Versata Software, Inc. v. Callidus Software, Inc.

United States Court of Appeals for the Federal Circuit ______________________ VERSATA SOFTWARE, INC., VERSATA DEVELOPMENT GROUP, INC., AND VERSATA, INC., Plaintiffs-Appellees, v. CALLIDUS SOFTWARE, INC., Defendant-Appellant. ______________________ 2014-1468 ______________________ Appeal from the United States District Court for the District of Delaware in No. 1:12-cv-00931-SLR, Judge Sue L. Robinson. ______________________ Decided: November 20, 2014 ______________________ CONOR M. CIVINS, Bracewell & Giuliani LLP, of Aus- tin, Texas, argued for plaintiffs-appellees. With him on the brief were MICHAEL CHIBIB and BRIAN C. NASH. DEBORAH E. FISHMAN, Dickstein Shapiro, LLP, of Palo Alto, California, argued for defendant-appellant. With her on the brief were ASSAD H. RAJANI and MICHAEL S. TONKINSON. ______________________ 2 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. Before CHEN, MAYER, and LINN, Circuit Judges. CHEN, Circuit Judge. Callidus Software Inc. appeals from the district court’s denial of a stay pending the United States Patent and Trademark Office’s (Patent Office) covered business method review of the patents-in-suit. For the reasons set forth below, we reverse and remand with instructions to stay the action. BACKGROUND Versata Software, Inc., Versata Development Group, Inc., and Versata, Inc. (collectively, Versata) brought suit against Callidus on July 19, 2012, alleging infringement of U.S. Patent Nos. 7,904,326 (the ’326 patent); 7,908,304 (the ’304 patent); and 7,958,024 (the ’024 patent). The asserted patents each concern the management and tracking of sales information by a financial services company. In September 2012, Callidus filed a motion to transfer the case from the District of Delaware to the Northern District of California and a separate motion to dismiss for failure to state a claim under Federal Rule of Civil Proce- dure 12(b)(6). In May 2013, seven months after briefing on these motions had concluded, 1 the district court issued orders denying both. Shortly thereafter, Callidus answered the complaint and counterclaimed, asserting three of its own patents against Versata. It informed Versata that it would file petitions for post-grant review of Versata’s asserted patents at the Patent Office under the Transitional Pro- gram for Covered Business Method (CBM) Patents. See America Invents Act (AIA), Pub. L. No. 112-29, § 18, 125 1 Briefing on these motions concluded in October 2012. VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 3 Stat. 284, 329–31 (2011). In early August 2013, Callidus also informed the district court that it intended to file CBM petitions and to seek a stay of all proceedings in the litigation pending the Patent Office’s resolution of those petitions. Later that month, Callidus filed a first set of CBM petitions with the Patent Office challenging every claim of the ’326 patent, every independent claim of the ’024 and ’304 patents, and several dependent claims of those two patents, all under 35 U.S.C. § 101. At that time, Versata had not yet identified its asserted claims in the litigation. Callidus simultaneously sought a stay of the district court proceedings. The district court declined to consider a stay until the Patent Trial and Appeal Board (PTAB) had made a deci- sion to institute CBM review. It set a Markman hearing for June 2015, and scheduled trial for October 2015. In the interim, Versata identified its asserted claims in December 2013, some of which were dependent claims of the ’024 and ’304 patents that Callidus had not included in its first set of CBM petitions. On March 4, 2014, the PTAB instituted CBM review for each patent, finding each challenged claim more likely than not directed to unpatentable subject matter under § 101. See Callidus Software, Inc. v. Versata Software, Inc., Nos. CBM2013-00052, CBM2013-00053, CBM2013- 00054 (P.T.A.B. Mar. 4, 2014) (decisions instituting CBM review). Callidus renewed its motion to stay the district court proceedings. In April 2014, Callidus also prepared and filed a second set of CBM petitions, challenging the validity of all of the remaining claims of the ’024 and ’304 patents under § 101 that Callidus had not raised in the first set of CBM petitions. On May 8, 2014, two months after the PTAB institut- ed CBM review for the first set of petitions, the district court granted a stay as to the ’326 patent but denied it as to the ’024 and ’304 patents. Callidus filed an interlocuto- 4 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. ry appeal from the district court’s denial of the stay in part. During the pending appeal, the PTAB instituted CBM review of Callidus’s second set of petitions, finding every remaining claim of the ’024 and ’304 patents more likely than not unpatentable under § 101. See Callidus Soft- ware, Inc. v. Versata Software, Inc., Nos. CBM2014-00117, CBM2014-00117 (P.T.A.B. Oct. 2, 2014) (decisions insti- tuting CBM review). We have jurisdiction over interlocu- tory appeals from decisions granting or denying stays pending CBM review under § 18(b) of the AIA. DISCUSSION Section 18(b) identifies four factors that a district court should consider when deciding whether to grant a stay: (A) whether a stay, or the denial thereof, will sim- plify the issues in question and streamline the trial; (B) whether discovery is complete and whether a trial date has been set; (C) whether a stay, or the denial thereof, would unduly prejudice the nonmoving party or present a clear tactical advantage for the moving party; and (D) whether a stay, or the denial thereof, will re- duce the burden of litigation on the parties and on the court. AIA § 18(b)(1). While we typically review district court decisions on motions to stay for abuse of discretion, the AIA also provides this court authority “to conduct more searching review of decisions to stay pending CBM review.” Benefit Funding Sys. LLC v. Advance Am. Cash Advance Ctrs. VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 5 Inc., 767 F.3d 1383, 1385 (Fed. Cir. 2014); see also AIA § 18(b)(2) (providing that this court “shall review the district court’s decision to ensure consistent application of established precedent, and such review may be de novo”). Even under the traditional abuse of discretion standard, the district court’s order denying a stay pending the PTAB’s review must be reversed. The district court considered each factor and conclud- ed that “[d]espite the presence of Congress’ thumb on the scales of justice,” the balance of factors did not favor a stay. See Versata Software, Inc. v. Callidus Software, Inc., No. CV 12-931-SLR, 2014 WL 1868869, at *3 (D. Del. May 8, 2014) (hereinafter, District Court Decision). We consider each factor in turn. Factor (A): Simplification of the Issues The district court noted at the outset that “Callidus chose to seek limited CBM review before the PTAB on a subset of the patent claims at issue.” Id. at *2. In addi- tion, because Callidus has asserted its own patent in- fringement counterclaims against Versata, the district court concluded that a stay will not simplify the issues. Id. On appeal, Callidus claims that a stay will simplify some issues for trial, even if the CBM review does not dispose of them all. On the other hand, Versata contends that the PTAB will address only patentable subject mat- ter under § 101 of a subset of claims, leaving the district court to address many other issues, including other inva- lidity challenges and the subset of claims that were not challenged in the first set of CBM petitions. We agree with Callidus, on the facts here, that a stay will simplify the issues and streamline trial, and this factor weighs in favor of a stay. The district court seem- ingly created a categorical rule that if any asserted claims are not also challenged in the CBM proceeding, this factor 6 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. disfavors a stay. For example, the district court stayed the litigation as to the ’326 patent even though the only distinction between that patent and the ’024 and ’304 patents was that Callidus challenged every claim of the ’326 patent in CBM review while only challenging a subset of claims for the ’024 and ’304 patents. But this type of categorical rule is inappropriate. Stays can be warranted even when a CBM proceeding does not address all asserted patents, claims, or invalidity defenses. See, e.g., VirtualAgility Inc. v. Salesforce.com, Inc., 759 F.3d 1307, 1312–13 (Fed. Cir. 2014) (staying case even though two prior art references not asserted in CBM review); Versata Software, Inc. v. Dorado Software, Inc., No. 2:13– cv–00920–MCE–DAD, 2014 WL 1330652, at *3 n.2 (E.D. Cal. Mar. 27, 2014) (staying case when CBM review granted for only one of three asserted patents and for less than all claims); Progressive Cas. Ins. Co. v. Safeco Ins. Co. of Ill., No. 1:10CV01370, 2013 WL 1662952, at *5 (N.D. Ohio Apr. 17, 2013) (staying case even though PTAB would not address every ground of invalidity and every prior art reference). 2 Certainly this simplification factor weighs more strongly in favor of a stay when all of the litigated claims are undergoing CBM review. See, e.g., VirtualAgility, 759 F.3d at 1314. But there can still be a simplification of the issues when only some, but not all, of the claims asserted in litigation are challenged in a CBM review. 3 In a situa- 2 We also note that, here, the district court stayed the litigation as to the ’326 patent even though the CBM petition for that patent did not assert any §§ 102, 103, or 112 challenges. 3 This is not to say, however, that the grant of a CBM petition on a subset of asserted claims automatically simplifies the issues. By its very nature, the CBM process will always simplify some issues. VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 7 tion like this one, a proper simplification analysis would look to what would be resolved by CBM review versus what would remain. See, e.g., Versata Software, Inc. v. Volusion, Inc., No. A-12-CA-893-SS, 2013 WL 6912688, at *2 (W.D. Tex. June 20, 2013) (noting “strong similarities” between claims invalidated in CBM review and asserted claims in litigation). We note, for example, that all of the independent claims of the ’024 and ’304 patents, as well as some dependent claims, are undergoing CBM review. That fact could be relevant to this analysis. The district court erred by not engaging in this inquiry. We also take judicial notice of the fact that the PTAB has recently instituted CBM review on the remaining claims of the ’024 and ’304 patents based on the subse- quent round of petitions filed by Callidus. 4 Although appellate courts may consider only the record made before the district court, we may also consider, under Federal Rule of Evidence 201(b)(2), the fact that the PTAB has made a decision instituting CBM review. See VirtualAgil- ity, 759 F.3d at 1312–13 (taking judicial notice under Rule 201(b)(2) of the fact that patent owner filed a motion to amend claims in the CBM review proceeding). The review of every claim of each of Versata’s asserted patents leaves little doubt that issues will be simplified. See id. at 1314 (holding simplification factor weighs heavily in favor of stay when CBM review instituted on all claims of only asserted patent). By granting CBM review, the PTAB has determined that every claim is more likely than not unpatentable under § 101. If the PTAB invalidates every 4 Based on the district court’s stay of the case as to the ’326 patent, it appears likely that Callidus would not have been in its present position—of having to appeal the partial denial of a stay as to the ’024 and ’304 patents— had it originally petitioned for CBM review of all claims of those two patents, as it had for the ’326 patent. 8 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. claim, the CBM review would dispose of Versata’s entire affirmative case. See id. at 1314 (“The CBM review could dispose of the entire litigation: the ultimate simplification of issues.”). The simplification factor would also favor Callidus more strongly had Callidus included other grounds for invalidity in its CBM petitions. As we explained, howev- er, it is not necessary for the CBM proceeding to simplify all invalidity issues to warrant a stay. See id. at 1314 (granting stay even though CBM review would not ad- dress two prior art references). Moreover, the statute explicitly allows petitioners, such as Callidus, to raise one invalidity defense without invoking estoppel as to other defenses that could have been raised. Benefit Funding, 767 F.3d at 1387; compare AIA § 18(a)(1)(D) (prohibiting petitioner in CBM review from asserting in litigation “that the claim is invalid on any ground that the petition- er raised during that transitional proceeding”), with 35 U.S.C. § 315(e)(2) (prohibiting petitioner in inter partes review from asserting in litigation “that the claim is invalid on any ground that the petitioner raised or rea- sonably could have raised during that inter partes re- view”) (emphasis added). Balancing these interests, we conclude that district court clearly erred, and the first factor strongly favors a stay. Factor (B): Whether Discovery Is Complete and Whether a Trial Date Has Been Set While not clearly stated, it appears that the district court believed that this factor weighed against a stay, or at best was a neutral factor. The district court acknowl- edged that “discovery is not complete,” but pointed out that “the trial date that is scheduled is within months of when the PTAB is expected to issue its decision.” District Court Decision, 2014 WL 1868869, at *2. First, the district court failed to explain why the fact that the ex- pected date of the PTAB’s decision, which is scheduled to VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 9 issue no later than seven months before the scheduled trial date, counsels against a stay. Second, and more importantly, the district court failed to analyze how much more remains to be done in litigation before reaching the trial date. See VirtualAgility, 759 F.3d at 1317 (evaluat- ing this factor by noting that discovery had not begun, no trial date had been set, eight months of fact discovery remained, and trial was a year away). For these reasons, we conclude that the district court clearly erred by finding that the relative timing of the PTAB’s expected decision alone disfavors a stay. When Callidus filed its first set of CBM petitions and concurrent motion to stay in August 2013, discovery had not yet started; hence, no documents or discovery requests had been exchanged, no depositions taken, no expert reports prepared. Because the Markman hearing and trial were two years away (scheduled for June and Octo- ber 2015), the parties had not yet exchanged claim terms or proposed constructions. The litigation was still in its infancy. See id. On appeal, Versata argues that the case advanced past its infancy between Callidus’s original motion to stay in August 2013 and the district court’s denial of that motion in May 2014. During that period, the parties exchanged discovery requests, issued third-party subpoe- nas, and produced documents, including source code. In Versata’s view, the state of the ongoing discovery under- mined the purpose of a stay: preservation of resources and judicial efficiency. Though unclear from the district court’s brief analy- sis, it appeared to evaluate the stage of litigation at the time of its decision. See District Court Decision, 2014 WL 1868869, at *2 (stating that “discovery is not complete”). This was error. As we explained in VirtualAgility, gener- ally the time of filing the motion will be the relevant stage at which to measure this factor. 759 F.3d at 1317. We 10 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. recognized in VirtualAgility that it is entirely appropriate and within the discretion of the district court to wait for the institution decision before ruling on the motion. Id. And courts are not obligated to ignore advances in the litigation that occur as of the date that the PTAB granted CBM review. See id. at 1317 n.6. But, like in VirtualAgility, whether we look at the August 2013 date when the motion to stay was filed or the March 2014 date when the PTAB granted CBM review, this factor strongly favors a stay. See id. at 1317. By March 2014, the case had not yet progressed to a point that disfavors a stay. Although many documents and discovery requests had been exchanged, fact discovery was still ongoing and no fact witnesses had been deposed. With the close of expert discovery still seven months away, the parties had not filed any expert reports or taken any expert depositions. Appellants’ Br. 26. And finally, the parties had not exchanged proposed terms or claim construction positions, for both the Markman hearing and the trial were well over a year away. While the parties have conducted some discovery, we must also be mindful of the burden on the parties and the court in completing both fact and expert discovery, resolving summary judgment motions, completing the Markman process, and preparing for trial. See VirtualAgility, 759 F.3d at 1317; see also Broad. Innovation, L.L.C. v. Charter Commc’n, Inc., No. 03-CV-2223-ABJ-BNB, 2006 WL 1897165, at *8 (D. Colo. July 11, 2006) (“[C]ourts consid- ering this factor do not stop at discovery and trial set- tings, but rather, routinely inquire as to the occurrence summary judgment arguments, rulings on summary judgment, and the status of the final pretrial order, among other elements.”). We therefore conclude that in this case, the timing factor strongly favors a stay, and the VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 11 district court clearly erred by not evaluating the status of the litigation. 5 Factor (C): Undue Prejudice or Tactical Advantage On this factor, the district court made no factual find- ings that Versata would suffer undue prejudice as a result of a stay. 6 Instead, the court determined that Callidus held an improper tactical advantage over Versata by seeking to stay Versata’s infringement claims while still pursuing its own infringement counterclaims: 5 Even if we evaluated this factor as of May 2014— the date when the district court denied the motion—the outcome would not change. Both fact and expert discov- ery remained ongoing, and the Markman hearing and trial date were still well over a year away. The case had not, in a matter of two months, progressed in any mean- ingful way. At that date, the preservation of the parties’ and the court’s resources would be served by a stay. The same is true of October 2014—the date of the PTAB’s institution of the second set of CBM petitions—as expert discovery was still ongoing, the parties had not meaning- fully engaged in claim construction, and trial was a year away. 6 Absent factual findings, we decline to consider whether either party would be unduly prejudiced. We note, however, that generic concerns raised by Versata in its appeal brief about undue prejudice, i.e., that a stay will raise issues with stale evidence, faded memories, and lost documents, are, without more, true of any stay and may affect either party’s preservation of its evidence. See VirtualAgility, 759 F.3d at 1319 (“It is undoubtedly true, as many courts have observed, that with age and the passage of time, memories may fade and witnesses may become unavailable. Without more, however, these asser- tions here are not sufficient to justify a conclusion of undue prejudice.”). 12 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. Given the fact that Callidus has affirmatively as- serted its patents against Versata, it is apparent that Callidus is playing the stay card as both a sword and a shield, moving forward on its inter- ests but denying Versata the opportunity to do the same, thus presenting a clear tactical advantage for Callidus, the moving party. District Court Decision, 2014 WL 1868869, at *2. The district court’s assertion that Callidus intended to “move forward on its [own] interests” while staying Versa- ta’s case-in-chief is not correct. Callidus made clear to the district court that it was seeking to stay the entire case. See Joint Appendix (J.A.) 0600 (“THE COURT: . . . Would you intend for us to go forward with [your patents]? [Callidus]: No, your Honor. In our motion for stay, we asked to stay the entire litigation.”). At oral argument, Versata did not dispute that Callidus sought a stay of the entire case. See Oral Argument at 30:20–31:56, available at http://www.cafc.uscourts.gov/oral-argument- recordings/2010-1648/all. The district court’s findings therefore were clearly erroneous. Although not clear, it appears that the district court may have also found that Callidus’s motions to transfer and dismiss for failure to state a claim amounted to improper tactics. But if this was a consideration in the district court’s analysis, the opinion fails to explain why such motions were improper. For example, Versata does not claim that the motions were frivolous. Litigants are within their rights to seek a proper forum or to dismiss a claim, and these motions by themselves are not improper, without some basis to find otherwise. This record does not suggest any undue prejudice to Versata or clear tactical advantage to Callidus. The district court clearly erred, and we conclude that this factor strongly favors a stay. VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. 13 Factor (D): Reduced Burden of Litigation Callidus’s motions practice influenced the district court in finding that this fourth factor weighs against a stay. By filing preliminary motions, the district court concluded that “Callidus’ tactics have actually increased the burdens of litigation, rather than reduced them.” District Court Decision, 2014 WL 1868869, at *2. 7 We conclude that the district court clearly erred in evaluating the burden-of-litigation factor exclusively through this backward-looking lens. The correct test is one that focuses prospectively on the impact of the stay on the litigation, not on the past actions of the parties. Cf. VirtualAgility, 759 F.3d at 1317 (noting that fact discov- ery was ongoing and the parties had not engaged in meaningful claim construction or jury selection); Sightsound Techs. v. Apple, Inc., No. 11-1292, 2013 WL 2457284, at *3 (W.D. Penn. June 6, 2013) (“Nonetheless, while much has been done thus far, there is more to come. The parties and Court will expend further substantial resources in this litigation, through completing discovery and trial.”). When framed appropriately, it becomes clear that a stay will indeed reduce the future burdens of litigation. 7 VirtualAgility recognized that the reduced burden of litigation factor may be implicated by a court’s familiar- ity with the patents at issue. 759 F.3d at 1314 n.4. It was not implicated in this case, however, by denial of Callidus’s motion to dismiss. The district court grounded the denial on Versata adequately identifying the accused products, asserting facially plausible claims under its pleading burden, and providing Callidus with adequate notice—not on any substantive discussion of the patents- in-suit. See Versata Software, Inc. v. Callidus Software Inc., 944 F. Supp. 2d 357, 362–63 (D. Del. 2013). 14 VERSATA SOFTWARE, INC. v. CALLIDUS SOFTWARE, INC. As in VirtualAgility, we note that the simplification analysis under the first factor often points in the same direction as reducing the burden of litigation under the fourth factor. 759 F.3d at 1313. Since the CBM proceed- ings will move forward on all asserted claims, if Callidus is successful, a stay will relieve the parties and the dis- trict court of having to expend substantial resources on claim construction and both noninfringement and invalid- ity defenses. See id. For these reasons and the same reasons identified for the first factor, we conclude that the district court clearly erred, and that this fourth factor strongly favors a stay. CONCLUSION Because the factors when taken together strongly fa- vor a stay pending CBM review, we reverse and remand with instructions to grant the motion to stay as to the ’024 and ’304 patents. REVERSED AND REMANDED