Zeta Global Corp. v. Maropost Marketing Cloud, Inc.

Opinion

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK --- --------------------------------------------------------- X : ZETA GLOBAL CORP., : Plaintiff, : : 20 Civ. 3951 (LGS) -against- : : OPINION AND ORDER MAROPOST MARKETING CLOUD, INC., : Defendant. : ------------------------------------------------------------ X

LORNA G. SCHOFIELD, District Judge: On July 7, 2022, the Court granted Defendant Maropost Marketing Cloud, Inc.’s (“Maropost”) motion for judgment on the pleadings on Plaintiff Zeta Global Corp.’s (“Zeta”) last remaining claim, and the Clerk of Court entered judgment for Maropost. On July 22, 2022, Maropost moved to declare this case exceptional for purposes of awarding attorney’s fees under 35 U.S.C. § 285. For the reasons below, Maropost’s motion is denied. BACKGROUND Familiarity with the underlying facts and procedural history is assumed. See Zeta Glob. Corp. v. Maropost Mktg. Cloud, Inc., No. 20 Civ. 3951, 2022 WL 2533182 (S.D.N.Y. July 7, 2022); Zeta Glob. Corp. v. Maropost Mktg. Cloud, Inc., No. 20 Civ. 3951, 2021 WL 2823563, at *6 (S.D.N.Y. July 7, 2021), reconsideration denied, 2021 WL 3682488 (S.D.N.Y. Aug. 19, 2021); Zeta Glob. Corp. v. Maropost Mktg. Cloud, Inc., No. 20 Civ. 3951, 2021 WL 1668134, at *1 (S.D.N.Y. Apr. 28, 2021). Only the facts relevant to the instant motion are recounted below. Zeta originally brought this action against Maropost for breach of contract, tortious interference with contract and declaratory judgment (the “Non-Patent Claims”) and infringement of three patents owned by Zeta, U.S. Patent Nos. 7,536,439 (the “’439 Patent”), 8,108,475 (the “’475 Patent”) and 10,257,672 (the “’672 Patent” and, collectively, the “Patent Claims”). Zeta initially filed substantially the same claims in the Central District of California. After the court tentatively concluded that it likely lacked personal jurisdiction over Maropost, Zeta voluntarily dismissed the case and refiled in this Court. Maropost moved to dismiss the Non-Patent Claims based on a forum selection clause in the relevant contract. The relevant contract passage states: “This Agreement is governed by the

laws of the Province of Ontario, Canada. All disputes arising out of or in connection with this Agreement shall be referred to and finally resolved under the laws of Ontario and in Toronto, Ontario.” On April 28, 2021, the Court granted Maropost’s motion. Zeta argued that the forum selection clause was only a choice of law provision, and that the reference to disputes being resolved “in Toronto, Ontario” simply selected the municipal law of Toronto. That argument was rejected because the Toronto municipal code addresses only municipal issues and does not include a body of substantive contract law. Zeta also argued, unsuccessfully, that the categorical, mandatory language of the forum selection clause was merely permissive. While Maropost’s partial motion to dismiss was pending, the parties proceeded with the

claim construction process. On November 30, 2020, Maropost served invalidity contentions on Zeta, in which Maropost contended that all of the asserted claims of all three patents were anticipated and also were not patent-eligible. According to Maropost, the parties exchanged proposed claim terms for construction on December 15, 2020, and Zeta notified Maropost that it was withdrawing its claims related to the ’672 Patent on January 12, 2021, the day before the parties were to exchange proposed constructions of disputed terms. On February 5, 2021, in the parties’ Joint Claim Construction Statement, Zeta withdrew its claims related to the ’672 Patent. On July 7, 2021, in its Claim Construction Opinion and Order, the Court dismissed Zeta’s claims related to the ’439 Patent on the grounds that the Patent’s relevant claims are indefinite. The Court found that the term “failure message” could not be understood in the ’439 Patent, even though neither party had pressed that argument. The Court found that the same problem did not affect ’475 Patent and rejected Maropost’s arguments that other terms in that patent were indefinite. On August 19, 2021, both parties’ motions for reconsideration were denied. In parallel with the claim construction process, the parties exchanged correspondence

about Zeta’s infringement contentions and discovery requests. Maropost repeatedly stated its belief that Zeta failed to assert a prima facie claim of infringement and demanded that Zeta withdraw its claims. Zeta refused. Briefly, the relevant claim of the ’475 Patent describes a three-step process of (1) receiving a “failure message” from an Internet Service Provider (“ISP”), (2) classifying a “failure type” using the failure message and (3) determining email invalidity based on the “failure type” and the identity of the ISP. In Maropost’s words, its non- infringement argument “hinged” on whether Maropost made a “determination of email invalidity using failure type and ISP information.” Maropost argued that its process (1) classifies failure messages into a “binary” failure type -- either a “hard bounce,” indicating that the email is

invalid, or a “soft bounce,” indicating that it is valid and the email should be re-sent after a deferral, and (2) determines whether to add an email to a “do not mail” list based only on whether the failure type is a hard or soft bounce. Maropost argued that it used ISP information only at the step of determining the “failure type” -- “hard” or “soft” -- and not at the third step of determining email invalidity, as Zeta’s process did. At the same time, Zeta continually pressed Maropost to produce more of its source code in discovery. Maropost preferred to produce only those pieces of code that Maropost viewed as relevant to Zeta’s claims, while Zeta sought source code for a large swath of Maropost’s functionality so it could all be evaluated together. One particular dispute arose over Zeta’s requests for code about Maropost’s “deferral” process, the process through which Maropost decided based on the failure message and ISP how frequently to re-try soft-bounced messages. Maropost contended that the deferral process was irrelevant to the dispute described above over Zeta’s infringement contentions. On July 7, 2022, Maropost’s motion for judgment on the pleadings was granted. At Step

One of the analysis required by Alice Corp. v. CLS Bank Int’l, 573 U.S. 208 (2014), the claims on the ’475 Patent were found to be patent-ineligible abstract ideas. Mere receipt of failure messages, like sending failure messages, solved a problem that “long permeated the pre-Internet, analog world” and had been addressed by basic email technology. See GoDaddy.com LLC v. RPost Commc’ns Ltd., No. 14 Civ. 126, 2016 WL 3165536, at *26 (D. Ariz. June 7, 2016), aff’d, 685 F. App’x 992 (Fed. Cir. 2017). Classification of failure messages and deciding whether to invalidate email addresses on that basis is also an abstract idea. See Intell. Ventures I LLC v. Symantec Corp., 838 F.3d 1307 (Fed. Cir. 2016). Like the email filtering system addressed in Intell. Ventures I, such classification is analogous to the “long-prevalent practice for people

receiving paper mail to look at an envelope and discard certain letters, without opening them, from sources from which they did not wish to receive mail based on characteristics of the mail.” Id. at 1314. The ’475 Patent merely addresses the Internet version of an age-old challenge -- failed mail delivery -- which is not “particular to the Internet” in any relevant way. Cf. DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1257 (Fed. Cir. 2014). Zeta tried to salvage its claim by arguing that its patented method involved communication with ISPs, but whether true or not, that assertion was found to have no basis in the claims of the ’475 Patent itself. At Step Two of the Alice analysis, the Court found no “inventive concept” that was “sufficient to ensure that the patent in practice amounts to significantly more than a patent upon the ineligible concept itself.” See PersonalWeb Techs. LLC v. Google LLC, 8 F.4th 1310, 1318 (Fed. Cir. 2021). Neither the Complaint nor the ’475 Patent itself suggested that it did anything more than “add[] computer functionality to increase the speed or efficiency of the process,” Intell. Ventures I, 838 F.3d at 1370, using “a generic computer to perform generic computer functions.” Id. at 1368 (quoting Alice, 573 U.S. at 225-26). Again, Zeta relied on a purported

enablement of communication with the ISP, which failed because it was not “evident in the claims.” Two-Way Media Ltd. v. Comcast Cable Commc’ns, LLC, 874 F.3d 1329, 1338 (Fed. Cir. 2017); accord Berkheimer v. HP Inc., 881 F.3d 1360, 1367, 1369 (Fed. Cir. 2018). STANDARD Section 285 of the Patent Act provides that a district court “in exceptional cases may award reasonable attorney fees to the prevailing party.” 35 U.S.C. § 285; accord Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545, 548 (2014). A prevailing party in patent litigation is a party that receives “at least some relief on the merits, which alters the legal relationship of the parties.” Winters v. Wilkie, 898 F.3d 1377, 1381 (Fed. Cir. 2018) (quoting

Buckhannon Bd. & Care Home, Inc. v. W. Virginia Dep’t of Health & Human Res., 532 U.S. 598, 606 (2001)) (internal quotation marks omitted); see also Sanofi-Synthelabo v. Apotex, Inc., No. 02 Civ. 2255, 2002 WL 1917871, at *3 n.2 (S.D.N.Y. Aug. 20, 2002) (“Federal Circuit law controls substantive and procedural issues that pertain to patent law.”). An award of attorney’s fees under § 285 requires a two-step inquiry: first, “a district court must determine whether a case is exceptional”; second, if the court determines the case is exceptional, “it must calculate the amount of the attorney’s fees award.” Large Audience Display Sys., LLC v. Tennman Prods., LLC, 660 F. App’x 966, 970 (Fed. Cir. 2016) (citing Octane Fitness, LLC, 572 U.S. at 554). The instant motion seeks only a declaration that the case is exceptional, which would permit Maropost separately to submit a claim for attorney’s fees and costs. Only the first step is addressed below. An exceptional case “is simply one that stands out from others with respect to the substantive strength of a party’s litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in which the case was litigated.” Octane Fitness,

LLC, 572 U.S. at 554. District courts must make this determination “in the case-by-case exercise of their discretion, considering the totality of the circumstances,” with relevant factors including, but not limited to, “frivolousness, motivation, objective unreasonableness (both in the factual and legal components of the case) and the need in particular circumstances to advance considerations of compensation and deterrence.” Id. at 554 n.6 (internal quotation marks omitted); accord In re Rembrandt Techs. LP Patent Litig., 899 F.3d 1254, 1277 (Fed. Cir. 2018). While exceptional cases are “rare,” a district court may award fees where “a party’s unreasonable conduct -- while not necessarily independently sanctionable -- is nonetheless so ‘exceptional’ as to justify an award of fees.” Octane Fitness, LLC, 572 U.S. at 555. A party must establish its entitlement to

fees by a preponderance of the evidence. Id. at 557-58; accord In re Rembrandt Techs. LP Patent Litig., 899 F.3d at 1269. DISCUSSION This is not an exceptional case. While Zeta’s merits arguments were unsuccessful, they were for the most part not frivolous or objectively unreasonable. Maropost’s arguments that Zeta litigated unreasonably amount to complaints that Zeta sought broader discovery than Maropost wished to give, and pressed for certain information over Maropost’s assertions that the requests were irrelevant. Maropost’s arguments that Zeta acted with improper motivations are unpersuasive. The totality of the circumstances, considering the need for compensation and deterrence, do not warrant taking the “rare” step of finding that this case is “exceptional.” A. Zeta’s Validity Positions Zeta’s positions on the validity of the ’439 and ’475 patents were not objectively unreasonable. In the end, Zeta’s patents claimed nothing more than using a generic computer to

perform long-familiar pre-Internet functions, but Zeta’s patents were not on all fours with any invalidated in Alice, Intell. Ventures I, or any other prior case cited by Maropost. It was not unreasonable for Zeta to attempt to distinguish the contrary but constantly evolving case law. In Intell. Ventures I, for example, the Federal Circuit held “that receiving e-mail (and other data file) identifiers, characterizing e-mail based on the identifiers, and communicating the characterization -- in other words, filtering files/e-mail -- is an abstract idea.” 838 F.3d at 1313. The court there relied on an analogy to the mental process of sorting mail based on its characteristics, “look[ing] at an envelope and discard[ing] certain letters, without opening them, from sources from which they did not wish to receive mail based on characteristics of the mail.”

Id. at 1314. Maropost cites several other cases invalidating patents aimed at using computers merely to receive and categorize data. See PersonalWeb, 8 F.4th at 1315-16; Content Extraction & Transmission LLC v. Wells Fargo Bank, N.A., 776 F.3d 1343 (Fed. Cir. 2014). But it was not unreasonable for Zeta to take the position that using a computer to understand and categorize ISP failure messages is distinguishable from the processes addressed by the cases cited above, in which a computer looks for and applies rules based on a fixed set of identifiers or data elements. According to the Complaint and the terms of Zeta’s patents, categorizing failure messages is different because it requires ISP-specific processes. That is, Zeta’s patents arguably claim a process for categorizing messages by looking at both the data itself and the metadata -- the information both on and inside the “envelope.” Zeta’s claim ultimately fails because there is no indication in the patents that such functionality in fact requires anything more than a slightly more complex set of business rules, or any “communication” with the ISP beyond receiving the message. See GoDaddy.com, 2016 WL 3165536, at *26. But Zeta did not act in an objectively unreasonable manner by attempting to draw a distinction that turned out to be unconvincing.

The cases cited by Maropost demonstrate that courts post-Alice commonly invalidate software patents similar to Zeta’s. But those cases do not demonstrate that courts have taken the “rare” step of finding analogous cases “exceptional.” Maropost principally relies on Gust, Inc. v. AlphaCap Ventures, LLC, 226 F. Supp. 3d 232 (S.D.N.Y. 2016), which was reversed by the Federal Circuit. Gust, Inc. v. AlphaCap Ventures, LLC, 905 F.3d 1321 (Fed. Cir. 2018). While the appeal formally addressed only the fee award under 28 U.S.C. § 1927 -- which had made plaintiff’s counsel jointly and severally liable for the award under 8 U.S.C. § 285 -- the Federal Circuit rejected the district court’s § 285 analysis in the process. Gust, 905 F.3d at 1330. The Federal Circuit cases cited by Maropost are not on point. Innovation Scis., LLC v.

Amazon.com, Inc., 842 F. App’x 555 (Fed. Cir. 2021), addressed the distinguishable situation in which a party’s litigation positions become unviable following claim construction. And Inventor Holdings, LLC v. Bed Bath & Beyond, Inc., 876 F.3d 1372 (Fed. Cir. 2017), addressed a category of patents that arguably is most clearly invalidated by Alice, the “computer-implemented business transaction.” Id. at 1379. In hindsight, it may seem obvious that receiving an ISP- generated failure message and then categorizing and acting on it based on ISP-specific rules has clear analogs in pre-Internet, abstract mental processes, but it was not objectively unreasonable for Zeta to have disagreed and urged a different view in this litigation. Maropost’s arguments that Zeta knew its patents were invalid are similarly unpersuasive. First, the fact that Maropost told Zeta that Maropost subjectively believed the patents to be invalid deserves little weight. Second, the fact that Zeta attempted to stretch a fair reading of its patent to salvage it is not, in this case, strong evidence that Zeta knew its patent was invalid as written. Zeta’s claim that the patented invention enables “communication” with ISPs is not an

absurd description of what Zeta’s system does when it receives failure messages. The problem is that, as discussed above, a system that merely receives failure messages is not patentable. There is no indication that the “communication” here goes beyond the patent-ineligible one-way process of receiving a message, to something more like “corresponding” with the ISP. Maropost also argues that Zeta’s assertion of the ’439 Patent was unreasonable because it was invalidated for indefiniteness at claim construction. This argument is unavailing because Maropost did not even press the argument that was ultimately adopted to invalidate the ’439 Patent. Even if Zeta’s claim construction on the ’439 Patent was frivolous, Zeta’s assertion of that patent would not have caused hardly any of the costs that Maropost now seeks. The ’475

Patent was substantively very similar to the ’439 Patent, though it fixed the indefiniteness problem identified in the latter. Thus, the course of the litigation would not have looked much different even if the ’439 Patent had never been asserted or if it had survived claim construction. B. Zeta’s Infringement Contentions and Discovery Tactics Zeta’s infringement positions were not objectively unreasonable. Maropost’s argument on this score is almost self-refuting. Maropost argues that Zeta “never cited evidence that Maropost’s Marketing Cloud platform uses ISP information along with failure type to invalidate an email address.” That is because, Maropost argues, “failure type determination was binary; failures were classified merely as either hard or soft bounces,” where “hard bounces” mean that an email is invalid and “soft bounces” do not. By defining “failure type” as a binary, Maropost argues that it used ISP information only to determine a failure type, not to decide whether to invalidate an email. But Maropost also concedes that, within the category of “soft bounces,” “Maropost used ISP information to set deferral intervals,” i.e., the time that Maropost waits before retrying to send a failed email. Maropost thus used ISP information both to make a non-

binary classification and a binary one, though Maropost chooses to call only the latter “failure type.” It is at least reasonable, however, to read Maropost’s own claims to say it uses ISP information both to classify the type of failure -- as a prerequisite to making the non-binary determination of deferral interval -- and to make the binary validate/invalidate decision. For similar reasons, Zeta’s discovery tactics were not particularly aggressive or improper. As the foregoing discussion demonstrates, Zeta had at least colorable bases on which to pursue broader discovery than initially was forthcoming from Maropost, including about Maropost’s deferral process. Maropost argues that Zeta’s approach to discovery was disproportional to the strength of Zeta’s patents, but that argument is unpersuasive based on the arguments about

validity above. As Zeta was entitled to pursue claims based on its patents, it was entitled to pursue reasonable discovery in an effort to prove those claims. Maropost also argues that Zeta’s litigation approach was disproportional to the value of the case. Maropost claims that it had changed its process for categorizing failure messages to eliminate any consideration of the ISP’s identity and thus any possibility of infringement, and that it told Zeta that. Maropost faults Zeta for failing to recognize that the Patent Claims became “worthless” after this change, and for seeking discovery on the “new method” in the face of Maropost’s assurances that it did not infringe. This argument is unpersuasive because Zeta was not required to take Maropost’s word for it, particularly given the circumstances of this case. Both Zeta and Maropost believed at one point that considering ISP information added value to the process of invalidating emails based on failure messages, since both did so at some step of the process. In the Non-Patent claims, Zeta also accused Maropost of poaching its customers in violation of a contract, at a time when Maropost used the allegedly infringing method. It was not objectively unreasonable for Zeta take the position that its claims were not worthless and to

probe both Maropost’s original process and how it changed, if at all. C. Zeta’s Motivations Maropost’s arguments that Zeta acted with improper motives are unavailing. Maropost argues that Zeta’s ulterior motive for pursuing the Patent Claims was its desire to obtain a favorable global settlement that also covered the Non-Patent Claims. Maropost cites (1) Zeta’s purportedly unreasonable discovery tactics, (2) repeated proposals to enter a global settlement, (3) the fact that the first three counts in the Complaint were Non-Patent Claims and (4) the fact that Zeta initially sought a preliminary injunction only on the Non-Patent Claims. None of the above is persuasive evidence of an ulterior motive. As discussed above,

Zeta’s approach to discovery is objectively unreasonable only if one accepts Maropost’s premise that Zeta’s merits positions were objectively unreasonable. Any rational litigant with parallel lawsuits against the same adversary -- particularly claims that initially were brought as part of a single suit -- might seek a global settlement. And whether Zeta saw the Non-Patent Claims as having some kind of priority over the Patent Claims is irrelevant. Maropost cannot fault Zeta for its failure to litigate the Patent Claims even more aggressively, by seeking a preliminary injunction on claims that Maropost argues were always, obviously doomed to fail. Under all the circumstances, considerations of compensation and deterrence do not favor a finding that this case is exceptional. Zeta pursued relatively weak Patent Claims that ultimately were unsuccessful, but its positions were not objectively unreasonable and there is little evidence that it acted with improper motivations. IV. CONCLUSION For the foregoing reasons, Maropost’s motion is DENIED. The Clerk of Court is respectfully directed to close the motion at Docket Number 134. Dated: January 23, 2023 New York, New York

UNITED STATES DISTRICT JUDGE