retention policy was instituted in bad faith." Filing 511 at 23. To that end, DatabaseUSA argues there is nothing in the record to suggest that DatabaseUSA's "routine deletion of data" was "instituted in bad faith," and as such, DatabaseUSA contends that the adverse instruction is improper under Lewy. Filing 511 at 23 (emphasis in original).
But that argument does exactly what DatabaseUSA claims the Court did—it plucks one "isolated quote" from a decision while ignoring the remainder. Had DatabaseUSA continued to refer the Court to related portions of the opinion, it would have cited the following:
Lewy, 836 F.2d at 1112 (citations omitted); see also Stevenson v. Union Pac. R. Co., 354 F.3d 739, 747 (8th Cir. 2004) (although retention policy was not unreasonable or instituted in bad faith, it was unreasonable and amounted to bad faith conduct for defendant to adhere to the principle in the circumstances of this case).
In other words, Lewy specifically provided guidance for a scenario where despite the reasonableness of the instituted policy, the destruction of documents gives rise to an unfavorable inference. And that scenario is nearly identical to this litigation: DatabaseUSA was sued in February 2014, but following the initiation of this lawsuit, it destroyed the database knowing it would be highly relevant to this particular litigation. Simply put, even if DatabaseUSA destroyed the database pursuant to its routine retention policy, the post-litigation destruction—and circumstances surrounding its destruction—create a sufficiently strong inference of an intent to destroy it for purpose of suppressing evidence. Stevenson, 354 F.3d at 748.
DatabaseUSA's second argument is peculiar at best. Indeed, that argument hinges on DatabaseUSA's contention that even without access to the 2014 database in its entirety, "Infogroup [] had sufficient access to DatabaseUSA's database to test its copyright claim." Filing 511 at 24. That is true, DatabaseUSA claims, because "there was absolutely nothing stopping Infogroup from using this data (or purchasing more) to conduct the necessary comparison." Filing 511 at 24. Yet, DatabaseUSA's entire argument—as discussed in more detail below—as to why, in its view, Infogroup's copyright claim fails as a matter of law, is premised on DatabaseUSA's contention that Infogroup was required to enter its entire database into evidence. Filing 508 at 7-8. That is required, DatabaseUSA argues, because to prove copyright infringement Infogroup must demonstrate "extensive verbatim copying" of the complete database. Filing 508 at 7-8. So, by DatabaseUSA's own logic, the Court is unsure how "Infogroup had sufficient access to DatabaseUSA's database to test its copyright claim" without the ability to forensically analyze the database. DatabaseUSA's Rule 50 motion will be denied on those grounds.
As a final matter, DatabaseUSA also argues that the adverse inference was too broad because it "should have been tailored to Infogroup's (single) copyright claim." Filing 511 at 25. But in making that argument, DatabaseUSA has failed to point the Court to any, much less persuasive, authority requiring the adverse inference to be attached to a particular claim. See filing 511 at 25-26. For instance, it's common in any case presenting multiple claims for relief for particular evidence to be relevant to some claims, but not others. It would, however, be uncommon to give a limiting instruction for such evidence, in the absence of any suggestion that the evidence could be misused—and DatabaseUSA is in effect saying that such a limiting instruction should have been given here. Instead, Infogroup was entitled to press the adverse inference with respect to any claim to which the 2014 database would have been relevant, had it not been destroyed. So, the Court concludes the instruction was proper under Eighth Circuit precedent, and the Court will deny DatabaseUSA's motions on those grounds. Stevenson, 354 F.3d at 750.
With those preliminary matters decided, the Court will address the substance of DatabaseUSA's Rule 50/59 motions. DatabaseUSA's Rule 50 motion generally alleges that there is no legally sufficient basis for the jury's verdict on the following categories of claims: (1) copyright infringement; (2) false advertising and mark infringement; (3) unjust enrichment and unfair competition; and (4) breach of the 2008 separation agreement. Filing 508 at 4. DatabaseUSA's Rule 59 motions similarly suggest that the jury's findings on each of those categories of claims is either excessive or against the clear weight of the evidence. See Filing 510 at 1. Because many of the arguments in support of those motions are based on similar issues of fact and law, the Court will generally consider those motions together.
DatabaseUSA's arguments underlying each of those claims generally fall into one of two categories—liability or damages. The Court will first consider whether the jury's finding that DatabaseUSA is liable to Infogroup on each of the above claims is supported by the evidence presented at trial. From there, the Court will determine if DatabaseUSA's arguments surrounding Infogroup's damages model have any merit.
The jury was asked to determine whether Infogroup owns a valid copyright in its 2011 database, and whether DatabaseUSA's 2014 database copied protected expression in Infogroup's copyrighted work. Taylor Corp. v. Four Seasons Greetings, LLC, 315 F.3d 1039, 1042 (8th Cir. 2003); Moore v. Columbia Pictures Indus., Inc., 972 F.2d 939, 941 (8th Cir. 1992); see filing 462 at 15. The jury determined that it did, and for the reasons set forth below, the Court finds that conclusion is supported by sufficient evidence. Filing 467.
The parameters of copyright law are clear—original works are copyrightable, but facts are not. Feist Publ'ns, Inc. v. Rural Tel. Serv., Co., 499 U.S. 340, 344-45, (1991); Experian Info. Sols., Inc. v. Nationwide Mktg. Servs. Inc., 893 F.3d 1176, 1185 (9th Cir. 2018); Schoolhouse, Inc. v. Anderson, 275 F.3d 726, 729-30 (8th Cir. 2002). At trial, the parties' arguments generally fell on opposite ends of that spectrum. That is, according to Infogroup, its database is not composed of facts—it is made of unique judgments. Filing 476 at 89; filing 477 at 171. And according to DatabaseUSA, Infogroup only sells facts— not unique or creative judgments. Filing 476 at 149. But in reality, this case lies somewhere in between: Infogroup sells a selection and arrangement of facts that may be subject to copyright protection if there is evidence that the information is selected, coordinated, or arranged in a creative way. 17 U.S.C. § 101; see also Feist, 499 U.S. at 357.
To begin, the kind of creativity that sufficiently establishes copyright protection in factual compilations is minimal. Feist, 499 U.S. at 348; Key Publ'ns, Inc. v. Chinatown Today Publ'g Enters., Inc., 945 F.2d 509, 512-13 (2d Cir. 1991); see Experian, 893 F.3d 1176, 1185. By way of example, the compilation of business names, addresses, and phone numbers of interest to the New York City Chinese-American community would be sufficiently creative to warrant copyright protection of a directory. Key Publ'ns, 945 F.2d at 512-13. And the compilation of data from a variety of sources, including catalogue purchase data, cable company records, real estate deeds, and warranty cards signed by consumers at retail stores is also sufficiently creative to warrant copyright protection. Experian, 893 F.3d at 1185.
But such compilations of factual information receive only limited protection. 17 U.S.C. § 103(b); Feist, 499 U.S. at 359. That means that a compiler may freely use the facts contained in a compilation when preparing a competing work, as long as the competing work does not exhibit the same selection or arrangement. See Feist, 499 U.S. at 349, see also Kregos, 937 F.2d at 702, 709. So, the question is whether there was sufficient evidence from which a jury could conclude that Infogroup's compilation of factual information was sufficiently creative to warrant copyright protection. Feist, 499 U.S. at 357.
According to DatabaseUSA, there was no way for the jury to determine whether the same selection or arrangement was used by both compilation companies. That is true, DatabaseUSA alleges, because Infogroup failed to introduce evidence of what its database actually looked like in November, 2011. See filing 508 at 7-13. And that oversight, DatabaseUSA argues, is not insignificant because the jury had no way to determine how, if at all, the database was creatively arranged. See filing 508 at 7. In support of that argument, DatabaseUSA extensively relies on the Ninth Circuit's decision in Experian. 893 F.3d at 1176.
But that argument is misplaced. The Experian court determined that the database in that case was entitled to copyright protection. Experian, 893 F.3d at 1176. More specifically, the court noted that "Experian's selection process in culling data from multiple sources and selecting the appropriate paring of addresses with names before entering them in the database involves a process of at least minimal creativity." Id. at 1185.
And that process is nearly identical to the way Mike Iaccarino and Amit Khanna described Infogroup's selection process. Iaccarino, Infogroup's CEO, testified to Infogroup's "merge/purge" process. Specifically, he explained,
Filing 475 at 60. In other words, according to Iaccarino, once Infogroup obtains factual information, that information is verified—either by making a phone call to the business or by comparing it to other information, and Infogroup's employees choose to include or exclude that information from the database. Filing 475 at 60; 745 at 61. To that end, Iaccarino also testified that:
Filing 475 at 61. Stated simply, Iaccarino provided the jurors with sufficient testimony from which they could assess that the process of determining what data, if anything, makes it into Infogroup's database involved some level of creativity.
And that conclusion is bolstered by Amit Khanna, who also testified that Infogroup's "database contains information that [Infogroup] compiled and decisions [Infogroup] made as to how to represent [information] in our database." Filing 477 at 170. In particular, Khanna stated that Infogroup's database was, in essence, its "interpretation of what information [they've] been able to gather and put in our database." Filing 477 at 170. Those interpretations and determinations necessarily involve some level of discretion, and in turn, some level of creativity. And so, contrary to DatabaseUSA's assertion, there was testimony explaining how Infogroup arranges, compiles, and verifies the information it gathers.
More fundamentally, though, the introduction of Infogroup's registered copyright created a rebuttable presumption of copyright validity. Four Seasons Greetings, 315 F.3d at 1042; E14. And because Infogroup introduced its certificate of registration which protects the "text, [and] Compilation" of its 2011 database, it is presumed that Infogroup's selection processes possess the creative attributes necessary for copyright protection.
And Infogroup's purported failure to introduce evidence of the actual 2011 database into evidence does not undermine the sufficiency of that evidence. Indeed, the failure to present all theoretically available evidence, or the best possible evidence, does not show that the evidence that was presented was legally insufficient. Nor does it, as DatabaseUSA seems to suggest, prohibit the jury from finding that DatabaseUSA's 2014 database (and all previous versions of that database) were identical copies of Infogroup's 2011 compilation and selection.
Again, DatabaseUSA relies on the Ninth Circuit's decision in Experian to support its contention that without introducing the copyrighted work or the allegedly infringing work, there can be no infringement. That is true, DatabaseUSA claims, because Experian required the jury to perform a sideby-side comparison of the copyrighted work and the allegedly infringing work. Experian, 893 F.3d at 1187. And without such comparison, DatabaseUSA claims, the jury had no way to determine whether DatabaseUSA's database was a bodily appropriation of Infogroup's database. Id.
But Experian is distinguishable for at least two reasons. First, DatabaseUSA intentionally destroyed its 2014 database and all previous versions following the commencement of litigation—a fact that is absent from the Experian decision. And that difference is not insignificant because it means the logic underlying the Ninth Circuit's insistence on the introduction of the entire database is no longer applicable. After all, the reason the court in Experian insisted that the entire database be entered into evidence was so the jury could perform a "side-by-side" comparison of the two works. Id. Here, however, even if Infogroup had introduced its copyrighted work, a "side-byside" comparison of the copyrighted work (i.e., Infogroup's 2011 database) and the allegedly infringing work (i.e., DatabaseUSA's 2014 database) would still not be possible. And there would be little, if any, value in giving the jury Infogroup's 2011 database when there was no infringing 2014 database for the jury to compare that database to.
Second, Experian did not, as DatabaseUSA seems to suggest, categorically foreclose the possibility of permitting a plaintiff to establish infringement on the basis of circumstantial evidence. See id. at 1187. Instead, the Experian court found that even assuming the pairings were exact copies, "the match rate would only be 80% and insufficient to establish bodily appropriation of [the copyrighted] work." Id. In reaching that decision, though, the Ninth Circuit relied on the Eighth Circuit's decision in Schoolhouse, 275 F.3d at 729-30. And Schoolhouse actually supports the jury's finding that the 2014 database circumstantially infringed on Infogroup's 2011 copyright.
In Schoolhouse, the plaintiff published a magazine with information concerning schools in the area. The defendant also compiled information concerning schools on his real estate website. Id. In that decision, the question was whether the defendant's use of the factual compilations demonstrated copyright infringement even though direct copying could not be proven. And in answering that question, the Eighth Circuit was clear—to establish copying by circumstantial evidence the analysis is twofold: (1) did DatabaseUSA have access to Infogroup's 2011 copyrighted work; and (2) is there substantial similarity—both in the ideas and expression of the copyrighted work and infringing work? Id. at 729; see also Hartman v. Hallmark Cards, Inc., 833 F.2d 117, 120 (8th Cir. 1987).
To establish access, Infogroup must show that DatabaseUSA had an "opportunity to view or to copy" its work. Moore v. Columbia Pictures Indus., Inc., 972 F.2d 939, 942 (8th Cir. 1992); 3 Nimmer on Copyright § 13.02[A]. Establishing a "bare possibility" of access is not enough; rather, Infogroup must prove that the DatabaseUSA had a "reasonable possibility" of viewing its work. Id. That is, there must be some evidence from which the jury could determine that DatabaseUSA or Gupta was able to view Infogroup's 2011 database.
The following testimony of Blake Van Gilder was adduced for the jury at trial:
Van Gilder Deposition at 65. That testimony, if believed by the jury, would support the inference that not only could DatabaseUSA access the database, but that it did. See Rottlund Co. v. Pinnacle Corp., 452 F.3d 726, 732 (8th Cir. 2006); see also Fisher-Price, Inc. v. Well-Made Toy Mfg. Corp., 25 F.3d 119, 123 (2nd Cir. 1994) (defendant's former employee testified that she heard the president say that he was sending plaintiff's dolls to a factory in China for copying and that the president bragged that the finished product was indistinguishable from the plaintiff's doll); Rogers v. Koons, 960 F.2d 301, 307 (2nd Cir. 1992) (defendant gave a copy of plaintiff's photograph to Italian artisans and instructed them to copy it); Koontz v. Jaffarian, 617 F.Supp. 1108, 1113-15 (E.D. Va. 1985), aff'd, 787 F.2d 906 (4th Cir. 1986) (defendant's former employee testified that defendant made a page by page display of the plaintiff's electrical estimating labor manual, making some modifications and including the same errors). There was conflicting evidence at trial with respect to DatabaseUSA's opportunity to view and copy Infogroup's work, but Van Gilder's testimony is plainly sufficient, if believed by the jury, to demonstrate that DatabaseUSA had access to Infogroup's 2011 database.
And access may also be inferred by proof of similarity which is so striking that the possibilities of independent creation, coincidence and prior common source are, as a practical matter, precluded. Selle v. Gibb, 741 F.2d 896, 901 (7th Cir. 1984). To establish substantial similarity in a factual compilation, there must be similarity, both in ideas and expression, between the original elements of Infogroup's database and DatabaseUSA's database. Schoolhouse, 275 F.3d at 729; Hartman v. Hallmark Cards, Inc., 833 F.2d 117, 120 (8th Cir. 1987).
Schoolhouse, 275 F.3d at 729.
There is legally sufficient evidence to support an inference of "essentially verbatim" copying with respect to Infogroup's 2014 database and its previous versions. Specifically, Khanna testified that Infogroup inserts "seed data" into its listings, containing fictitious combinations of name, address, and telephone number. Filing 476 at 200. This "seed data" allows Infogroup to monitor its information, and if its seeds appear in a competitor's list, Infogroup knows that the competitor has somehow gained access to Infogroup's information. Filing 476 at 201. Sometime in June 2013, Infogroup's audit team found nine of Infogroup's 2011 seeds in a sample of DatabaseUSA's database. Filing 476 at 203; 210; E108. That discovery resulted in further investigation—which also revealed that DatabaseUSA's database contained numerous entries with the exact same misspellings and typographical errors as Infogroup's entries. Filing 109; filing 476 at 223.
Based on this information, Infogroup performed a 2014 data audit by Infogroup comparing the overlap of information in its database with information in two of its competitors: (1) DatabaseUSA and (2) Dun & Bradstreet ("D&B"). Filing 476 at 225. Khanna testified that "the idea . . . this is comparing our data to two sample sets that we've acquired, one from obviously DatabaseUSA, one from our leading competitor, Dun & Bradstreet." Filing 476 at 226. And this comparison revealed that DatabaseUSA's match percentage to Infogroup's information was "95 percent and D&B's was 59.3" percent. Filing 476; E111. Specifically, according to Khanna,
Filing 476 at 233. Van Gilder also testified to the virtually identical nature of the two databases claiming that DatabaseUSA's database and Infogroup's database were "extremely similar." Van Gilder Deposition at 32-33. More specifically, Van Gilder testified that the two databases included identical, indepth, information such as email addresses, employee size, and sales volume. See Van Gilder Deposition at 33.
Based on that evidence, it is difficult to imagine why it would be unreasonable for the jury to conclude that the 2014 database is a verbatim copy of Infogroup's 2011 database. After all, a 95 percent match rate, the presence of in-depth, hard-to-come-by information, the finding of every fictitious "seed" entry in the data sample, and nearly identical typographical and spelling errors is sufficient to circumstantially demonstrate "extensive verbatim copying." Compare Experian, 893 F.3d 1176, 1187 (9th Cir. 2018) (finding an 80 percent match rate insufficient to demonstrate "bodily appropriation"); Ross, Brovins & Oehmke, P.C. v. Lexis Nexis Grp., 463 F.3d 478, 483 (6th Cir. 2006) (finding a 61 percent match rate insufficient to establish infringement); Schoolhouse, 275 F.3d at 729 (finding approximately 74 percent match rate insufficient to demonstrate substantial similarity).
Even so, DatabaseUSA contends that evidence cannot, as a matter of law, support Infogroup's claim of infringement. That is true, DatabaseUSA suggests, because the "audits" and "seed" evidence were based only on snapshots of DatabaseUSA's 2013 database rather than the entire database. And those snapshots, DatabaseUSA claims, were essentially non-copyright protected matter (i.e., facts). See filing 508 at 10-16. This argument, however, is an attempt by DatabaseUSA to circumvent the effects of the adverse inference. The Court remains unconvinced. The jury was entitled to infer, based on all the evidence before it, that had Infogroup been allowed to forensically analyze the 2014 database (or any previous version of that database) that every field, creative selection, and arrangement of DatabaseUSA's database is arranged in the exact same way as Infogroup's database. And as such, the Court finds there is more than sufficient evidence to support the jury's finding that the database infringed on Infogroup's copyright. DatabaseUSA's Rule 50 motion will be overruled on those grounds.
As a final matter, however, nearly all Infogroup's evidence of "verbatim copying" was found in either DatabaseUSA's 2013 or 2014 database. There was no evidence adduced at trial illustrating how, if at all, DatabaseUSA's post-2014 databases infringed on Infogroup's 2011 copyright. And without the adverse inference, which applies only to the 2014 database (and previous versions of that database), there is no reason why Infogroup did not, and could not, introduce the 2011 database for a side-by-side comparison with the available databases. See Experian, F.4d 893 at 1187. Nor is the Court convinced that Infogroup circumstantially proved infringement after 2014. There is no evidence that Infogroup's seeds appear in any post-2014 versions of DatabaseUSA's database, nor is there evidence of similar typographical errors in the subsequent versions of the Database. Filing 476 at 22; E109, 153. And the Court is not persuaded that the 91 percent match rate, without more, is sufficient to prove "extensive verbatim copying" needed for factual compilations—particularly given the availability of those databases in their entirety. See generally id.
In all events, the Court will deny DatabaseUSA's Rule 50 motion. There is more than sufficient evidence that at least some versions of DatabaseUSA's database (i.e., the 2013 and 2014 databases) infringed on Infogroup's 2011 copyright. Relatedly, the Court will deny DatabaseUSA's Rule 59 motions— the jury's liability finding is not against the clear weight of the evidence.
The jury was also asked to decide two separate, but related, Lanham Act claims: false advertising and mark infringement. After considering all the evidence, the jury determined that DatabaseUSA had falsely advertised and committed mark infringement. And according to DatabaseUSA, that conclusion is not supported by the evidence. See filing 508 at 24-28; see also filing 511 at 6-8. The Court disagrees and will deny DatabaseUSA's motions.
With respect to Infogroup's mark infringement claim, the jury specifically found that DatabaseUSA infringed on three of Infogroup's service marks: (1) infoUSA; (2) REFERENCEUSA; and (3) SALESGENIE.com. The principle underlying trademark protection is that distinctive marks—words, names, symbols, and the like—can help distinguish a particular artisan's goods from those of others. Sturgis Motorcycle Rally, Inc. v. Rushmore Photo & Gifts, Inc., No. 17-1762, 2018 WL 5726690, at *3 (8th Cir. Nov. 2, 2018). To prove a trademark infringement claim, a plaintiff must show that it has a valid, protectible mark and that there is a likelihood of confusion between its mark and the marks that the defendants were using. Id.; B&B Hardware, Inc. v. Hargis Indus., Inc., 569 F.3d 383, 389 (8th Cir. 2009); see also 15 U.S.C. § 1125(a)(1)(A).
DatabaseUSA does not challenge the validity of Infogroup's marks, nor does it contend that it did not use Infogroup's marks. Instead, DatabaseUSA argues that there is not sufficient evidence suggesting Infogroup's customers were likely confused between the marks. Filing 508 at 26. In determining whether a likelihood of confusion exists, courts consider: (1) the strength of the owner's mark; (2) the similarity between the owner's mark and the alleged infringer's mark; (3) the degree to which the products compete with each other; (4) the alleged infringer's intent to "pass off" its goods as those of the owner; (5) incidents of actual confusion; and (6) the type of product, its costs, and conditions of purchase. B&B Hardware, Inc., 569 F.3d at 389; Insty*Bit, Inc. v. Poly-Tech Indus., Inc., 95 F.3d 663, 667 (8th Cir. 1996).
DatabaseUSA takes issue with the perceived lack of evidence suggesting "incidents of actual confusion." Filing 508 at 26. That is, DatabaseUSA claims that the only evidence of actual confusion was testimony that a handful of customers thought DatabaseUSA's use of Infogroup's marks meant that the two entities were actually the same company or were affiliated in some way. Filing 477 at 14, 16-17. That limited evidence, DatabaseUSA claims, cannot support a finding of mark infringement. Filing 508 at 26.
Even though this is not the strongest part of Infogroup's case, the Court disagrees. DatabaseUSA's argument rests solely on its assumption that Infogroup must demonstrate that its customers were actually confused. But that assumption is not accurate. SquirtCo. v. Seven-Up Co., 628 F.2d 1086, 1091 (8th Cir. 1980) (actual confusion is not essential to a finding of trademark infringement, although it is positive proof of likelihood of confusion). As the Eighth Circuit made clear in its most recent infringement case, Sturgis Motorcycle Rally, Inc., "there is no reason to foreclose the possibility that in some actions a mark's owner may rely on a visual inspection, without any corroboration from consumer surveys or examples of actual confusion, to prove there is a likelihood of confusion." Id. at *14 (internal citations omitted). Stated differently, confusion can be established simply by comparing the overall impression the use of Infogroup's mark created. Id. And here, there was evidence that DatabaseUSA used Infogroup's exact same marks, compare E16, 17, 18 with E83, 88, 94, to promote DatabaseUSA's nearly identical business. See filing 476 at 253-254; filing 477 at 47-48.
And that determination is bolstered by the testimony of actual confusion on behalf of some of Infogroup's customers. See filing 477 at 14, 16-17. Indeed, when determining whether a likelihood of confusion exists, weight is given to the instances and extent of actual confusion. Duluth News-Tribune, a Div. of Nw. Publ'ns, Inc. v. Mesabi Pub. Co., 84 F.3d 1093, 1098 (8th Cir. 1996); Life Techs., Inc. v. Gibbco Sci., Inc., 826 F.2d 775, 777 (8th Cir. 1987). And Infogroup submitted evidence that it received numerous emails and phone calls from customers questioning whether Infogroup and DatabaseUSA were the same, or related entities. See E92; 95; 102; 174; see also Van Gilder Deposition at 40.
But even if there was a complete lack of evidence from which the jury could infer actual confusion, the remaining "likelihood of confusion factors" support the jury's determination. Indeed, neither party disputes that the marks used were exactly the same.
Relatedly, the jury was also asked to decide whether DatabaseUSA engaged in false advertising. To establish a Lanham Act false advertising claim, a plaintiff must prove (1) a false statement of fact by the defendant in a commercial advertisement about its own or another's product; (2) the statement actually deceived or has the tendency to deceive a substantial segment of its audience; (3) the deception is material, in that it is likely to influence the purchasing decision; (4) the defendant caused its false statement to enter interstate commerce; and (5) the plaintiff has been or is likely to be injured as a result of the false statement. Buetow v. A.L.S. Enters., Inc., 650 F.3d 1178, 1182-83 (8th Cir. 2011); United Indus. Corp. v. Clorox Co., 140 F.3d 1175, 1180 (8th Cir. 1998).
DatabaseUSA, in particular, takes issue with the final requirement— that Infogroup has been or is likely to be injured. To support why, in its view, Infogroup was not injured by DatabaseUSA's false advertising, DatabaseUSA primarily relies on the Court's previous Memorandum and Order (filing 88) denying Infogroup's request for preliminary injunction. To that end, DatabaseUSA argues that "[t]he evidence presented at trial is much the same as that presented at the outset of this case, when the Court observed `there is not persuasive evidence of injury to Infogroup.'" Filing 508 at 25.
But just because there was not persuasive evidence before the Court at the preliminary injunction stage, does not mean there was not any evidence presented at trial nearly four years later.
More specifically, Iaccarino testified that DatabaseUSA distributed marketing materials suggesting that it has been direct marketing "leaders since 1972." Filing 476 at 61; 280; E81. He also testified that DatabaseUSA sent a letter suggesting that DatabaseUSA has been "helping businesses grow since 1982." E82. But it was Infogroup, not DatabaseUSA, that started in 1982, Filing 476 at 61. And DatabaseUSA's marketing materials also purported to offer "the same product [as Infogroup] at a lower price" Filing 476 at 267; E83. From that evidence, the jury could infer that DatabaseUSA lost sales or goodwill on the grounds that some customers were misled into thinking that DatabaseUSA was actually Infogroup.
Next, DatabaseUSA argues that the evidence does not support a finding that Gupta breached the 2008 Separation Agreement. Filing 508 at 34. To recover in an action for breach of contract, the plaintiff must prove the existence of a promise, its breach, damage, and compliance with any conditions precedent that activate the defendant's duty. Solar Motors v. First Nat. Bank of Chadron, 545 N.W.2d 714 (Neb. 1996). The parties in this case do not dispute that a valid contract existed, but the parties do dispute whether there was sufficient evidence for the jury to determine that Gupta actually breached that contract. See filing 508 at 34-37.
According to Infogroup, Gupta's conduct violated the relevant two provisions of the parties' 2008 Separation Agreement:
E22. To support why, in its view, Gupta breached that agreement, Infogroup generally makes two arguments. First, Infogroup claims that Gupta used Infogroup's confidential information when creating his new company, DatabaseUSA. See filing 518 at 32-34. And second, Infogroup claims that Gupta disparaged Infogroup. See filing 518 at 32-34.
With respect to Infogroup's first contention, breach of the Confidential Information provision, there was evidence at trial that Gupta used the same marketing strategies and targeted potential customers based on customer lists that he took from Infogroup. In particular, Van Gilder testified to the following:
Van Gilder Deposition at 43. And Van Gilder also testified that Gupta specifically told him that the marketing pieces and marketing lists were the same as those used at Infogroup. See Van Gilder Deposition at 61. There was also evidence that Gupta used Infogroup's trade and service marks in violation of the agreement. E83, 88, 94. That evidence, if believed by the jury, would support its conclusion that Gupta took information specifically protected under the 2008 agreement.
In addition to the breach of the Confidential Information provision, there was also evidence that Gupta breached the Non-Disparagement portion of the 2008 Separation Agreement. Specifically, in 2011, an article was published where Gupta was quoted saying "[t]he problem with Infogroup is they cannot compete in the marketplace . . . [t]hey have no leadership, no brains and their product is obsolete." E173. In the same article, Gupta claimed that Infogroup "laid off 900 people in the company in Omaha alone." E173. Those comments, the jury could conclude, were "disparaging" and "critical" of Infogroup. And as such, there was sufficient evidence for the jury to determine that Gupta breached the 2008 agreement.
Accordingly, the Court finds that the jury's finding of liability on Infogroup's 2008 breach of contract claim is sufficiently supported by the evidence. Thus, the Court will deny DatabaseUSA's Rule 50 and Rule 59 motions on those grounds.
DatabaseUSA also claims that the breach of contract, unfair competition, and unjust enrichment claims fail as a matter of law. To support why those claims necessarily fail, DatabaseUSA relies on the same arguments previously rejected by the Court. Specifically, DatabaseUSA argues that these claims are preempted, and alternatively, that there was no evidence of copying, nor was there any evidence of actual deception. See filing 508 at 29-33. For the reasons discussed above, the Court will deny DatabaseUSA's Rule 50 motion and 59 Motions with respect to Infogroup's remaining state law claims.
To summarize: the jury's findings on each of Infogroup's causes of action—copyright infringement, mark infringement and false advertising under the Lanham Act, breach of the 2008 Separation Agreement, breach of the 2012 Settlement Agreement, unfair competition, and unjust enrichment— are not against the great weight of the evidence, but rather, are sufficiently supported by the evidence. Damages, however, require more discussion.
But before addressing the sufficiency of the jury's award, it is helpful to begin the discussion with the evidence that was actually adduced at trial. As briefly discussed above, Infogroup's damages evidence was presented primarily through the testimony of John Hofmann—former chief financial officer of Infogroup. see Filing 479 at 6-79, E113, 114. Hofmann's testimony reached two relevant conclusions: (1) DatabaseUSA's revenue attributable to copying Infogroup's 2011 database, false advertising, mark infringement, breach of the 2012 settlement agreement, unfair competition, and unjust enrichment was $39.6 million, see filing 114; filing 479 at 34-35; 44-45; and (2) Infogroup's damages as a result of Gupta's breach of the 2008 separation agreement were $10 million. Filing 479 at 34-35;45. In other words, according to Infogroup's own expert, the maximum amount Infogroup could recover in this lawsuit was $49.6 million: $10 million for Gupta's breach of the 2008 separation agreement, and $39.6 million for the remaining allegations of wrongdoing. See filing 479 at 34-35; 44-45.
But the jury nonetheless awarded Infogroup $53.6 million in damages. $39.6 million of that award was based on the jury's finding that DatabaseUSA infringed on Infogroup's 2011 copyright, $4 million was based on the jury's finding that DatabaseUSA had engaged in false advertising, and $10 million of that award was based on the jury's finding that Gupta had violated the 2008 separation agreement.
The Court will first consider the jury's $39.6 million award on Infogroup's copyright infringement claim. A prevailing plaintiff in an infringement action is entitled to recover the infringer's profits to the extent they are attributable to the infringement. 17 U.S.C. § 504(b); Frank Music Corp., 772 F.2d at 514. In establishing the infringer's profits, the plaintiff is required to prove only the defendant's sales; the burden then shifts to the defendant to prove the elements of costs to be deducted from sales in arriving at profit. 17 U.S.C. § 504(b). Any doubt as to the computation of costs or profits is to be resolved in favor of the plaintiff. Shapiro, Bernstein & Co. v. Remington Records, Inc., 265 F.2d 263 (2d Cir. 1959). If the infringing defendant does not meet its burden of proving costs, the gross figure stands as the defendant's profits. See Russell v. Price, 612 F.2d 1123, 1130-31 (9th Cir. 1979), cert. denied, 446 U.S. 952 (1980).
Infogroup presented evidence that DatabaseUSA's total revenue after the copying of Infogroup's 2011 database is approximately $46.6 million. Hofmann took that number and reduced it to $39.6 million by multiplying an overlap factor determined by Infogroup's data compilation team and subtracting a commission figure. Filing 479 at 37; E114. But the problem with Hofmann's testimony is that Hofmann's damage calculation assumes Infogroup is entitled to profits in years that DatabaseUSA did not establish any sort of infringement. And a key component of Infogroup's damage burden is to establish the infringer's profits to the extent they are attributable to the infringement. 17 U.S.C. § 504(b). That is, there must be some sort of causal nexus between the profits Infogroup seeks to recover and the actual infringement.
By way of example, if a new book were published about the adventures of boy wizard Harry Potter, his friends Ron and Hermione, and his lawyer Themis Blackacre, J.K. Rowling would be entitled to the profits made from that infringing publication, despite the introduction of a new original character. But if the same author were to publish a second book based solely on the legal adventures of Blackacre, there is nothing to suggest that J.K. Rowling would be entitled to those profits—even if Blackacre's popularity or profitability was due to her initial association with Harry Potter and his wizard friends. That is true because if the new work does not infringe the original copyright, J.K. Rowling would not be entitled to the profits attributable to the second book.
And although the infringement of a database is a bit more convoluted given its nature, the underlying premise remains the same: to recover DatabaseUSA's profits from 2013-2018, Infogroup needed to present evidence that those databases continued infringing on Infogroup's 2011 copyright in the years following DatabaseUSA's actual infringement—which occurred in 2014. And although the Court can imagine a scenario where subsequent iterations of DatabaseUSA's post-2014 database contained data taken from the 2011 database . . . there was no evidence of that presented to the jury at trial. Instead, the evidence adduced was that Infogroup's databases were constantly changing. Filing 477 at 156.
More specifically, Iaccarino testified that Infogroup's database goes through hundreds of changes on a daily basis. See filing 476 at 109-113. And Khanna also testified that Infogroup's database is modified, updated, and changed on at least a monthly basis. See filing 477 at 157. Given these frequent changes, to recover DatabaseUSA's profits in 2015 and beyond, Infogroup was required, as a matter of law, to provide the jury with some evidence from which it could infer that DatabaseUSA's 2015 database, and any subsequent databases, continued to infringe on Infogroup's 2011 copyright.
But by 2015 there is no evidence that any of Infogroup's 2011 seeds remained in DatabaseUSA's database.
And that failure is not insignificant, because the post-2014 databases were readily available to Infogroup—yet the jury still was presented no evidence of how, if at all, those databases continued to infringe on Infogroup's 2011 copyright.
Stated more simply, the evidence supports the jury's conclusion that DatabaseUSA's 2014 database (and its previous versions) infringed on Infogroup's 2011 copyright.
As a final matter, DatabaseUSA claims that the Court should have instructed the jury that it may award only nominal damages on Infogroup's copyright claim. Filing 511 at 28. But nominal damages are generally reserved for instances where a breach of duty or an infraction of the plaintiff's right is shown, but no serious loss is proved. See generally Welch v. Spangler, 939 F.2d 570, 573 (8th Cir. 1991). And specifically in the context of copyright infringement, a nominal damage instruction is appropriate when there is evidence suggesting that the infringer's costs unrelated to its infringement would reduce damages to zero. Branch v. Ogilvy & Mather, Inc., 772 F.Supp. 1359, 1363 (S.D.N.Y. 1991); see generally Peer Int'l Corp. v. Pausa Records, Inc., 909 F.2d 1332, 1337 (9th Cir. 1990). But here, DatabaseUSA failed to present any evidence that its profits were the result of factors unrelated to its copyright infringement.
The jury's finding that $4 million would fairly and reasonably compensate Infogroup for DatabaseUSA's false advertising is also problematic. Where a defendant misrepresented its own product (i.e., falsely advertised), the plaintiff may be only one of many competitors, and without proof of causation and specific injury the plaintiff might receive a windfall unrelated to its own damage. Porous Media Corp., 110 F.3d at 1335-36. As such, when assessing actual damages, courts may consider the difficulty of proving an exact amount of damages from false advertising, as well as the maxim that the wrongdoer shall bear the risk of the uncertainty which his own wrong has created. Id. at 1336; 15 U.S.C. § 1117(a). But the court must ensure that the record adequately supports all items of damages claimed and establishes a causal link between the damages and the defendant's conduct, lest the award become speculative or violate the Lanham Act's prohibition against punishment. Id.
Here, the jury's $4 million is entirely speculative. There is no evidence that DatabaseUSA's false advertising reaped $4 million in undeserved profits.
In other words, Infogroup's missing link is the introduction of any evidence allowing a reasonable inference as to the amount of business diverted to DatabaseUSA that is attributable to false advertising.
The jury's awards for breach of the 2012 Settlement Agreement, unfair competition, and unjust enrichment are duplicative of the jury's false advertising and copyright infringement awards. See filing 464 at 4-6. Those claims are either based on DatabaseUSA's use of Infogroup's marks, its intent to pass off its goods or services, or deceive its customers. More specifically, the jury awarded $4 million on Infogroup's breach of the 2008 agreement. The jury then reduced that award by $4 million to prevent double recovery—because Infogroup had already recovered those damages under its false advertising award. See filing 464 at 7. And the jury's awards for unfair competition and unjust enrichment stem from Infogroup's false advertising and copyright infringement claims. See filing 464 at 5, 8. In particular, the jury awarded $43.6 million—but reduced that awarded by the entire $43.6 million ($39.6 million to prevent double recovery on Infogroup's copyright infringement and $4 million for false advertising). See filing 464 at 5, 8.
But even if those awards were not reduced by the jury to prevent double recovery, there are inherent causation problems associated with awarding damages on Infogroup's state law claims. With respect to the breach of the 2012 Settlement Agreement, as discussed in the previous section, the $4 million award was not even a figure presented by Infogroup's expert, much less an amount Infogroup connected to DatabaseUSA's breach of the contract.
The jury's $43.6 million award on unjust enrichment and unfair competition is equally problematic. That award is in direct conflict with Hofmann's testimony that the $39.6 million encompassed all Infogroup's damages. See filing 479 at 34-5, 44-45. And even assuming—for sake of argument—that award were supported by the evidence, Hofmann's $39.6 million figure, E114, is based entirely on an overlap percentage purporting to connect DatabaseUSA's profits to its infringement (i.e., the copying of Infogroup's database). It has no bearing on whether Infogroup's unfair competition or unjust enrichment resulted in any undeserved profits. See E114; filing 479 at 34-5; 44-45.
In sum, the Court finds that the jury's $4 million award for breach of the 2012 Settlement Agreement, and $43.6 million award for unfair competition and unjust enrichment, suffer from the same causation problems discussed at length above. As such, the Court will grant DatabaseUSA's Rule 50 motion on those grounds. But because the jury already reduced those awards to prevent double recovery, see filing 464 at 5, 7, 8, the Court need not remit those awards any further.
That leaves the jury's $10 million award for Gupta's breach of the 2008 separation agreement. The proper measure of damages in a contract action is the losses sustained by reason of a breach. See Bachman v. Easy Parking of Am., Inc., 562 N.W.2d 369, 373 (Neb. 1997); see also Lone Cedar Ranches, Inc. v. Jandebeur, 523 N.W.2d 364, 370 (Neb. 1994). That is, the ultimate objective of a damages award is to put the injured party in the same position that the injured party would have occupied if the contract had been performed, that is, to make the injured party whole. See Radecki v. Mutual of Omaha Ins. Co., 255 Neb. 224, 583 N.W.2d 320 (1998); see also Larsen v. First Bank, 245 Neb. 950, 515 N.W.2d 804 (1994). And as a general rule, a party injured by a breach of contract is entitled to recover all damages which are reasonably certain and which are naturally expected to follow the breach. Ed Miller & Sons, Inc. v. Earl, 502 N.W.2d 444 (Neb. 1993).
Infogroup sought damages reflecting its $10 million severance payment to Gupta in exchange for the promises contained in the 2008 Separation Agreement. And at trial, Hofmann testified that Gupta's failure to uphold his end of the bargain was worth $10 million to Infogroup. Filing 479 at 60. That is true because under the terms of the separation agreement, Infogroup's priority was preserving confidentiality and ensuring non-disparagement. Filing 476 at 146-149; Filing 479 at 60. And because Gupta did not abide by those fundamental terms, Infogroup, in essence, suggests that the damage it suffered is quantified by the entire amount it paid to Gupta. Simply put, in Hofmann's opinion, $10 million would put Infogroup back in the position it would have been in had Gupta not breached the contract (i.e., $10 million would make Infogroup whole). Filing 479 at 60.
DatabaseUSA's argument about why that award cannot stand as a matter of law is not that Hofmann's testimony is flawed—rather, DatabaseUSA contends that the $10 million award does not fit into the theory of damages upon which the jury was instructed. More fundamentally, DatabaseUSA claims that the $10 million award "effected a one-way recession of the contact" despite the fact that Gupta fulfilled other portions of the contract. Filing 508 at 38. And because that recession theory was neither pleaded, nor instructed on, DatabaseUSA claims that the $10 million award must be remitted. Filing 508 at 38.
But that argument is not evidentiary—it is instructional. And DatabaseUSA neither asked for that instruction nor objected to the instruction actually given. See filing 454 at 35; filing 481 at 25-57. Instead, by seeking an actual damage instruction—which, to be clear, both parties agreed (and even insisted) upon giving—DatabaseUSA opened the door for the jury to accept Hofmann's testimony as true, and conclude that $10 million would in fact make Infogroup whole. See filing 479 at 59-60.
So, while the Court agrees that the jury's $10 million award arguably fits better under a recession theory, the Court cannot say, as a matter of law, that the jury's $10 million award fails to put Infogroup back in the position it would have held had Gupta not breached the 2008 contract. Indeed, the entire purpose of the Separation Agreement is evinced by its name: Infogroup wanted to cut ties with Gupta entirely. And to accomplish that goal, Infogroup agreed to pay Gupta $10 million plus some additional benefits in exchange for Gupta's resignation, and a promise that he would not engage in future conduct that would harm the company. Filing 476 at 142-148; E22.
But despite that agreement, Gupta continued using Infogroup's marks, customer lists, marketing materials, and also spoke negatively about Infogroup and its management. See E83, 88, 94. Simply put, Gupta did exactly what he was paid $10 million (plus some) not to do. And it cannot be beyond all bounds of reason that the reimbursement of that sum is "naturally expected to follow the breach." Ed Miller & Sons, 502 N.W.2d at 451. After all, the only evidence of the value of the "services" Gupta was being paid for is what Infogroup was willing to pay for them. Thus, the jury's determination that reimbursement of the $10 million would make Infogroup whole is supported by sufficient evidence.
To summarize the jury's award, the Court will deny DatabaseUSA's Rule 59 motion on the grounds that the jury's award was against the clear weight of the evidence. The Court will, however, grant DatabaseUSA's Rule 50 motion in part and deny it in part. Specifically, Court finds that Infogroup's copyright infringement award is remitted to $11.2 million (the incremental margin from 2013 and 2014, see E114) and Infogroup's award for false advertising and mark infringement claim under the Lanham Act is remitted to $0. Infogroup's breach of the 2012 settlement agreement, unfair competition, and unjust enrichment claims suffer from the same causational deficiencies as portions of Infogroup's copyright infringement claim, and its Lanham Act claims—but because those claims have already been reduced to prevent double recovery, the Court need not remit that award any further. Infogroup's breach of contract claim against Vinod Gupta will stand in its entirety. Accordingly, the Court will remit the jury's verdict to a total $21.2 million, of which DatabaseUSA is liable for $11.2 million for copyright infringement and Gupta is personally liable for $10 million as a result of his breach of the 2008 Separation Agreement.
Infogroup asks the Court to award it reasonable attorney's fees for its claims under the Copyright Act and the Lanham Act. Filing 495 at 2. In a copyright action, a district court "in its discretion may . . . award a reasonable attorney's fee to the prevailing party." 17 U.S.C. § 505. Attorney's fees, however, are not awarded to the prevailing party automatically or as a matter of course. See Fogerty v. Fantasy, Inc., 510 U.S. 517, 533, (1994) (noting no presumption for fee awards in Copyright Act claims). When determining whether an award of attorney's fees is appropriate, the Court considers factors such as whether the lawsuit was frivolous or unreasonable, the losing litigant's motivations, the need in a particular case to compensate or deter, and the purposes of the Copyright Act. Action Tapes, Inc. v. Mattson, 462 F.3d 1010, 1014 (8th Cir. 2006).
But, as DatabaseUSA correctly points out, no award for attorney's fees under the Copyright Act is permitted when "any infringement of copyright in an unpublished work commenced before the effective date of its registration." 17 U.S.C. § 412. And here, Infogroup's 2011 copyright was registered on September 10, 2015—at least one year after DatabaseUSA's infringement occurred. E14; Ez-Tixz, Inc. v. Hit-Tix, Inc., 919 F.Supp. 728 (S.D.N.Y. 1996) (collecting cases) (infringement "commences" when the first act of infringement in a series of on-going infringements occurs). Thus, the Court determines that an award for attorney's fees is not available for Infogroup's copyright claim. See Feldhacker v. Homes, 173 F.Supp.3d 828, 832 (S.D. Iowa 2016); Dutch Jackson IATG, LLC v. Basketball Mktg. Co., 846 F.Supp.2d 1044, 1052 (E.D. Mo. 2012).
But Infogroup has also moved for attorney's fees under the Lanham Act. A prevailing plaintiff is entitled, subject to the principles of equity, to recover the costs of the action. 15 U.S.C. § 1117(a)(3). In addition, in "exceptional cases," a court may award reasonable attorney's fees to the prevailing party. Id.; see B & B Hardware, Inc. v. Hargis Indus., Inc., 716 F.3d 1020, 1027 (8th Cir. 2013); First Nat. Bank in Sioux Falls v. First Nat. Bank S.D., 679 F.3d 763, 771 (8th Cir. 2012); Devon Park, 634 F.3d at 1013. Where a defendant's conduct was willful and deliberate, a court may well determine that it is the type of exceptional case for which an award of attorney's fees is appropriate. First Nat. Bank, 679 F.3d at 771; Devon Park, 634 F.3d at 1013.
The Court determines that this is such a case. DatabaseUSA willfully and deliberately used Infogroup's marks, after promising it would not do so in the parties' 2012 settlement agreement. See Devon Park, 634 F.3d at 1013-14. Infogroup has submitted evidence of $426,250.10 in costs and fees incurred related to its Lanham Act claims.
The Court bears the responsibility of scrutinizing attorney's fees requests, and the burden rests with counsel to establish a factual basis to support the award. Johnston v. Comerica Mortg. Corp., 83 F.3d 241, 246 (8th Cir. 1996). In cases such as this, the Court uses the "lodestar" approach. Id. Under the "lodestar" methodology, the hours expended by an attorney are multiplied by a reasonable hourly rate of compensation so as to produce a fee amount which can be adjusted, up or down, to reflect the individualized characteristics of a given action. Id.
The standards to be considered in calculating attorney's fees under a "lodestar" approach are (1) the number of hours spent in various legal activities by the individual attorneys, (2) the reasonable hourly rate for the individual attorneys, (3) the contingent nature of success, and (4) the quality of the attorneys' work. Jorstad v. IDS Realty Trust, 643 F.2d 1305, 1312-13 (8th Cir. 1981); see also Grunin, 513 F.2d at 127. The "reasonable hourly rate" for purposes of a lodestar analysis is the "hourly amount to which attorneys of like skill in the area would typically be entitled for a given type of work on the basis of an hourly rate of compensation." Jorstad, 643 F.2d at 1313. The starting point is multiplying attorneys' hours and typical hourly rates; only after such a calculation do other, less objective factors come into the equation. Grunin, 513 F.2d at 127.
The Court has carefully considered the hours spent on the case by Infogroup's attorneys, and their hourly rates. See filing 494-1 at 14-26; see also filing 494-1; filing 494-2; filing 494-3; filing 494-4; filing 494-5; filing 494-6; filing 494-7. The Court has also considered the challenging nature of this litigation, and the extensive preparation and care reflected in the briefings and evidence submitted over the course of the litigation by Infogroup's capable counsel. Considered under the lodestar approach, the Court finds Infogroup's requested amount to be fair and reasonable, and will enter its award accordingly.
Infogroup also seeks injunctive relief to permanently enjoin DatabaseUSA and Gupta from using its marks. According to well-established principles of equity, a plaintiff seeking a permanent injunction must satisfy a four-factor test before a court may grant such relief. Specifically, a plaintiff must demonstrate: (1) that it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction. eBay Inc. v. MercExchange, L.L.C., 126 S.Ct. 1837, 1839 (2006).
On balance, the Court finds that such proof is present here. Infogroup has provided the Court with sufficient evidence that it suffered irreparable harm. See Ferring Pharm., Inc. v. Watson Pharm., Inc., 765 F.3d 205, 216 (3d Cir. 2014), holding modified by Reilly v. City of Harrisburg, 858 F.3d 173 (3d Cir. 2017). And that is particularly true given the extensive evidence adduced at trial demonstrating that DatabaseUSA and Gupta remain undeterred by Infogroup's registration of its marks or their contractual obligations to refrain from using those marks. Additionally, Infogroup's inability to recover monetary damages for DatabaseUSA and Gupta's unlawful conduct bolsters the need for injunctive relief: while the Court was not persuaded in the preliminary stages of this case that monetary relief would be unavailable, that plainly didn't bear out—so, the difficulty Infogroup faced in proving the amount of its damages weighs in favor of finding that its injuries are irreparable by money damages. And finally, there is nothing to suggest that anyone would be harmed by an injunction—in fact, preventing the defendants from misleading future potential customers would enhance, not disserve, the public interest. Accordingly, the Court will grant permanent injunctive relief.
The scope of that relief, however, is something of a challenge. On the one hand, the injunction sought by Infogroup, filing 498 at 4-5, is broader than warranted by the evidence. And an injunction must not be broader than necessary to remedy the underlying wrong. Gerlich v. Leath, 861 F.3d 697, 710 (8th Cir. 2017). Moreover, an injunction cannot be too vague and must give fair and precisely drawn notice of what the injunction actually prohibits. Bennie v. Munn, 822 F.3d 392, 397 (8th Cir. 2016), cert. denied, 137 S.Ct. 812 (2017), and cert. denied, 137 S.Ct. 814 (2017). And Infogroup's proposed injunction with respect to its marks "or any substantially similar mark" is neither fairly nor precisely drawn.
On the other hand, an injunction which does little or nothing more than order the defendants to obey the law is not specific enough either. Id. And the Court has authority to issue a broad injunction in cases where a proclivity for unlawful conduct has been shown. Capitol Records, Inc. v. Thomas-Rasset, 692 F.3d 899, 906 (8th Cir. 2012). More precisely, the Court may enjoin certain otherwise lawful conduct where the defendant's conduct has demonstrated that prohibiting only unlawful conduct would not effectively protect the plaintiff's rights against future encroachment. Id. When a party has violated the governing statute—here, the Lanham Act—the Court may enjoin the conduct that allowed the prohibited actions to occur, even if that conduct would have been lawful standing alone. See id.
The Court has previously observed that many of Gupta's uses of Infogroup's marks might fall within the scope of nominative fair use. Filing 88 at 18-21. But even then, the Court noted that many of Gupta's uses of those marks "come very close to the line." Filing 88 at 20. And the jury, having been instructed on nominative fair use, filing 462 at 25, found mark infringement. That verdict, as explained above, was supported by the evidence, because the defendants have proven unable—or more likely, unwilling—to carefully toe the line between fair use and mark infringement. So, the Court concludes that the appropriate scope of injunctive relief is to move that line to a place where it is easier for the defendants to see. The Court will not only enjoin further deceptive use of Infogroup's marks, but will also enjoin Gupta and DatabaseUSA from using Infogroup's marks when describing Gupta's experience or qualifications in DatabaseUSA's marketing.
As a final matter, DatabaseUSA claims various instructions were improper and as such, contends that a new trial is warranted. A new trial may be appropriate when a jury has been improperly instructed. See, e.g., McKay v. WilTel Commc'n Sys., Inc., 87 F.3d 970, 976 (8th Cir. 1996). The Court examines whether, taken as a whole and viewed in the light of the evidence and applicable law, the instructions fairly and adequately submitted the issues in the case to the jury. Gill v. Maciejewski, 546 F.3d 557, 563 (8th Cir. 2008). A party is entitled to an instruction on its theory of the case so long as it is legally correct and there is factual evidence to support it. Boesing, 540 F.3d at 890. But the instructions need not be technically perfect, Gill, 546 F.3d at 563, and a party is not entitled to any particular wording in the instructions. Ryther v. KARE 11, 108 F.3d 832, 847 (8th Cir. 1997).
The Court has reviewed the instructions and finds no error that substantially effected DatabaseUSA's rights. As such, the Court will deny DatabaseUSA's Rule 59 motion on those grounds.
For the foregoing reasons, the Court will grant injunctive relief, and will award attorney's fees in the total amount of $426,250.10 payable jointly and severally by Gupta and DatabaseUSA. In addition, pursuant to this order, judgment is amended and is entered in favor of Infogroup, Inc. and against DatabaseUSA in the amount of $11,200,000.00 and in favor of Infogroup, Inc. and against Vinod Gupta in the amount of $10,000,000.00.
