1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 CITY OF HOLLYWOOD FIREFIGHTERS Case No. 3:23-cv-00519-WHO PENSION FUND, et al., 8 Plaintiffs, ORDER GRANTING MOTION TO 9 DISMISS SAC v. 10 Re: Dkt. No. 61 ATLASSIAN CORPORATION, et al., 11 Defendants.

12 13 In this putative securities class action, the lead plaintiffs amended their complaint to 14 attempt to plead that the defendants made materially false and misleading statements to investors 15 and shareholders, and that the defendants had the requisite scienter required under the securities 16 laws. Despite a previous dismissal without prejudice, the plaintiffs added few substantial 17 allegations to their newly amended complaint. The bulk of the new allegations come from a new 18 confidential witness—but the facts related to this witness are confusing, inconsistent, and 19 ultimately unhelpful to the plaintiffs’ case. For that and the following reasons, the plaintiffs again 20 fail to plead the requisite elements for this securities case. Because the plaintiffs could plausibly 21 amend their allegations to address the inconsistencies and explain their theory of scienter, they are 22 granted leave to amend—one more time—to do so. 23 BACKGROUND 24 The background of this case is outlined in detail in my Order Granting the Motion to 25 Dismiss the Plaintiffs’ First Amended Complaint. (“Prior Order”) [Dkt. No. 53]. This Order 26 assumes familiarity with those facts, and includes new allegations below, as well as some relevant 27 background for context. 1 Pension and Retirement Systems (“the plaintiffs”) filed this securities lawsuit against defendants 2 Atlassian Corporation, Atlassian Corporation PLC, Atlassian’s co-founder and co-CEO Michael 3 Cannon-Brookes, Atlassian’s other co-founder and co-CEO Scott Farquhar, Atlassian’s current 4 President and former Chief Operating Officer Anu Bharadwaj, and Atlassian’s former Chief 5 Revenue Officer Cameron Deatsch. Second Amended Complaint (“SAC”) [Dkt. No. 59] ¶¶ 1, 3, 6 17–26. 7 Atlassian is a software company that uses a sales and growth strategy of providing free 8 versions of its software products to up to ten users, and then charging customers for the eleventh 9 and subsequent users. Id. ¶¶ 35-38. Given this strategy, Atlassian has two main growth metrics: 10 “Free to Paid Conversions,” which consist of customers that used the free version of its software 11 and then upgraded to a paid version; and “Paid User Expansion,” which consists of customers that 12 used a paid version of the products then added “seats” or “heads”—for example, a customer had 13 50 paid users at its company and then added 10 more paid users. See id. About ninety percent of 14 Atlassian’s revenue comes from “Paid User Expansion.” Id. ¶ 38. 15 In a nutshell, the plaintiffs allege that the defendants made various false and misleading 16 statements to the market, reproduced below, that said Atlassian was not feeling the impact of the 17 macroeconomic pressure and constraints that other software companies faced throughout 2022. 18 When the defendants “revealed” the truth to the market in November 2022, announcing that 19 “cloud growth came in at just 49%” and that the company missed its earnings guidance for the 20 first time, the stock price fell, and the plaintiffs lost a collective $7 billion. Id. ¶¶ 84, 93. 21 Their new allegations, added to the SAC after dismissal of the First Amended Complaint, 22 (“FAC”) [Dkt. No. 40], focus on the experience of Confidential Witness 3 (“CW3”), who asserts 23 that he was the Head of Product Strategy & Business Operations, Work Management from June 2022 24 through June 2023. SAC ¶ 49 n.6. He alleges that when he joined in June 2022, “things” were 25 already “slowing down” at Atlassian, “it was clear growth rates were not what they were,” and 26 Atlassian “was not growing at 50% year-over-year growth.” Id. ¶¶ 72–77. He also asserts that he 27 was told of a hiring freeze in July 2022, id. ¶ 75, and that “paid seat expansion numbers had 1 “everyone” was aware of the metrics, and that he directly or indirectly presented “these metrics” to 2 Bharadwaj. Id. And, he says that “Project Big Fish” was started in July or August to “understand 3 some of the ‘big things’” and “find[] ways to increase Atlassian’s user expansion.” Id. ¶¶ 79–80. 4 Bharadwaj was part of Project Big Fish, though CW3 did not join until September. Id. ¶¶ 80–81. 5 The SAC brings two causes of action: first under Section 10(b) of the Securities Exchange 6 Act, and second under 20(a) of the Act. Id. ¶¶ 151–66. The defendants moved to dismiss the 7 SAC. (“Mot.”) [Dkt. No. 61]. The plaintiffs opposed. (“Oppo.”) [Dkt. No. 62]. The defendants 8 replied. (“Repl.”) [Dkt. No. 63]. I held a hearing at which counsel for both parties appeared. 9 LEGAL STANDARD 10 Under Federal Rule of Civil Procedure (“FRCP”) 12(b)(6), a district court must dismiss a 11 complaint if it fails to state a claim upon which relief can be granted. To survive a Rule 12(b)(6) 12 motion to dismiss, the plaintiff must allege “enough facts to state a claim to relief that is plausible 13 on its face.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007). A claim is facially plausible 14 when the plaintiff pleads facts that “allow the court to draw the reasonable inference that the 15 defendant is liable for the misconduct alleged.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) 16 (citation omitted). There must be “more than a sheer possibility that a defendant has acted 17 unlawfully.” Id. While courts do not require “heightened fact pleading of specifics,” a plaintiff 18 must allege facts sufficient to “raise a right to relief above the speculative level.” Twombly, 550 19 U.S. at 555, 570. 20 In deciding whether the plaintiff has stated a claim upon which relief can be granted, the 21 court accepts the plaintiff's allegations as true and draws all reasonable inferences in favor of the 22 plaintiff. Usher v. City of Los Angeles, 828 F.2d 556, 561 (9th Cir. 1987). However, the court is 23 not required to accept as true “allegations that are merely conclusory, unwarranted deductions of 24 fact, or unreasonable inferences.” In re Gilead Scis. Sec. Litig., 536 F.3d 1049, 1055 (9th Cir. 25 2008). 26 A complaint alleging a violation of section 10(b) of the Securities Exchange Act “must 27 meet both the heightened pleading requirements for fraud” under FRCP 9(b) and the “exacting 1 (9th Cir. 2017) (citations omitted). This requires that the complaint “state with particularity the 2 circumstances constituting fraud,” to satisfy Rule 9(b), and “state with particularity facts giving 3 rise to a strong inference that the defendant acted with the required state of mind,” to meet the 4 PSLRA’s standard. Id. (citations omitted). 5 With respect to falsity, “the complaint [must] specify each statement alleged to have been 6 misleading, [and] the reason or reasons why the statement is misleading.” 15 U.S.C. § 78u- 7 4(b)(1)(B). With respect to scienter, “the complaint shall, with respect to each act or omission 8 alleged . . . state with particularity facts giving rise to a strong inference that the defendant acted 9 with the required state of mind.” 15 U.S.C. § 78u-4(b)(2)(A). “[F]alsity and scienter in private 10 securities fraud cases are generally strongly inferred from the same set of facts, and the two 11 requirements may be combined into a unitary inquiry under the PSLRA.” In re Daou Sys., Inc., 12 411 F.3d 1006, 1015 (9th Cir. 2005), abrogated on other grounds as recognized by In re Genius 13 Brands Int’l, Inc. Sec. Litig., 97 F.4th 1171, 1183 (9th Cir. 2024) (citation and internal quotation 14 marks omitted). 15 “To adequately demonstrate that the defendant acted with the required state of mind, a 16 complaint must allege that the defendants made false or misleading statements either intentionally 17 or with deliberate recklessness.” Zucco Partners, LLC v. Digimarc Corp., 552 F.3d 981, 991 (9th 18 Cir. 2009) (quotation marks and citation omitted). “Facts showing mere recklessness or a motive 19 to commit fraud and opportunity to do so provide some reasonable inference of intent, but are not 20 sufficient to establish a strong inference of deliberate recklessness.” In re VeriFone Holdings, Inc. 21 Sec. Litig., 704 F.3d 694, 701 (9th Cir. 2012) (citation omitted). Accordingly, “a court must 22 consider plausible, nonculpable explanations for the defendant’s conduct, as well as inferences 23 favoring the plaintiff.” Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 324 (2007). 24 “[A]n inference of scienter must be more than merely plausible or reasonable—it must be cogent 25 and at least as compelling as any opposing inference of nonfraudulent intent.” Id. at 314. 26 If the court dismisses the complaint, it “should grant leave to amend even if no request to 27 amend the pleading was made, unless it determines that the pleading could not possibly be cured 1 this determination, the court should consider factors such as “the presence or absence of undue 2 delay, bad faith, dilatory motive, repeated failure to cure deficiencies by previous amendments, 3 undue prejudice to the opposing party and futility of the proposed amendment.” Moore v. Kayport 4 Package Express, 885 F.2d 531, 538 (9th Cir. 1989). 5 DISCUSSION 6 My Prior Order explained that the plaintiffs plausibly alleged that there was a slowdown to 7 Paid User Expansion in mid-August 2022 based on the defendants’ statements to that effect in 8 their November shareholder update. See Prior Order. But of the nine statements alleged to be 9 misleading, I determined that the plaintiffs plausibly alleged only that Bharadwaj misleadingly 10 omitted information about the slowdown during a conference in September 2022; the remaining 11 statements were not plausibly pleaded as false or misleading. See id. But the plaintiffs failed to 12 plead the requisite scienter for Bharadwaj’s statement, so the complaint was dismissed. See id. 13 As I will explain at length below, the plaintiffs’ complaint is again insufficient. They rely 14 almost entirely on CW3’s allegations to remedy the issues from the FAC, but his assertions do not 15 show that any of the other statements were false or misleading because they are too vague and 16 nonspecific to show that the Paid User Expansion slowdown happened before mid-August 2022. 17 And, his assertions do not show that Bharadwaj had the requisite scienter in making her 18 misleading omissions because CW3 failed to establish his reliability or personal knowledge for the 19 allegations he makes. I address these issues in turn. 20 I. STATEMENTS / FALSITY 21 A. August 4 Statements 22 The first three statements were made by Deatsch during the August 4, 2022 earnings call. 23 First: 24 The good news as of to date is we have yet to see any specific trend geographically or even in industry segments or in customer size that gives us 25 pause or worry to date. So, something we continue to watch like a hawk, but there is no new news to share today. 26 SAC ¶ 108; Mot. Ex. B at 6. 27 Second: “[W]e have not seen any significant shift in customer demand across our product 1 lines.” SAC ¶ 108; Mot. Ex. B at 7. 2 The third statement consists of the bolded words below, in context: 3 As far as that slight thing that we’ve mentioned in the shareholder letter, just so you all realize that we land all of our net new customers in Free plans. . . . And the 4 one thing that’s worth calling out that we’ve seen literally just in the last month is that the cohort of customers that came in, in the April, May, and June timeframe 5 are converting to those paid plans at a slightly slower rate than what we’ve seen in previous quarters. . . . There’s no specific customer segment there. . . . [T]hose 6 cohort of customers that are signing up in the last quarter are using the products, 7 . . . but they simply just haven’t put those credit cards and hit those paid walls yet. So that’s one of those areas that we continue to be vigilant. We have multiple 8 analytics teams, multiple growth teams, as well as our onboarding and R&D teams that are focused on ensuring that those customers remain active, and that gives us 9 more chances to convert them to paid plans in the future. That does not take 10 away from the continued growth we see in our existing customer base. That also drives more than 90% of our revenue in existing year. 11 Mot. Ex. B at 9 (emphases added); see also SAC ¶ 111. And in the report published alongside the 12 call, the defendants stated: 13 As mentioned earlier, we are not insulated from broader macroeconomic impacts. We are seeing a modest decrease in the conversion rates of Free instances 14 upgrading to paid plans, but nothing we are seeing right now is changing our 15 outlook. As a reminder, over 90% of our revenue in FY22 was derived from existing customers, and this dynamic has been consistent in our business over the 16 years. We will remain vigilant and continue to inspect all aspects of our business relying on our well refined analytics and understanding of our funnels. 17 Mot. Ex. C at 14 (emphasis added). 18 In my Prior Order, I explained that the plaintiffs failed to plead that these were false or 19 misleading for two reasons. First, the defendants clearly told the investors that Free to Paid 20 Conversions were slowing down, and because this was a small part of their business, the plaintiffs 21 did not plead that the “slight thing” comment was misleading. Prior Order 10–15. Second, the 22 defendants’ November announcement to shareholders and investors stated that the slowdown to 23 Paid User Conversions did not begin until mid-August—meaning August 16 at the earliest—so the 24 plaintiffs failed to plead that Deatsch’s August 4 comments were misleading or false when he 25 failed to inform investors of a slowdown that had not yet begun. See id. To remedy these 26 deficiencies, the plaintiffs must have alleged in the SAC that the Free to Paid Conversion 27 slowdown was not a small part of the business, and they must have alleged that the Paid User 1 Conversion slowdown began before the August 4 statements and that Deatsch knew about the 2 slowdown. See generally id. 3 With respect to Free to Paid Conversion, to remedy the deficiencies in the SAC the 4 plaintiffs again rely on the theory that the defendants’ disclosure of this slowdown on August 4 5 was misleading because the defendants did not specifically label it a “trend.” See SAC ¶ 112. I 6 rejected this identical argument based on an identical theory and nearly identical allegations in my 7 Prior Order, and it fails for the same reasons here. See Prior Order 10:27–11:13. Deatsch’s 8 November statements merely confirmed that the Free to Paid Conversion metric “was slowing 9 down going into August.” Mot. Ex. G at 12. That is exactly what he told the market in August. 10 With respect to Paid User Expansion, to remedy the deficiencies in the SAC the plaintiffs 11 rely almost entirely on the new allegations from CW3 that they say show the slowdown began in 12 June or July 2022. CW3 alleges that he started working at Atlassian in June 2022 and saw 13 immediately that “things were slowing down,” “they were not growing like gangbusters,” “growth 14 rates were not what they were,” Atlassian “was not growing at 50% year-over-year growth,” and 15 the company was not growing at the same rate as prior years. SAC ¶¶ 8, 72, 74. 16 The allegations are insufficient for three reasons. First, they are vague. It is unclear from 17 the SAC, the Opposition, and the other context what metric or metrics “things” refers to and what 18 “not growing like gangbusters” means. See In re Quality Sys., 865 F.3d at 1140 (requiring 19 plaintiffs to state fraud with particularity). It is also not clear from the SAC when exactly CW3 20 thinks the slowdown happened—his allegations about the timing seem to range from June, see 21 SAC ¶ 74 (“back in June, things were slowing down”), sometime in June or July, id. ¶ 75 (“[I]n 22 June and July, he observed the slowdown in the Company’s business”), sometime in July or 23 August, see id. ¶ 74 (noting July and August 2022 growth rates “were not what they were” in the 24 prior year), or even September, id. ¶ 80 (noting CW3 was not involved in Project Big Fish until 25 Labor Day). Without the requisite specificity, not only do the plaintiffs fail to meet the legal 26 standard, but they also fail to put the defendants (and me) on notice of their theory of liability. 27 Second, the plaintiffs do not reconcile CW3’s allegations with the reported metrics about 1 were,” implying they were much lower. See SAC ¶¶ 72, 74. Based on the Opposition—which I 2 look to because, again, CW3’s statements are too vague to know for sure what he intended to 3 say—CW3 seems to be referring to revenue growth. See Oppo. 7:19–28, 10:16–11:11. But 4 Atlassian reported that its year-over-year (“YOY”) revenue for Q4 2022 saw 36 percent growth, 5 while Q4 2021 saw only 30 percent growth.1 Mot. Ex. C at 11. Without more specifics attached 6 to CW3’s statements, then, the only way to reconcile his statement that 2022 growth rates “were 7 not what they were” with the report showing the rates increased is to assume that CW3 was 8 correct—they were not what they were, because they were higher. CW3 also said that Atlassian 9 “was not growing at 50% year-over-year growth,” SAC ¶¶ 72, 74, but he does not specify whether 10 he is talking about total revenue or cloud growth. If the former, he fails to explain why the 11 company should have been growing at 50 percent YOY total revenue when it was reporting 36 12 percent growth to the SEC. If the latter, he fails to explain why Atlassian reported that, during the 13 same time period, its cloud business grew 55 percent. Mot. Ex. C. at 12. These allegations 14 therefore do not support the plaintiffs’ theory of falsity or misrepresentations, regardless of the 15 timing of the start of the slowdown. 16 Third, the plaintiffs do not successfully rely on the existence of Project Big Fish to 17 plausibly allege that there was a slowdown in Paid User Expansion before mid-August. See Oppo. 18 22:24–23:10. For support they cite SAC ¶¶ 72, 77, 79–80, which contain the same confusing and 19 vague allegations addressed above, about how growth rates “were not what they were” and also 20 the 30 percent versus 50 percent rates. These also contain the vague and nonspecific allegations 21 that CW3 said that “[t]he tech market was getting killed at that time” and the executive team was 22 having discussions with other leadership teams to address “slowing” growth and to “understand 23 . . . ‘big things’ that they could implement to reverse these negative trends.” SAC ¶ 79. With 24 respect to Project Big Fish, CW3 said it involved about 100 “upper-level” employees who 25 brainstormed “ways to increase Atlassian’s user expansion through product enhancements and 26 1 I previously took judicial notice of this and other publicly available documents that formed the 27 basis for integral allegations in the complaint. See Prior Order 6–8. I took notice of the facts 1 new features.” Id. ¶ 80. It not clear that Big Fish was initiated because of or targeted slowing 2 Paid User Expansion. Indeed, though CW3 alleges that the project was formed because executives 3 knew they would have to “walk back” earlier statements and guidance during the upcoming 4 November earnings call, his allegations about that earlier guidance are again vague and confusing. 5 See id. ¶ 82. He seems to assert that the executives wanted a plan in place to address the fact that 6 the cloud revenue fell below 50 percent YOY growth, see id., but again fails to reconcile that 7 number with the reported number for that time period—55 percent YOY growth, Mot. Ex. C at 8 12—and with the ultimate reported number for Q1 FY23—49 percent YOY, Mot. Ex. F at 11. 9 Finally and importantly, CW3 did not join Big Fish until September and so has no basis for 10 knowledge about what was happening in June, July, or August. Cf. In re Plantronics, Inc. Sec. 11 Litig., No. 19-CV-07481-JST, 2022 WL 3653333, at *14 (N.D. Cal. Aug. 17, 2022) (inferring 12 witness had “reliable information” where they personally participated in the requisite meetings). 13 None of CW3’s allegations, then, show that Deatsch’s August 4 statement contradicted 14 what he knew at the time, see Khoja v. Orexigen Therapeutics, Inc., 899 F.3d 988, 1008 (9th Cir. 15 2018), or that these statements created any impression that the state of affairs differed “in a 16 material way from the one that actually exists,” Brody v. Transitional Hosps. Corp., 280 F.3d 997, 17 1006 (9th Cir. 2002). They do not help the plaintiffs to plead falsity or misrepresentations. 18 Finally, the plaintiffs emphasize that Deatsch himself said in November that the slowdown 19 to Paid User Conversion began in July, rather than in mid-August. See SAC ¶¶ 87–88. But 20 Deatsch did not say this; what he said was that “the net new customers, the free customers 21 converted to paid was slowing down going into August”—confirming that the Free to Paid 22 Conversions were slowing by August—and “July and August tend to see user growth slow down 23 across our customer base”—which says nothing about what happened to Paid User Conversions in 24 July or August. SAC ¶ 87 (emphases omitted). He also said, “It just tends to be a normal seasonal 25 trend where July and August are slower as people are not upgrading instances, adding more 26 users,” id. ¶ 88 (emphases omitted), but again this says nothing about whether or when there was a 27 slowdown to Paid User Conversions. Accordingly, it is not clear how these statements help the 1 The plaintiffs therefore fail to allege that the defendants’ August 4 statements about Free to 2 Paid Conversions were false or misleading, and they fail to allege that the slowdown to Paid User 3 Expansion occurred before mid-August 2022. Their challenges to the August 4 statements again 4 fail. 5 B. August 19 Statements 6 Next, the plaintiffs again argue that Atlassian’s Annual Report for FY 22, published on 7 August 19, 2022, contained the following misleading statement: 8 Other than as disclosed elsewhere in this Annual Report, we are not aware of any trends, uncertainties, demands, commitments or events for the current fiscal year 9 that are reasonably likely to have a material effect on our revenues, income, profitability, liquidity or capital reserves, or that caused the disclosed financial 10 information to be not necessarily indicative of future results of operations or financial conditions. 11 SAC ¶ 114; Ex. A at 66. 12 In my Prior Order, I explained that the plaintiffs failed to plead that this statement was 13 misleading regarding any trend to Paid User Expansion because they failed to plead that this 14 metric saw any change before August 16, and they did not plead that the three days between the 15 beginning of the slowdown and the publication of the Annual Report was enough to see or report a 16 trend. Prior Order 15:28–16:21. As explained above, supra Part I.A, they again fail to plead that 17 the Paid User Expansion slowdown began before August 16, and so this argument again fails for 18 the same reasons. 19 I also previously determined that the plaintiffs failed to plead that the Annual Report 20 statements were misleading with respect to Free to Paid Conversions because they failed to 21 explain why these statements contradicted the defendants’ knowledge or the state of affairs at that 22 time. Prior Order 16:22–17:1 (citing Khoja, 899 F.3d at 1008; Brody, 280 F.3d at 1006). It seems 23 that the plaintiffs’ theory now is that the defendants called this slowdown a “trend” in November, 24 so they should have specifically labeled it as a trend that was “reasonably likely to have a material 25 effect” in the Annual Report, but instead referred to the slowdown only as “hypothetical.” See 26 Oppo. 23:22–24:10; SAC ¶¶ 114–15. 27 1 November, the defendants told the market, “Last quarter [Q4 FY22], we shared that we saw a 2 decrease in the rate of Free instances converting to paid plans. That trend became more 3 pronounced in Q1 [FY23].” Mot. Ex. F at 2. In other words, the defendants confirmed that Free 4 to Paid Conversions were slowing in Q4 FY22—which they told the market in early August—and 5 they labeled it as a “trend” for Q1 FY23. They did not say it was a trend in Q4 FY22, nor did they 6 say it had a material effect on their revenue or any other metric. And because no other allegations 7 show that this was a trend or had a material effect on their revenue in Q4 FY22, or that the 8 defendants knew either of these, it is not clear how it could be false or misleading to not 9 specifically label it as a material trend in the Annual Report. See Khoja, 899 F.3d at 1008; Brody, 10 280 F.3d at 1006. Nor does the Annual Report say the slowdown to Free to Paid Conversions was 11 merely “hypothetical,” as the plaintiffs assert. Indeed, it says that Atlassian was facing 12 “macroeconomic headwinds and slower revenue growth.” Mot. Ex. A at 66. Again, the plaintiffs 13 ignore the context. 14 Accordingly, the plaintiffs again fail to plead that the August 19 statements “directly 15 contradict[ed]” anything the defendants knew, Khoja, 899 F.3d at 1008, or affirmatively 16 represented a state of affairs that materially differed from the one that “actually exist[ed],” Brody, 17 280 F.3d at 1006. They fail to plead the statements were false or misleading. 18 C. October 4 Statements 19 I previously rejected the plaintiffs’ challenges to the defendants’ statements in the October 20 4 Post-Effective Amendment to their Form S-8 Registration Statement with the SEC. Prior Order 21 20:6–21:7. Again the plaintiffs fail to plead that the August 19 statements were false or 22 misleading, so again to the extent that their theory relies on the falsity of those statements, the 23 claim fails. 24 The plaintiffs now ground their legal theory in a different regulation from their FAC. It 25 provides in relevant part that a successor issuer must file its registration statement and “set[] forth 26 any additional information . . . necessary to keep the registration statement from being misleading 27 in any material respect.” 17 C.F.R. § 230.414(d). 1 because it included the Annual Report but did not include what they say the defendants knew by 2 October, which is that a slowdown was affecting its Paid User Expansion metric. See SAC 3 ¶¶ 122–23; Oppo. 24:11–19. But it is not clear why this would be materially misleading. The 4 plaintiffs do not allege that the Registration Statement said that there were no slowdowns or trends 5 in October, or that the defendants submitted information about their growth from August to 6 October but not their lower performing metrics. See Khoja, 899 F.3d at 1009 (“[O]nce defendants 7 [choose] to tout positive information to the market, they [are] bound to do so in a manner that 8 wouldn’t mislead investors, including disclosing adverse information that cuts against the positive 9 information.” (citations omitted)). Submitting the Annual Report confirms the defendants’ 10 information as of August 19. It does not purport to provide information beyond that point. Nor do 11 the plaintiffs point to law or cases that require anything else. Accordingly, the plaintiffs again fail 12 to plead that the October 4 statements were false or misleading. 13 * * * 14 Accordingly, the plaintiffs again fail to plead that the August or October statements were 15 false or misleading and so the motion is GRANTED for this argument. I turn next to whether they 16 successfully pleaded scienter for Bharadwaj’s September statements. 17 II. SCIENTER 18 A. Statements Previously Found Misleading 19 Bharadwaj’s plausibly misleading statements were made on September 14, 2022, at a 20 conference. First: 21 Just in the last shareholder letter, we talked about conversion of users from free to paid. We are seeing a bit of softness over the past couple of months. But it’s 22 important to illustrate that point in the overall picture as over 90% of our revenue comes from existing customers. So the conversion from free to paid is really a very 23 small part of our business. And also, the number of free users that are coming in that create new free instances continues to grow steadily. We’re not seeing any 24 trends there. So across the board, across existing and new, that's overall what we 25 are seeing in different parts of the business. SAC ¶ 118; Mot. Ex. D at 9. And second: 26 So to address your question, the way our pricing model works is the first 10 users 27 are free and then the 11th user onwards, it's paid. So in year 1, you can imagine the paid user there. So the overall quantum looks pretty low in year 1, but it builds up 1 and compounds over time. So we think of upsell as really across free to standard, 2 standard to premium, premium to enterprise. And we’ve seen some very encouraging progress there over the last 4 quarters, and I haven’t really seen any 3 discernible trend there in terms of the macroeconomic impact. SAC ¶ 119; Mot. Ex. D at 10.2 4 I previously explained that the plaintiffs plausibly alleged that these statements were 5 misleading because, while the defendants “had no affirmative duty to disclose the beginning signs 6 of a slowdown before the quarter ended, they created the duty to disclose by ‘tout[ing] positive 7 information to the market’ about the rest of their business at that time.” Prior Order 19:3–9 8 (quoting Khoja, 899 F.3d at 1009). Though I did not find that the plaintiffs pleaded there was a 9 “trend” at this point as opposed to merely a slowdown, I found the omission plausibly misleading, 10 particularly when combined with the other allegations that the defendants told investors that they 11 were vigilant monitors and open communicators and that they were monitoring Paid User 12 Expansion. Id. 19:17–20:2. But the plaintiffs did not allege scienter because there was no logical 13 motive and the allegations about the comments to analysts, linear sales cycle, and stock trades 14 simply did not show deliberate recklessness. Id. at 22:22–24:4. 15 The defendants ask me to reconsider the finding in my Prior Order that the plaintiffs 16 pleaded these statements were misleading. Mot. 15:17–17:25. They did not file a motion for 17 leave to file a motion for reconsideration, see Civ. Loc. R. 7-9(a), let alone a motion for 18 reconsideration, nor do they cite (or meet) the standard for granting one. They do not point to 19 “newly discovered evidence,” do not argue that I “committed clear error or the initial decision was 20 manifestly unjust,” and they do not assert that there was “an intervening change in controlling 21 law.” Eng. v. Apple Inc., No. 14-CV-01619-WHO, 2016 WL 1108929, at *2 (N.D. Cal. Mar. 22, 22 2016) (citation omitted). Their request is denied. Instead I turn to whether the plaintiffs pleaded 23 Bharadwaj made the statements with the requisite scienter. 24 25 26 2 The plaintiffs appeared to allege another statement was misleading in the first round on the 27 motion to dismiss. See Prior Order 17:14–16. But the SAC specifically alleges that the statements 1 B. Whether the Plaintiffs Alleged These Statements Were Made With Scienter 2 The PSLRA requires that plaintiffs “state with particularity facts giving rise to a strong 3 inference that the defendant acted with the required state of mind.” 15 U.S.C. § 78u-4(b)(2)(A); 4 E. Ohman J:or Fonder AB v. NVIDIA Corp., 81 F.4th 918, 937 (9th Cir. 2023) (“NVIDIA”), cert. 5 granted sub nom. NVIDIA Corp. v. Ohman J, No. 23-970, 2024 WL 3014476 (U.S. June 17, 2024) 6 (same). With respect to false or misleading statements or omissions, “the complaint must 7 plausibly allege, with the particularity required by the PSLRA, that the maker of the statements” 8 knew about the material information and “intentionally or recklessly” failed to disclose it. In re 9 Alphabet, Inc. Sec. Litig., 1 F.4th 687, 705 (9th Cir. 2021). 10 “[T]he required state of mind is a mental state that not only covers intent to deceive, 11 manipulate, or defraud, but also deliberate recklessness.” NVIDIA, 81 F.4th at 937 (quoting In re 12 Quality Sys., Inc., 865 F.3d at 1144). “[D]eliberate recklessness” requires plaintiff to show that a 13 defendant made “a reckless omission of material facts . . . reflect[ing] some degree of intentional 14 or conscious misconduct.” In re Alphabet, Inc., 1 F.4th at 701 (citations omitted). This means 15 that the misrepresentations or omissions were “an extreme departure from the standards of 16 ordinary care, which presents a danger of misleading buyers or sellers that is either known to the 17 defendant or is so obvious that the actor must have been aware of it.” Id. (internal citations and 18 quotation marks omitted). 19 “Even if no single allegation, standing alone, is ‘sufficient to give rise to a strong inference 20 of scienter,’ a holistic review of all the allegations may ‘combine to give rise to a strong inference 21 of scienter.’” NVIDIA, 81 F.4th at 940 (quoting Glazer Cap. Mgmt. L.P. v. Forescout Techs., Inc., 22 63 F.4th 747, 766 (9th Cir. 2023)); see also Tellabs, 551 U.S. at 326 (“The court’s job is not to 23 scrutinize each allegation in isolation but to assess all the allegations holistically.”). The court 24 “must consider plausible, nonculpable explanations for the defendant’s conduct, as well as 25 inferences favoring the plaintiff.” Tellabs, 551 U.S. at 324. The inference of scienter “need not be 26 irrefutable,” but “must be more than merely ‘reasonable’ or ‘permissible’—it must be cogent and 27 compelling, thus strong in light of other explanations.” Id. A “court must compare the malicious 1 complaint to survive a motion to dismiss if the malicious inference is at least as compelling as any 2 opposing innocent inference.” In re Sorrento Therapeutics, Inc. Sec. Litig., 97 F.4th 634, 643 (9th 3 Cir. 2024) (quoting Zucco, 552 F.3d at 991). 4 “Where . . . Plaintiffs’ scienter allegations rely on the statements of confidential witnesses, 5 the complaint ‘must pass two hurdles to satisfy the PSLRA pleading requirements.’” NVIDIA, 81 6 F.4th at 937 (quoting Zucco, 552 F.3d at 995). “First, the confidential witnesses whose statements 7 are introduced to establish scienter must be described with sufficient particularity to establish their 8 reliability and personal knowledge. Second, those statements which are reported by confidential 9 witnesses . . . must themselves be indicative of scienter.” Id. (quoting Zucco, 552 F.3d at 995). 10 The plaintiffs’ new allegations about scienter rely mostly on CW3’s allegations that the 11 slowdown started in June (or July, or August, or September), that Bharadwaj participated in 12 Project Big Fish, and that she reviewed declining metrics, so she knew of the slowdown and 13 intentionally withheld that information in September. Oppo. 12:12–16:3. They also add 14 allegations to try to explain why Bharadwaj would withhold this information, only to share it 15 weeks later in the quarterly report. See SAC ¶ 82. 16 For many of the reasons explained above, most of CW3’s allegations are too nonspecific, 17 vague, or irreconcilable with the published reports to establish reliability and personal knowledge 18 of what Bharadwaj knew and when. See NVIDIA, 81 F.4th at 937. The plaintiffs try to show that 19 Bharadwaj had the requisite scienter because of her review of certain reports, and it is true that a 20 plaintiff can allege scienter by pointing to the defendant’s “access and review of contemporaneous 21 reports” with the relevant information. Id. at 940. But in NVIDIA, the confidential witnesses 22 explained in detail how they knew that the defendant knew the relevant information—the first 23 witness personally prepared presentations about the specific information, “had frequent 24 communications” with high-level executives about that information, and wrote weekly sales email 25 related to that information, while the second witness “personally met” with the defendant who the 26 plaintiffs alleged had the requisite knowledge and scienter. 81 F.4th at 940. 27 In comparison, CW3 says he personally reviewed Paid User Expansion and Free to Paid 1 executives generally reviewed objectives, key performance indicators, and metrics for the “Jira” 2 product, id. ¶ 50, but he fails to make the connection or allege that Paid User Expansion or Free to 3 Paid Conversions were discussed at those meetings.3 He also seems to say that these metrics were 4 discussed at Big Fish meetings in June, July, and August, but simultaneously says that he was not 5 a member of Big Fish until September.4 Id. ¶ 80; cf. In re Plantronics, Inc. Sec. Litig., No. 19- 6 CV-07481-JST, 2022 WL 3653333, at *14 (N.D. Cal. Aug. 17, 2022) (inferring the witness had 7 “reliable information” where they personally participated in the requisite meetings). Even if he 8 had personal knowledge about Big Fish, the complaint says the project was designed generally “to 9 increase Atlassian user expansion,” not to address declining Paid User Expansion. SAC ¶ 80. His 10 personal knowledge of a hiring freeze, id. ¶ 75, also does not show that Bharadwaj knew of a 11 slowdown to Paid User Expansion. 12 And while CW3 says that he or his team “presented” metrics to Bharadwaj that “paid seat 13 expansion numbers had slowed down,” he does not say when and is very vague about how— 14 “either directly in meetings or in a report that he submitted for her review, [or] on other occasions 15 his team’s report was incorporated into a document that was presented to [her].” Id. ¶ 76. These 16 are a far cry from the particular and comprehensive allegations in NVIDIA that established the 17 witnesses’ personal knowledge and reliability.5 See also In re Alphabet, 1 F. 4th at 705–06 18 (plausibly pleading scienter where plaintiffs pointed to specific memos with specific data and 19 alleged that the executives actually read those memos); Wozniak v. Align Tech., Inc., 850 F. Supp. 20 3 The Opposition says “Bharadwaj attended weekly meetings beginning in July 2022 to discuss the 21 poor Expansion trend, which was so important that Defendants devoted over 100 top-level employees to help address the slowdown.” Oppo. 18:3–5. With this sentence, the plaintiffs 22 conflate and confuse their own separate allegations about weekly planning meetings, the metrics discussed, and the 100-person operation (Big Fish), none of which are connected in the SAC itself. 23 Tellingly, the Opposition does not cite the SAC for this proposition.

24 4 Because CW3 does not successfully establish his reliability or personal knowledge about Project Big Fish before September, the plaintiffs fail to allege that Big Fish addressed or even was formed 25 to address Paid User Expansion. For this reason, I am unpersuaded by their citations to cases where scienter was established by alleging that defendants instituted various projects in response 26 to issues or problems that they were hiding from the market. See Oppo. 12:27–13:17.

27 5 The plaintiffs also include allegations from CW4 and CW5, but they merely “confirm[]” what 1 2d 1029, 1043 (N.D. Cal. 2012) (failing to plead scienter by pointing to calls, meetings, and 2 reviews of “key metrics” without providing specific numbers or data that “contradict[ed]” the 3 defendant’s statements). 4 The “core operations” theory is also insufficient. Oppo. 18:1–11. In In re Apple Inc. 5 Securities Litigation, for example, Tim Cook said that Apple’s business in some emerging markets 6 was slowing but its business in China was not and also that iPhone sales were doing well, but soon 7 after, Apple missed its earnings guidance by $9 billion, mostly due to the Chinese iPhone market. 8 No. 19-CV-02033-YGR, 2020 WL 6482014, at *2–3 (N.D. Cal. Nov. 4, 2020). The plaintiffs 9 created a “strong inference” that Cook knew the Chinese iPhone market would materially affect 10 the business by pleading that the Chinese market constituted 20 percent of Apple’s revenue, Cook 11 frequently travelled to China, the Chinese economic issues were reported widely in the media so 12 Cook likely knew, and the iPhone was a core operation of Apple’s business. Id. at *9. Here, the 13 plaintiffs plausibly allege that Paid User Expansion is 90 percent of Atlassian’s business, but their 14 facts about Bharadwaj’s knowledge and especially the decline in this metric—including what the 15 slower growth was, when it occurred, who communicated it, who learned about it, and what 16 impact it had on the business—are insufficiently specific. And while it would be reasonable to 17 infer Apple’s CEO would know when his company was about to miss its revenue guidance by $9 18 billion, the plaintiffs here do not point to a similar statistic; the alleged $7 billion impact to the 19 stock price is a wholly different metric, and plaintiffs do not try to assert that Bharadwaj knew or 20 could know the stock price in advance. 21 Even assuming the plaintiffs plausibly alleged that Bharadwaj knew about the declining 22 Paid User Expansion metric, they again fail to plead “intentional or conscious misconduct” 23 because they do not plausibly allege that Bharadwaj engaged in an “extreme departure from the 24 standards of ordinary care” or knew or must have been aware that she was intentionally 25 misleading buyers or sellers. In re Alphabet, Inc., 1 F.4th at 701 (citations omitted). This is in 26 large part due to lack of a coherent theory about why she would mislead the market in mid- 27 September, knowing that the truth would come out at the beginning of November. CW3 asserts 1 Expansion, so that when Atlassian ultimately told the market, it could tell its shareholders and 2 investors that they were already working on the problem. SAC ¶ 82. As discussed above, he does 3 not sufficiently allege that Big Fish was intended to address declining Paid User Expansion. And 4 importantly, he provides no basis whatsoever for his reliability or personal knowledge of this as 5 Bharadwaj’s rationale. Cf. In re Plantronics, 2022 WL 3653333, at *14–15 (finding confidential 6 witnesses reliable and with personal knowledge about particular projects they worked on and 7 meetings they attended). 8 The theory itself also does not show the requisite mental state for scienter. The plaintiffs 9 point to In re Alphabet, 1 F.4th at 706–07, where the complaint alleged that Google executives 10 withheld information about severe cybersecurity vulnerabilities for six months, when the 11 (separate) Facebook-Cambridge Analytica data breach scandal was happening at the same time, 12 during which Facebook faced intense public and Congressional backlash. The Ninth Circuit found 13 plausible the inference that the decision to not tell the market something similar happened at 14 Google was intentional, “to ‘buy time’ by avoiding putting Google in the spotlight alongside [that 15 scandal] and at the time of heightened public and regulatory scrutiny.” Id. The “competing 16 inference” that the defendants were merely negligent in not disclosing the vulnerabilities for six 17 months was “not plausible.” Id. Here though, the alleged reason to wait to disclose is much 18 flimsier—Bharadwaj wanted a solution in place before announcing an issue—which is 19 substantially less “malicious,” see In re Sorrento, 97 F.4th at 643, than keeping quiet about “the 20 largest data-security vulnerability” in Facebook’s history to avoid regulatory scrutiny for as long 21 as possible, see In re Alphabet, 1 F.4th at 706. And critically, the “competing inference”—that 22 Bharadwaj held this mid-quarter call, not required by law, to try to provide open and regular 23 communication to shareholders during uncertain economic times and was merely negligent in 24 failing to specifically call out Paid User Expansion—is much stronger. Compare id. 25 Indeed, considering all the other allegations in the complaint, the theory of scienter is 26 simply weak “in light of other explanations.” Tellabs, 551 U.S. at 324; see also NVIDIA, 81 F.4th 27 at 940 (conducting a holistic review of scienter allegations). The plaintiffs again point to the 1 dramatically out of line with . . . prior trading practices” stock sales of two other defendants, and 2 || the defendants’ violation of their obligations under Item 303. But these allegations are 3 |} unconvincing for all the reasons laid out in my Prior Order, which the plaintiffs failed to address 4 || upon amendment. See Prior Order 22:22-25:4. 5 Finally, because the plaintiffs again failed to plead that any statements other than 6 || Bharadwaj’s were false or misleading, I do not need to address the plaintiffs’ arguments or 7 allegations about any other executives having the requisite scienter. 8 Accordingly, the plaintiffs failed to plead scienter. As for the statements above, it seems 9 || possible that the plaintiffs can clarify and replead their allegations, specifically concerning CW3 10 and their theory of liability, and so they may have leave to amend once more.°® 11 CONCLUSION 12 For those reasons, the motion is GRANTED, and the complaint is DISMISSED. The 13 plaintiffs may have one final opportunity to amend their allegations and state a claim. Any 14 amended complaint must be filed within 21 days of the Order. 3 15 IT IS SO ORDERED. a 16 Dated: August 13, 2024 3 17 . 18 ® Hiam H. Orrick 19 United States District Judge 20 21 22 23 24 25 26 27 © The parties agree that if the plaintiffs fail to plead a Section 10(b) claim, they also fail to plead a 28 Section 20(a) claim. See Mot. 25:17-19; Oppo. 25 n.14. For the same reasons, then, the Section 20(a) claim is DISMISSED with leave to amend.