City of Miami Fire Fighters' and Police Officers' Retirement Trust v. Cerence Inc.

• Court: District Court, D. Massachusetts
• Decided: March 25, 2024
• Docket: 1:22-cv-10321
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF MASSACHUSETTS

* CITY OF MIAMI FIRE FIGHTERS’ and * POLICE OFFICERS’ RETIREMENT * TRUST, Individually and on Behalf of All * Others Similarly Situated, * * Plaintiff, * Civil Action No. 22-cv-10321-ADB * v. * * CERENCE INC., SANJAY DHAWAN, and * MARK J. GALLENBERGER, * * Defendants. * *

MEMORANDUM AND ORDER

BURROUGHS, D.J. This is a federal securities class action lawsuit concerning alleged violations of Sections 10(b) and 20(a) of the Securities Exchange Act of 1934, 15 U.S.C. §§ 78j, 78t, and Rule 10b-5, 17 C.F.R § 240.10b-5, by Defendants Cerence Inc. (“Cerence”), Cerence’s former CEO Sanjay Dhawan and former CFO Mark Gallenberger. [ECF No. 37 (“Amended Complaint” or “Am. Compl.”)]. Plaintiff, on behalf of itself and others, claims that Cerence and its executives made false and misleading statements concerning demand for Cerence’s software and sales of its product. See [Id. ¶¶ 3–6]. Currently before the Court is Defendants’ motion to dismiss. [ECF No. 39]. For the reasons discussed below, the motion to dismiss is DENIED. I. FACTS AS ALLEGED For purposes of this motion to dismiss, the Court, as it must, “accept[s] as true all well-pleaded facts alleged in the complaint and draw[s] all reasonable inferences therefrom in the pleader’s favor.” A.G. ex rel. Maddox v. Elsevier, Inc., 732 F.3d 77, 80 (1st Cir. 2013) (quoting Santiago v. Puerto Rico, 655 F.3d 61, 72 (1st Cir. 2011)). A. The Parties Cerence is “an artificial-intelligence software company that operates almost exclusively in the automobile market . . . [selling] voice-operated virtual assistant software to . . .

approximately 60 automobile manufacturers, including Toyota, BMW, Daimler, and Ford.” [Am. Compl. ¶ 32]. Defendant Dhawan was the CEO and a board member of Cerence from June 7, 2019 to December 15, 2021. [Am. Compl. ¶ 29].1 As CEO, he ran Cerence’s day-to-day operations and was responsible for its financial performance. [Id.]. More specifically, he “personally reviewed and approved each [Cerence] contract for the sale of $1 million or more in licenses,” and “regularly spoke to investors and securities analysts regarding [Cerence]’s revenues.” [Id.]. Defendant Gallenberger was the CFO of Cerence from July 1, 2019 to February 7, 2022. [Am. Compl. ¶ 30].2 As CFO, he “personally reviewed and approved each [Cerence] contract for the sale of $1 million or more in licenses,” and “regularly spoke to investors and securities

analysts regarding [Cerence]’s revenues.” [Id.]. Both Dhawan and Gallenberger “directly participated in the management of Cerence’s operations, had direct and supervisory involvement in Cerence’s day-to-day operations, and had the ability to control and did control financial reporting and Cerence’s statements to investors.” [Am. Compl. ¶ 31]. Further, they “were involved in drafting, reviewing, publishing, and making

1 Dhawan was the CEO of Cerence’s predecessor, Nuance, from June 7, 2019 to October 1, 2019, when Cerence was spun off from Nuance. [Am. Compl. ¶ 29].

2 Gallenberger was the CFO of Cerence’s predecessor, Nuance, from July 1, 2019 to October 1, 2019, when Cerence was spun off from Nuance. [Am. Compl. ¶ 30]. [Cerence]’s statements to investors, including the false and misleading statements and omissions alleged” in the Amended Complaint. [Id.]. B. Cerence’s Business Cerence became a public company in 2019. [Am. Compl. ¶ 2]. From that date forward, Defendants told the public, investors, and analysts that the company was focused on growth and

had strong growth prospects. [Id.]. For example, on a February 11, 2020 earnings call for its first quarter as a public company, Dhawan said that Cerence was “laser-focused on profitably growing the business,” and Gallenberger stated that Cerence was “poised for strong revenue growth and profit performance for the fiscal year and beyond.” [Id. ¶ 33]. Shortly after, at Cerence’s first Analyst Day on February 18, 2020, Dhawan said the company was “very focused on growth,” and Gallenberger explained that one of the key takeaways that I want you to walk away from today with as it relates to Cerence is [how] the company historically has had a very good growth trajectory even in light of the fact that we’re part of an auto industry that has low-single-digit growth rates. We’ve been growing 10%, 15%. And so, there’s a secular tailwind that we have associated with the Cerence story. [Id. ¶ 34]; see also [id. (Gallenberger touting growth rates and the fact that company “expect[ed] that trend to continue,” with revenues nearly doubling by 2024). Gallenberger further said that Cerence had “‘visibility’ into future revenues . . . based on [a] ‘large amount of backlog’” (i.e. expected sales in the future). [Id.]. Analysts responded well to these comments and publicly noted confidence in the company’s prospects. [Id. ¶ 35]. To sell its software, Cerence enters “license agreements” with its customers, which give auto manufacturers the right to install Cerence’s software into their automobiles’ infotainment systems. See [Am. Compl. ¶¶ 3–4, 32]. The customer purchases an individual license for every automobile in which it installs the software, which allows the unlimited use of the software in that particular vehicle. See [id. ¶¶ 3–4]. In other words, a “license agreement” may give the auto manufacturer the right to many individual “licenses,” reflecting the fact that each vehicle requires its own license. See [id.] More than half of Cerence’s revenue comes from its license agreements. [Am. Compl. ¶ 35]. In 2021, for example, license revenue was 52% of the company’s total revenue. [Id.].

Relevant here, the license agreements come in two general forms. [Am. Compl. ¶ 37]. First, under variable license agreements, customers pay for each individual license when they actually install Cerence’s software in an automobile. [Id.]. When a variable license agreement is struck, the customer pays nothing up front. [Id.]. Rather, Cerence initially records expected revenue from the deal in its backlog (because it expects to get that revenue in the future in light of the agreement), and thereafter, as the customer puts the software into its vehicles, the customer sends a quarterly report to Cerence indicating the number of vehicles it produced and shipped with Cerence’s software. [Id.]. Based on those royalty reports, Cerence records revenue in the fiscal quarter during which the cars are produced. [Id. ¶ 38]. As it records that revenue, it also removes it from the backlog. [Id.]. It ultimately receives payment in the quarter after the

company records the revenue. [Id.]. Variable licenses generate steady quarterly revenue, [Am. Compl. ¶ 38], and are thus Cerence’s preferred license type, [id. ¶ 37]. Moreover, Plaintiff alleges that variable licenses are “an important barometer” of Cerence’s growth, [id. ¶ 38], apparently because the backlog shows expected future revenues. In contrast, the second type of agreements are fixed license agreements, which involve an upfront payment for a set number of licenses (i.e., cars that have Cerence’s technology). [Am. Compl. ¶ 39]. There are two types of fixed license agreements: “prepaid” and “minimum commitment” agreements. See, e.g., [id. ¶¶ 39, 67–68]. First, under “prepaid” agreements, customers purchase Cerence’s software in “bulk” (e.g., the right to multiple individual licenses at once), at a discounted price, and Cerence records revenue when the parties sign the agreement and Cerence is paid for all of the licenses in the agreement (e.g., “up front”—before the software is implemented in automobiles). [Am. Compl.

¶ 39]. Plaintiff alleges that Cerence offered only this type of fixed license deal for “most of the Class Period.” [Id.]. Although prepaid fixed license deals allowed Cerence to get cash upon signing the agreement, Plaintiff alleges that overreliance on them “harm[ed] future revenue and [Cerence]’s long term stability” because if the Company sells too many prepaid licenses, manufacturers will have an excess supply of licenses to work through and will not generate any new revenue for the Company until after they have used all of those licenses. In this way, prepaid deals are a drag on future revenue – and this effect is exacerbated by the discounted nature of the deals. [Am. Compl. ¶ 40]. Thus, “Cerence repeatedly underscored to investors the importance of decreasing prepays, or at least holding them flat as compared to historical levels,” [id. ¶ 41], and these fixed license deals were of “keen[] interest[]” to analysts who commented on and asked about them during earnings calls, [id. ¶ 42]. Second, under “minimum commitment” agreements, customers similarly agree to “purchas[e] a set number of licenses within a designated timeframe,” [Am. Compl. ¶ 68], typically at a discount, [id. ¶ 79], and Cerence records revenue “up front,” when the agreement is struck, [id. ¶ 68]. Unlike conventional prepaid deals, however, customers “pa[y] zero cash up front.” [Id.]. Thus, although Cerence reports an immediate positive financial result upon entering a minimum commitment deal, it is not actually paid until the customer puts the software into a vehicle, which could happen years later. See [id. ¶ 79]. As described in more detail below, Plaintiff alleges that during the Class Period, Cerence began offering these licenses without disclosing them to investors. [Id. ¶¶ 65–67]. Plaintiff further avers that they “cannibalized future revenue, and did so at a discount, yet yielded no cash up-front.” [Id. ¶ 68]. They were therefore “the worst kind of deal for the Company.” [Id.]. C. The Alleged Fraud

The class period began on November 16, 2020. [Am. Compl. ¶ 43]. At the time, there was a global semiconductor shortage due to the COVID-19 pandemic, and Plaintiff alleges that investors and analysts were concerned about the impact of the shortage on Cerence’s revenues and growth. [Id.]. Against this backdrop, in November 2020, Cerence announced its fiscal year and fourth quarter 2020 results. [Am. Compl. ¶ 45]. It reported that “Q4 revenue increased 21% from last quarter and [] 10% from the year prior, setting new quarterly and full year records,” and that it was “expecting another year of growth supported by a strong backlog and a solid pipeline of new business opportunities.” [Id.]. Investors and analysts responded positively to Cerence’s results and comments about revenues going forward, [id. ¶ 48], and Cerence’s stock increased by about

10% in one day, [id. ¶ 49]. In reality, Plaintiff alleges, Defendants had begun a scheme in which they were “personally instruct[ing] sales force personnel to dramatically increase their sales of fixed licenses,” [Am. Compl. ¶ 64], which hurt future growth opportunities. For example, “FE1,” an anonymous former sales manager in Japan until September 2021, [id. ¶ 69],3 reported that Dhawan “strongly pushed the prepayment deals,” [id. ¶ 70], “all over the world,” [id. ¶ 71], beginning in January 2020, and “direct[ed] sales personnel to sell more prepays, including by

3 Japan was a “very important geography” for Cerence. [Am. Compl. ¶ 71 n.2]. By Q2 2022, it accounted for nearly 30% of Cerence’s revenue. [Id.]. converting variable deals to prepays,” [id. §] 72], “every quarter” thereafter at least until FE] left Cerence in September 2021, [id. § 70]. Plaintiff alleges that the pressure to “push customers to buy prepays” came directly from Dhawan or from his direct reports. [Id. 72—73]. In the second quarter of 2021, for example, FE1 received calls and emails directly from Dhawan putting pressure on him to increase prepaid contracts. [Id. § 74]. FE1 also reported that Dhawan “required all Sales [teams] to convert all (typical) license contracts to Prepayment to get the cash.” [Am. Compl. § 77]. Specifically, under this “pull- forward scheme,” [id. § 79], “im the second month of each quarter, Dhawan added pressure” to convert variable contracts to prepay contracts “[d]espite knowing that prepays ‘crashed the future opportunity,” [id. § 77]. Overall, FE] explained that fixed license deals were used by “Dhawan and Gallenberger to ‘fill’ or ‘hide’ [a] ‘big gap’ between ‘Goal and Real’ performance.” [Id. 78}. In light of these allegations, Plaintiff alleges that the following statements from the November 2020 announcement were false and misleading:*

137 [Mr. Gallenberger:] “So going into fiscal year ‘21. I certainly would expect us to be within that range. If you recall, in the past, I have said that we’re sort of biased towards reducing prepays. However, that’s not always inside our control because we have our customers’ demand as well. And so that sometimes ebbs and flows. So I think going into FY ‘21, my view is that it would be down from fiscal year ‘20. But

* The parties both provided an exhibit identifying the challenged statements and their arguments as to why the statements are actionable, or not actionable. [ECF Nos. 40-1, 47-1]. The parties do not appear to dispute the list of challenged statements, but have emphasized different portions of them. See generally [ECF Nos. 40-1, 47-1]. The Court reproduces the challenged statements identified in the parties’ exhibits, showing what is emphasized by Plaintiff because it 1s required to draw inferences in its favor. °“AC refers to the paragraph in the Amended Complaint in which Plaintiff alleges that the referenced statement was materially false and misleading.

certainly, I think it’s going to still be within the range that we have seen over the last several years, which is low 40s to high — or low 40s to low 50s.” 139 [Mr. Dhawan:] “In particular, non-GAAP earnings per share at $0.61 per share was 88% above the midpoint of our guidance of $0.33 per share. The outperformance was primarily driven by great adoption of our products and services by the auto OEMs, the strong recovery in the auto market, coupled with the prudent financial controls we have implemented in recent quarters.” 143 [Mr. Dhawan:] “We had a stronger than expected start to the fiscal year as auto production continued to recover from the impact of Covid-19. Our 23% revenue growth, compared to the same quarter last year, reflects our strong competitive position enabled by our continued focus on innovation and speed of execution.”

[Mr. Dhawan:] “We expect continued year-over-year revenue growth in our second quarter as the auto industry recovers from Covid-19. However, our second quarter guidance accounts for the expected impact of semiconductor shortages on auto production in the first half of the calendar year.

According to IHS Markit’s current forecast, these shortages should be resolved by mid-year resulting in auto production growth of 13.7% for the 2021 calendar year. Overall the company is progressing well in all directions; introducing a steady stream of new products, winning new customers, successfully entering adjacent markets, and increasing revenue and profitability.” 144 [Mr. Dhawan:] “We expect continued year-over-year revenue growth in our second quarter as the auto industry recovers from Covid-19. However, our second quarter guidance accounts for the expected impact of semiconductor shortages on auto production in the first half of the calendar year. According to I Markit’s current forecast, these shortages should be resolved by mid-year resulting in auto production growth of 13.7% for the 2021 calendar year. Overall the company is progressing well in all directions; introducing a steady stream of new products, winning new customers, successfully entering adjacent markets, and increasing revenue and profitability.” For the next three quarters, Plaintiff alleges that “Defendants persisted in keeping investors under the illusion that Cerence continued to achieve its ‘record’ results and [was] build[ing] a ‘strong pipeline’ while holding prepays toward the lower end of its historical range.” [Am. Compl. ¶ 54]. First, on February 8, 2021, Cerence reported “strong” financial results and made “assurances about limiting prepay revenue.” [Am. Compl. ¶ 50]. Plaintiff specifically alleges that the following statements were false and misleading: Statement 146 [Mr. Dhawan:] “So in the last quarter, we booked 2 deals, which will contribute revenue to that line. So these are real bookings. But we don’t break them down from a number standpoint, but there are significant bookings, right? So that was good. At least it has started now, right? This current quarter, we have a strong pipeline, and I’m expecting more than 2 deals that will be booked this current quarter, that will also contribute towards that revenue line.” 148 [Mr. Gallenberger:] “So in summary, the business delivered another solid quarter. Our new products and technologies continue to enhance our competitive position, & which is enabling us to maintain our strong market share. And the business model continues to perform well as evidenced by our Q1 results and by raising our revenue 149 and profit metrics for the year. A combination of strength in our core business along with opportunities for new business from our new products and adjacent markets makes a bright future for Cerence. This concludes our prepared remarks, and now we will open it up for questions.” 151 [Mr. Gallenberger: ] “And so with that said, I think we’ll defer our commentary on bookings until our next earnings call. However, pipeline remains strong as we continue to expand our product offerings and also into adjacent markets. And that is naturally expanding the number of opportunities that we have going into our pipeline. And so that’s the color I can provide at this point.” 154 [Mr. Gallenberger:] “I think it’s just going to ebb and flow. There’s a few things to consider, right? One is prepays, we’re planning to be lower this year. So that’s & going to have to be reflected in that spread. The legacy — our legacy connected business, that’s a flattening effect this year. As we all know, that program is 155 coming to an end. So we’re not projecting any year- over-year growth for fiscal year ‘21 for that business versus fiscal year ‘20, it’s flat. So that has an impact as well.” [Mr. Gallenberger:] “So prepays can be lumpy. It’s more concentrated. So as we've seen in the past, they can go up and down and so forth. Last year, we did about $54 million in prepays, and we do expect prepays to be down this year. We — historically, we’ve been in that range of low 40s to low 50s, and I think we’re going to stay in that range. So last year, we were at the higher end of that range. This year, I would estimate we’ll probably be around in the middle of that range. And so that’s where we see it trending this year.” Analysts and investors responded well and raised their estimates for the company based on “statements that Cerence would ‘return to more normalized levels [of prepays] in ’21.’” [Am. Compl. § 52]. Four days later, Cerence stock reached an all-time high. [Id. § 53].

Second, on May 10, 2021, Defendants reported “revenue growth and strong profitability,’ [Am. Compl. § 54], and made the following statements that Plaintiff alleges were false and misleading:

159 [Mr. Dhawan: “Once again our results were ahead of expectations as we delivered the highest revenue for any quarter in the company’s history. Our core license business, in particular, performed better than expected as the global auto recovery takes shape and as our conversational AI and connected services expand into more car makes and models. We are proud to deliver both revenue growth and strong profitability.” [Mr. Gallenberger: | “I think last quarter we were around 10-ish or so in this quarter. So we were kind of a little bit below the run rate this quarter. We were above the run rate, really driven by I customer that we had accounted for over 50% of the entire fixed amount. So that kind of drove us towards the higher portion. So I think for the balance of the year, just because of what happened this quarter, we’re probably now going to be flat to up from the prior year. That would be my rough guess right now. But it’s difficult to give guidance on that line item. But last year, we did about 54 — about $54 million in total. And for the full — for the first 6 months, we’re at about $27-ish million, call it. So that kind of feels like we’re sort of on that same run rate as last year. And you’re right, we did say we’d probably be flat to down this year. J think now it’s probably going to be at the flat level and possibly up. It’s difficult to predict.” 163 [Mr. Dhawan: □ “Additional first half booking highlights include strategic customer wins with Hyundai and a prominent Japanese car company. It is important to note that bookings for our new applications products were approximately $33 million. This is a solid start for building the bookings foundation to support our $75 million of applications revenue expectations in our target 2024 model. We expect these bookings to deliver initial revenue before the end of the calendar year. You may have seen our press release last week announcing P97 networks as a partner in the ecosystem of our Cerence Pay application. This further solidify Cerence Pay in the market and will connect drivers to save seamless payments through the P97 mobile commerce platform currently available to more than 30% of the retail fuel sites across the U.S.” [Mr. Dhawan:] “I think we do think we will — we don’t guide our bookings number, Chris, as you know. But our internal goals definitely are to meet or beat last year’s bookings number. We do have the pipeline to achieve that, so it’s not just a pipe dream. It’s — there is — it’s supported by facts and a strong pipeline and a strong start to 03. Like I said in my prepared remarks, in last year, Q2 was a big — first half was bigger than second half. This year — this fiscal year, it all appears that it will be reversed. The second half will be a bigger bookings half as compared to first half. And the timing of that, as you all know, it’s totally driven by kind of OEMs and

their cycles and so on and so forth, right? So we don’t control that. It’s a big-- very important decision that OEMs have to take. But to summarize pipeline, is there and we surely hope that, that will meet or exceed last year’s $800 million number.” 164 [Mr.Dhawan:] “Overall, our pipeline continues to be strong, and our win rate remains extremely high. We had no competitive loss of note in the first half, and we won back an important multinational customer that had chosen a competitor when the business was still part of Nuance.”

[Mr. Dhawan:] “So that $30 million that we mentioned comes from 2 customers, 1 of them we announced, which was Xevo, that works with a number of OEMs; and the second one is a direct contract with another European OEM. And for the first one, its — the Cerence Pay is the main product. For the second one, there are multiple products included in that bookings, would be — which is the one with the OEM directly. The — in terms of the momentum, since the — since Q2 into Q3, we have 1 more OEM that has signed up with our — for our apps products. And that one is for the — travel guide product basically, and that is included in — we mentioned in our remarks that there is over $100 million of bookings in Q3 that has already happened, which is a very strong start. And there is one more OEM, which is going to launch the travel pro app. The pipeline remains strong. I can — there are probably half a dozen to 10-plus RFP/core conversations going on with a number of OEMs right now.”

[Mr. Dhawan:] “Yes. I think we do think we will — we don’t guide our bookings number, Chris, as you know. But our internal goals definitely are to meet or beat last year’s bookings number. We do have the pipeline to achieve that, so it’s not just a pipe dream. It’s — there is — it’s supported by facts and a strong pipeline and a strong start to Q3. Like I said in my prepared remarks, in last year, Q2 was a big — first half was bigger than second half. This year — this fiscal year, it all appears that it will be reversed. The second half will be a bigger bookings half as compared to first half. And the timing of that, as you all know, it’s totally driven by kind of OEMs and their cycles and so on and so forth, right? So we don’t control that. It’s a big-- very important decision that OEMs have to take. But to summarize pipeline, is there and we surely hope that, that will meet or exceed last year’s $800 million number.” In light of these representations, despite “concern” that “prepays came in higher than expected,” analysts “remained optimistic” about Cerence’s revenues in part because “the company continue[d] to focus on driving more Variable versus Prepays,” and also purportedly had “an exceptionally deep pipeline to support sustainable revenue growth.” [Am. Compl. ¶ 58 (internal quotations omitted)]. Third, on August 9, 2021, Defendants again “touted strong results for the third quarter of 2021.” [Am. Compl. § 59]. “Analysts reacted positively,” [id. § 61], and the stock price increased approximately 9% between August 6 and the close of trading on August 9, [id. §] 62]. But by the third quarter of 2021, “customers were balking at even discounted prepaid deals,” and Plaintiff alleges that, in response, Defendants began offering “customers a new type of fixed license deal — the ‘minimum commitment’ deal.” [Am. Compl. § 79]. As explained by FE1, “Dhawan first began discussing minimum commitment deals at the end of 2020 and [] by mid-2021 (more specifically, before the fourth quarter of 2021), minimum commitments were being offered to customers.” [Id. § 80]. According to FE1, “when the ‘customer hesitated’ to enter into a prepaid deal,” Dhawan or his direct report pushed sales personnel to propose minimum commitment deals, and these were “the deals that “Cerence pushed mainly’” during that time. [Id.]. With this context, Plaintiff alleges the following statements in August 2021 were false and misleading:

[Mr. Dhawan:] “Our multifaced growth strategy to deliver sustainable growth continues to play out. It starts with a strong core of leading conversational AI technology for the car and extends to new applications in adjacent markets. We continue to push the innovation envelope for our customers so that they can offer their customers the safest, most enjoyable experience inside the car, which also represents a seamless transition of their digital life from outside the car to inside the car.” 170 [Mr. Gallenberger: | “So in summary, we had another quarter of excellent financial performance. While we remain cautious in the near term due to the semiconductor shortages impacting the auto industry, we are continuing to benefit from the secular tailwinds caused by the digital transformation of the auto industry. Our long-term prospects remain strong and as demonstrated in our updated target model, and our focus on innovation and growth, while at the same time crafting a profitable business model will benefit the company and our shareholders well into the future.”

172 Mr. Gallenberger:] “Prepays, we historically, we typically see around $40 million to $55 million per year. And so, we’re going to see if we’re going to keep that those prepays, which we call fixed volume contracts within that historical range. So that’s what we built into the forecast as well.” After these announcements, in August and September of 2021, ““Cerence participated in a series of industry conferences, during which [Dhawan and Gallenberger] continued to tout the Company’s purportedly strong core business and sustainable revenue growth.” [Am Compl. § 63]. Plaintiff alleges that the following statements from those conferences were false and misleading: Statement 171 [Mr. Gallenberger:] “Prepays, we historically, we typically see around $40 million to $55 million per year. And so, we’re going to see if we’re going to keep that those prepays, which we call fixed volume contracts within that historical range. So that’s what we built into the forecast as well.” 174, “[A]n analyst asked ‘how [to] think about prepaids in the context of [the Company’s] 175 overall license forecast’ given that “the forecast assume[d] lower production’ of vehicles. Gallenberger again represented that the Company was committed to its historical prepaid revenue range, stating, “‘Prepays, historically, we typically see around $40 million to $55 million per year. And so we’re going to see if we're going to keep that — those prepays, which we call fixed volume contracts within that historical range.””® 178 [Mr. Gallenberger:] “[I]t’s a very difficult thing to predict. Even the experts are having a hard time predicting it and they’re constantly revising downwards some of their auto production estimates for this year and now even going into next year. So, it’s achallenge. It’s a definite challenge. The good news as you mentioned, we do — we don’t have 100% exposure, about one-third of our business is directly tied to what we get from our customers each quarter which we call quarterly royalty reports and we take revenue, our licensed revenue based upon these quarterly royalty reports. And that tells us how many cars did our customers ship in that given three-month window and based upon those reports, we obviously invoice in revenue based upon those reports. And that accounts for about a third of the company’s total revenue. The other two-thirds is a combination of fixed contracts — volume contracts, which some people call as prepays.”

® The parties do not provide a specific quote in their exhibits, but refer to the statement in § 174 of the Amended Complaint. See [ECF No. 47-1 at 17]. The language identified above is from ¥ 174 of the Amended Complaint.

In November 2021, when Cerence announced financial results for the fourth quarter and fiscal year 2021, Plaintiff alleges that investors began to “learn the truth.” [Am. Compl. § 87]. Specifically, on November 22, 2021, Cerence announced that fixed deals had “ballooned to $71 million, vastly exceeding the approximately $40 to $55 million range that Defendants continually reiterated.” [Id.]. By this time, “Cerence’s fixed license deals . . . [had] increase[d by] approximately 41% over the Company’s average historical amount,” and by Q4 2021, “minimum commitment deals accounted for 87% of fixed license revenue.” [Am. Compl. 82, 91]. In fact, “minimum commitment deals . . . in the first quarter of 2022 [] accounted for 100% of Cerence’s total fixed revenues.” [Id. § 82]. Further, Plaintiff avers that analysts and investors had been left “in the dark about what minimum commitments were and how much of them Cerence had sold,” [Am. Compl. § 93], because Defendants had “affirmatively mischaracterized ‘minimum commitment’ deals as ‘prepaid’ deals,” [id. § 94]. The only explicit reference to the minimum commitment deals, for example, was in a footnote of an investor presentation, described below, which provided only that “[f]ixed license revenue includes prepaid and minimum commitment deals,” [id. FJ 93, 190]. In light of this background, Plaintiff alleges that the following statements in November 2021 and shortly thereafter were false and misleading:

challenges our customers are facing due to semiconductor shortages. Our total company revenue grew 17% compared to the auto production growth of 9% over the same time-period, which is testament to the secular tailwinds, as well as, the innovative products and services we continue to bring to market.”

183 [Mr. Dhawan:] “We finished the year strong, especially considering the production challenges our customers are facing due to semiconductor shortages. Our total company revenue grew 17% compared to the auto production growth of 9% over the same time-period, which is testament to the secular tailwinds, as well as, the innovative products and services we continue to bring to market.” 185 [Mr. Gallenberger:] “Yes. So as I mentioned in my prepared remarks, it was driven by 2 larger than typical deals that we had closed in the quarter. And if I look at historically, we may have maybe one large deal in any given quarter, which tends to — as I mentioned before, it tends to swing those numbers around, and they’re difficult to predict, the size of those deals. And so it’s very unusual to have 2 happen at the same time. And that’s really what sort of drove the spike in Q4.

[Mr. Gallenberger:] “Like I said, typically, it’s one customer or it’s a series of customers on smaller deals, which, typically, would keep us in that $10 million to $15 million type of range. And so it was just that timing which drove it. I think if you look into fiscal ‘22, we do expect it to recede. We certainly don’t think it’s going to be a repeat of last year where we had a $71 million record. And if you look at our historical range, we’ve typically been in that low 40s to mid-50 type range. If you go back 3 or 4 years, that’s typically been the range from one year to the next.” 186 [Mr. Gallenberger:] “Like I said, typically, it’s one customer or it’s a series of customers on smaller deals, which, typically, would keep us in that $10 million to $15 million type of range. And so it was just that timing which drove it. I think if you look into fiscal ‘22, we do expect it to recede. We certainly don’t think it’s going to be a repeat of last year where we had a $71 million record. And if you look at our historical range, we’ve typically been in that low 40s to mid-50 type range. If you go back 3 or 4 years, that’s typically been the range from one year to the next.” 187 [Mr. Gallenberger:] “So because we were outside that range, that does put a little bit of a damper on growth rates for next year and possibly into fiscal ‘23 as well as those licenses get consumed. Right now, it’s hard to predict exactly where that number is going to be, but I would say that it’s going to be down $12 million, $13 million, $14 million or so year-over-year. So that kind of gets you back into our historical range, but at the higher end of the historical range. That’s what I’m anticipating.” 189 [Mr. Dhawan:] “So let me start, and then I’ll ask Mark to add in, Chris. So from my standpoint, no, nothing has changed. We stand by our guide for fiscal ‘24 and we feel good about it. As you heard in my prepared remarks, the new products contribute a lot towards that guide and 20% of our bookings, about $120 million was set from a bookings standpoint. We also have — we’ll be announcing some new aftermarket products. We have received an award letter for one of them already, which is not part of our bookings yet. They will be part of our fiscal quarter 1 bookings. So from that standpoint, I feel good that our new products are further contributing towards the contribution.” 190 : Revenue Growth Powered by License and Pro Services & 19] Priors era Ps] Nohara 1 & Sea ere □□□ eae 194 Semtteie ta $9.5M 0%

winsome “ToaiRewne: □□□ | soa | toe ‘Fixed license revenue includes prepaid and minimum commitment deals. ?Excluding a one-time accounting adjustment of $1.7 to correct an amortization schedule, year-over-year growth would have been 18% “I Fixed license revenue includes prepaid and minimum commitment deals.” “? Excluding a one-time accounting adjustment of $1.7M to correct an amortization schedule, year-over-year growth would have been 18%.” 193 [Mr. Gallenberger:] ““So we — you have to look at what we are sort of modeling internally for our Q1] revenues. And if you look at the Q4 revenues, we did have a large amount of fixed contract revenue, which we don’t expect to repeat to that same level. And so in last quarter, in Q4, if you look at the slides, we had $25 million of fixed revenue — fixed license revenue that is. And so that’s a pretty substantial number, and we don’t expect that to repeat. So when you factor that down, that number down quarter-over-quarter, that’s really what’s driving it. So we do expect variable licenses, which is most tightly coupled to auto production. We expect that number to increase sequentially.” [Mr. Gallenberger:] “Like I said, typically, it’s one customer or it’s a series of customers on smaller deals, which, typically, would keep us in that $10 million to $15 million type of range. And so it was just that timing which drove it. I think if you look into fiscal ‘22, we do expect it to recede. We certainly don’t think it’s going to be a repeat of last year where we had a $71 million record. And if you look at our historical range, we’ve typically been in that low 40s to mid-50 type range. If ou go back 3 or 4 years, that’s typically been the range from one year to the next.” 196 [Analyst:] “Just one last one, the 10-K’s out. We have some questions coming in around this from some of the investors with just around the amount of revenue that’s been recognized but unbilled has been increasing and we just see that in the 10-K. So, Mark, I don’t know if you have anything you can share around? Are you guys [indiscernible] (00:40:02) business practices or what’s leading to that?” [Mr. Gallenberger]: “Yeah. The biggest driver of that, Mark, is the fact that our fixed contract license revenue went up substantially this year versus last year. And with that, components of payment terms, some of those payment terms, if they’re not scheduled yet, then it becomes unbilled, natural movement up in that fixed

license contract’s line from last year to this year. This year, we’re forecasting the fixed contract revenue to actually come down. So, I would expect by the end of next or this fiscal year, a year from now, call it, that that number would probably start to come back down because of what we’re forecasting this year.” Plaintiff alleges that despite this optimism, the Defendants knew the risks of increasing fixed licenses. [Am. Compl. ¶ 87]. Gallenberger himself, for example, explained that “fixed license sales [were] ‘good for short-term, . . . [but] also . . . create[d] a little bit of pressure on our next year and sometimes the year after growth because we have to consume – or the customer has to consume those licenses.’” [Id. ¶ 88]. He said that the fixed deals “put a little bit of a damp around growth rates for [2022] and possibly into fiscal 2023 as well, as those licenses get consumed.” [Id.]. Ultimately, analysts expressed surprise and concern at Cerence’s results. [Am. Compl. ¶ 89]. For example, one explained that “‘prepaid license was up 40% sequentially,’ which ‘raises concerns about potentially pulling forward revenue.’” [Id. (emphasis in Am. Compl.)]. As a result of the disclosures, Cerence’s stock dropped by 20% between November 19 and November 22, 2021. [Id. ¶ 90]. About three weeks after the disclosure of financial results for the fourth quarter and fiscal year 2021, on December 15, 2021, Dhawan unexpectedly resigned as CEO of Cerence. [Am. Compl. ¶ 97]. Analysts and the market responded poorly to this announcement, attributing it to Cerence’s poor financial disclosures weeks earlier. [Id. ¶ 98]. Cerence’s stock declined 11% on

the day of the announcement. [Id. ¶ 99]. Then, on February 7, 2022, the first business day after the end of the class period (which is February 4, 2022), Cerence announced that it did $20.1 million in fixed deals during the first quarter of 2022, putting it on pace to surpass a “$71 million record” high from the prior year. As of the first quarter of 2022, the Company’s fixed license revenue had increased year-over-year by almost 100%, while its variable license revenue – i.e., its most important and valuable source of revenue – had declined year-over-year by over 40%. Along with its earnings announcement, the Company lowered its guidance for fiscal year 2022 by 9%, withdrew its guidance for fiscal year 2024, unexpectedly announced the “retirement” of Defendant Gallenberger, and for the first time disclosed the nature and impact of “minimum commitment” deals. [Id. ¶¶ 100–101]. The new CEO also highlighted that [Cerence’s] past sales of fixed licenses had “caused” a year- over-year “variable license [revenue] decline,” materially impairing the Company’s most valuable stream of revenue. Based on these troubling facts, [the new CEO] conceded that the Company needed now to define a “new vision and strategy for growth,” which would require the Company to “set[] a stronger foundation for long- term sustainable growth.” [Id. ¶ 102]. Further, Gallenberger said we are starting to experience the impacts of the larger-than-planned fixed license deals that we did last year and the year before, which is now creating a significant headwind to our variable license revenue growth. The reason is because those fixed licenses need to be consumed and netted out against the gross number of licenses consumed by customers each quarter. [Id. ¶ 103 (emphasis in Am. Compl.)]. In response to questions from analysts, Gallenberger also confirmed that all of the fixed licenses in Q1 2022 were minimum commitment deals, not “prepays.” [Id. ¶ 105]. As a result, Cerence’s stock dropped by 30% in one trading day between February 4 and 7, 2022. [Id. ¶ 110]. D. Dhawan and Gallenberger’s Alleged Scienter Regarding Dhawan and Gallenberger’s involvement and knowledge, according to FE1, they “approved all prepaid and minimum commitment deals over $1 million.” [Am. Compl. ¶ 83]. In addition, Cerence tracked consumption of prepaid licenses, and “Cerence’s royalty report team in Burlington (where Dhawan and Gallenberger were based) reviewed the [quarterly license] reports.” [Id. ¶ 84]. Plaintiff further alleges that “Dhawan and Gallenberger profited handsomely from their scheme” in two primary ways. [Am. Compl. ¶ 124]. First, both Dhawan and Gallenberger received payments under incentive compensation plans that were tied to Cerence’s revenue performance. [Am. Compl. ¶ 126]. Specifically, if

Cerence booked $385 million in revenues in fiscal year 2021, they would receive millions of dollars in additional compensation. [Id.]. Cerence booked $387.2 million in fiscal year 2021, just 0.5% more than that $385 million target, and as a result, both Dhawan and Gallenberger received additional stock and cash compensation. [Id. ¶¶ 127–128].7 Second, Plaintiff alleges that Dhawan and Gallenberger “enriched themselves” by selling stock at an inflated price. [Am. Compl. ¶ 130]. Specifically, Dhawan sold shares worth more than $24 million during the class period. [Id.]. The number of shares he sold during the class period was 5.7 times the number he sold between August 2019 and November 2020. [Id.]. Similarly, Gallenberger sold more than 50% of his vested stock at the beginning of the class period, receiving $1.1 million in proceeds, and the number of shares he sold during the class

period was 1.4 times the number he sold between August 2019 and November 2020. [Id. ¶ 132].8

7 Plaintiffs do not provide an exact amount of additional compensation, but state that

Dhawan obtained a substantial cash bonus and also achieved a total award of 382,781 shares, which were valued, on vesting, at a staggering $26,919,174. If Defendants had not pulled forward revenues, Cerence would have missed its guidance, and failed to meet [Dhawan and Gallenberger’s] compensation revenue target, which would have caused this award to be significantly reduced. [Am. Compl. ¶ 127]; see also [id. ¶ 128 (Gallenberger also received a “cash bonus” and “a total award of 77,646 shares, which were valued, on vesting, at $4,501,014” that, had Cerence missed the revenue target, “would have caused this award to be significantly reduced”)]. 8 Defendants argue that their trading was not suspicious because, among other reasons, trades were made pursuant to Rule 10b5-1 trading plans. [ECF No. 40 at 12–15; ECF Nos. 40-3, 40-4]. E. Cerence’s Post-Class Period Statements After the class period, Plaintiff alleges that Cerence and its new CEO, Stefan Ortmanns, confirmed the fraud. [Am. Compl. ¶ 111]. For example, Ortmanns said that (1) Dhawan approved fixed license deals, [id. ¶ 112]; (2) accelerating the “backlog” with fixed license deals was “not a good thing” and caused the “drop[] down [in] revenue,” [id. ¶¶ 113, 122]; and (3)

making up for the pulled forward deals could take 6 to 10 quarters, [id. ¶ 116]. Plaintiff also alleges that Ortmanns “admitted” that, going forward, Cerence “need[ed] to distinguish when it comes to fixed license deals” between the “two variants” (prepay and minimum commitment). [Id. ¶ 117]. Further, despite having previously conflated prepaid and minimum contract deals, see [id. ¶¶ 94, 114], Cerence’s disclosures after the class period “confirmed that for ‘Minimum Commitment’ deals, the Company received ‘no cash upfront,’ further demonstrating that Defendants’ characterizations of minimum commitments as ‘prepaid’ were misleading,” [id. ¶ 118]. II. PROCEDURAL HISTORY Plaintiff filed its initial complaint on February 25, 2022, [ECF No. 1], and its amended

and operative complaint on July 26, 2022. [Am. Compl.]. On September 9, 2022, Defendants filed the instant motion to dismiss. [ECF No. 39]. Plaintiff opposed on October 24, 2022, [ECF No. 47], and Defendants replied on November 23, 2022. [ECF No. 48]. III. LEGAL STANDARD “Section 10(b) of the Securities Exchange Act of 1934 forbids the ‘use or employ, in connection with the purchase or sale of any security . . . , [of] any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the [SEC] may prescribe as necessary or appropriate in the public interest or for the protection of investors.’” Tellabs, Inc. v. Makor Issues & Rts., Ltd., 551 U.S. 308, 318 (2007) (alterations in original) (quoting 15 U.S.C. § 78j(b)). In turn, “SEC Rule 10b-5 implements § 10(b) by declaring it unlawful,” id., “in connection with the purchase or sale of any security,” (a) To employ any device, scheme, or artifice to defraud, (b) To make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in light of the circumstances under which they were made, not misleading, or (c) To engage in any act, practice, or course of business which operates or would operate as a fraud or deceit upon any person. 17 C.F.R. § 240.10b-5. Therefore, [t]o survive a motion to dismiss under Rule 12(b)(6), a complaint alleging securities fraud under section 10(b) of the Exchange Act and Securities and Exchange Commission Rule 10b-5 must plead six elements: “(1) a material misrepresentation or omission; (2) scienter, or a wrongful state of mind; (3) a connection with the purchase or sale of a security; (4) reliance; (5) economic loss; and (6) loss causation.” Kader v. Sarepta Therapeutics, Inc., 887 F.3d 48, 56 (1st Cir. 2018) (quoting ACA Fin. Guar. Corp. v. Advest, Inc., 512 F.3d 46, 58 (1st Cir. 2008)).9

9 “Claims brought under section 20(a) of the [Securities Exchange] Act, 15 U.S.C. § 78t(a), are derivative of 10b-5 claims.” Hill v. Gozani, 638 F.3d 40, 53 (1st Cir. 2011). Section 20(a) provides that once a company has been found to have violated the Exchange Act’s substantive provisions, “[e]very person who, directly or indirectly, controls” the company “shall also be liable jointly and severally with and to the same extent as [the company] . . . unless the controlling person acted in good faith and did not directly or indirectly induce the act or acts constituting the violation or cause of action.” 15 U.S.C. § 78t(a). Here, Plaintiffs allege Section 20(a) claims against Dhawan and Gallenberger on the grounds that they acted as controlling persons of Cerence within the meaning of Section 20(a). . . . By virtue of their high-level positions, participation in and/or awareness of the Company’s operations, direct involvement in the day-to-day operations of the Company, and/or intimate knowledge of the Company’s actual performance, and their power to control public statements about Cerence, Defendants Dhawan and Gallenberger had the power and ability to control the actions of Cerence and its employees. [Am. Compl. ¶ 260]. Defendants do not contest that, if Plaintiff has stated a viable claim under Section 10(b), Dhawan and Gallenberger are liable as control persons under Section 20(a). See [ECF No. 40 at 20]. Thus, because the Court finds that Plaintiff has stated a claim under Section To survive a motion to dismiss, the complaint must contain “enough facts to state a claim to relief that is plausible on its face.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007). Further, because this case involves claims of securities fraud, Plaintiff must additionally satisfy the Federal Rule of Civil Procedure 9(b) standard for alleging fraud with particularity and

comply with the heightened pleading requirements imposed by the Private Securities Litigation Reform Act (the “PSLRA”). See Advest, Inc., 512 F.3d at 58. IV. DISCUSSION Defendants generally assert that the Amended Complaint should be dismissed because it fails to plead two of the elements of a § 10(b) claim: (1) a material misrepresentation or omission and (2) scienter. [ECF No. 40 at 9, 17]. The Court considers each argument in turn. A. Material Misrepresentation or Omission For a statement to be a material misrepresentation or omission, Plaintiffs must show “that defendants made a materially false or misleading statement or omitted to state a material fact necessary to make a statement not misleading . . . .” Geffon v. Micrion Corp., 249 F.3d 29, 34 (1st Cir. 2001). To show a material misrepresentation or omission under the heightened Federal

Rule of Civil Procedure 9(b) standard, Kader v. Sarepta Therapeutics, Inc., No. 14-cv-14318, 2016 WL 1337256, *13 (D. Mass. Apr. 5, 2016), plaintiff must provide “each [specific] statement alleged to have been misleading [and] the reason or reasons why the statement is misleading,” 15 U.S.C. § 78u-4(b)(1), including the “who, what, when, where, and how of the alleged fraud,” SEC v. Spivak, 194 F. Supp. 3d 145, 151 (D. Mass. 2016) (quoting United States ex rel. Ge v. Takeda Pharm. Co., 737 F.3d 116, 123 (1st Cir. 2013)).

10(b), see infra, it has also done so under Section 20(a). See Miller v. Sonus Networks, Inc., 636 F. Supp. 3d 245, 253 (D. Mass. 2022). The PSLRA, however, has “safe harbor” provisions that “sharply limit liability of companies and their management for certain ‘forward-looking statements,’ . . . when such statements are accompanied by appropriate cautionary language.” In re Smith & Wesson Holding Corp. Sec. Litig., 669 F.3d 68, 71 n.3 (1st Cir. 2012); see 15 U.S.C. § 78u-5.10 “[T]he

definition of a forward looking statement includes ‘a statement of the plans and objectives of management for future operations, including plans or objectives relating to the products or services of the issuer.’” Meyer v. Biopure Corp., 221 F. Supp. 2d 195, 203 (D. Mass. 2002) (quoting 15 U.S.C. § 78u-5(i)(1)(B)); see Carvelli v. Ocwen Fin. Corp., 934 F.3d 1307, 1324 (11th Cir. 2019) (“A forward-looking statement is what it sounds like—a prediction, projection, or plan.”). “On any motion to dismiss based upon subsection (c)(1), the court shall consider any statement cited in the complaint and any cautionary statement accompanying the forward-looking statement, which are not subject to material dispute, cited by the defendant.” 15 U.S.C. § 78u-5(e). Here, the allegedly fraudulent statements are set forth in twenty-three paragraphs of the

Amended Complaint. See [Am. Compl. ¶¶ 136, 138, 142, 145, 147, 150, 152, 153, 158, 160,

10 The safe harbor provision provides that in any private action arising under this chapter that is based on an untrue statement of a material fact . . . , a person . . . shall not be liable with respect to any forward-looking statement, whether written or oral, if and to the extent that . . . the forward-looking statement is . . . identified as a forward-looking statement, and is accompanied by meaningful cautionary statements identifying important factors that could cause actual results to differ materially from those in the forward-looking statement . . . . 15 U.S.C. § 78u-5(c)(1)(A)(i). 162, 167, 169, 171, 174, 176, 177, 182, 184, 188, 190, 192, 195].11 Defendants generally argue that the challenged statements are not actionable because they are (1) forward-looking and accompanied by cautionary language, (2) vague statements of corporate optimism that are not actionable, and/or (3) not supported by “particularized facts supporting [a] strong inference of [a]

scheme to increase fixed license contracts” (e.g. the statements were not misrepresentations or omissions). See generally [ECF No. 47-1]. 1. Forward-Looking Statements

“A forward-looking statement is what it sounds like—a prediction, projection, or plan.” Carvelli, 934 F.3d at 1324. A forward-looking statement is non-actionable if (1) it is “identified as forward-looking[] and [] accompanied by [a] meaningful cautionary statement,” or (2) a plaintiff does not show that the alleged misstatement “was made with actual knowledge.” 15 U.S.C. § 78u-5(c)(1). Here, Defendants argue that the statements discussed in ¶¶ 137, 144, 146, 148–149, 151, 154–155, 161, 163–164, 170–172, 175,12 and 186–187 are protected forward-looking statements under the PSLRA safe harbor. See [ECF Nos. 40-1, 47-1]. Specifically, they aver that ¶¶ 137, 154–155, 161, 175, and 186–187 are not actionable because they were, generally, statements that “the company intended to limit its use of discounted fixed contractual arrangements,” and “investors and analysts who followed Cerence were on notice that the company’s actual results

11 The parties have identified the statements by the paragraphs in which Plaintiff alleges that the statements were false or misleading. See [ECF Nos. 40-1, 47-1]. The Court adopts the parties’ numbering for clarity here.

12 In the parties’ arguments for ¶ 175, they refer back to arguments for ¶ 174, but there are no arguments for ¶ 174. See [ECF No. 47-1 at 17]. In any event, the statement in ¶ 174 appears to be forward looking, and thus the Court considers it here. may or may not conform to that goal.” [ECF No. 40 at 19]. Similarly, Defendants argue that the statements about “the company[’s] hope[] to close deals in its ‘pipeline’” at ¶¶ 146, 151, and 164 concerned future contract prospects and were therefore not actionable. [Id.]. First, the statements referenced at ¶¶ 154–155, 171, 172, 174, 175, 186, and 187 involve

“financial projections” that “squarely fall within the statutory safe harbor for ‘forward-looking’ statements concerning projected earnings and ‘future economic performance,’” In re Biogen Inc. Sec. Litig., 193 F. Supp. 3d 5, 40 (D. Mass. 2016), and are therefore not actionable. For example, in the statement discussed in ¶¶ 154 and 155, Gallenberger said the following: I think it’s just going to ebb and flow. There’s a few things to consider, right? One is prepays, we’re planning to be lower this year. So that’s going to have to be reflected in that spread. The legacy – our legacy connected business, that’s a flattening effect this year. As we all know, that program is coming to an end. So we’re not projecting any year-over-year growth for fiscal year ‘21 for that business versus fiscal year ‘20, it’s flat. So that has an impact as well. . . . So prepays can be lumpy. It’s more concentrated. So as we’ve seen in the past, they can go up and down and so forth. Last year, we did about $54 million in prepays, and we do expect prepays to be down this year. We — historically, we’ve been in that range of low 40s to low 50s, and I think we’re going to stay in that range. So last year, we were at the higher end of that range. This year, I would estimate we’ll probably be around in the middle of that range. And so that’s where we see it trending this year.” [ECF No. 47-1 at 8 (emphasis in original); see also ECF No. 41-23 at 4 (“I would like to remind you that this call may involve certain forward-looking statements. These statements are subject to risks and uncertainties as described in the press release preceding today’s call”); ECF No. 41- 22 at 7 (“Statements in this presentation regarding Cerence’s future performance, results and financial condition, expected growth, business and market trends, and innovation and our management’s future expectations, beliefs, goals, plans or prospects constitute forward‐looking statements within the meaning of the [PSLRA] . . . .”)]. The statements that Defendants were “planning” that “prepays” would “be lower this year,” or would stay in the “range of low 40s to low 50” million,” can be understood to relate to future revenue estimates, which could very well have been true or believed to be true.13 See [ECF No. 47-1 at 8]; cf. Sousa v. Sonus Networks, Inc., 261 F. Supp. 3d 112, 118 (D. Mass. 2017) (plaintiff did not adequately plead that “defendants were not in fact ‘comfortable’ with analyst’s . . . revenue estimates” despite telling “the market otherwise.”). Thus, the statements referenced at ¶¶ 154–155, 171, 172, 174, 175,

186, and 187 are protected by the PSLRA safe harbor. Many of the other paragraphs, on the other hand, refer to statements that are a mix of present facts and future projections, in which the present statements of fact are affected by the alleged fraud. See [Am. Compl. ¶¶ 137, 146, 151, 161, 163–164; ECF No. 47-1 at 1, 5, 7, 10– 12]. In these cases, the Court’s focus must be on the aspects of the statement are alleged to be false. The mere fact that a statement contains some reference to a projection of future events cannot sensibly bring the statement within the safe harbor if the allegation of falsehood relates to non- forward-looking aspects of the statement. The safe harbor, [the First Circuit] believe[s], is intended to apply only to allegations of falsehood as to the forward- looking aspects of the statement. . . . [Thus,] where the falsehood relates to a representation of present fact in the statement, it will not necessarily come within the statute’s safe harbor, even though the statement might also contain a projection of future financial experience. In re Stone & Webster, Inc. Sec. Litig., 414 F.3d 187, 213 (1st Cir. 2005). Put another way, “[t]he statutory protection will only apply where the claim of fraud is based upon the future projection.” In re Biogen Idec, Inc. Sec. Litig., No. 05-cv-10400, 2007 WL 9602250, at *10 (D. Mass. Oct. 25, 2007) (finding safe harbor did not apply to “statements of the future earnings or

13 The Court notes that its analysis would likely be different if discovery revealed that Defendants knew, based on actual data, for example, that Cerence’s fixed license revenue would not be in the historical range. The allegation, however, that Cerence tracked consumption of prepaid licenses, and “Cerence’s royalty report team in Burlington (where Dhawan and Gallenberger were based) reviewed the [quarterly license] reports,” [Am. Compl. ¶ 84], does not sufficiently suggest that Dhawan and Gallenberger actually knew that they would not achieve their projections. future financial posture” of the defendant where the alleged fraud related to “the failure to disclose the adverse clinical data”). Here, for example, in the statement referenced at ¶ 137, Gallenberger said the following: So going into fiscal year ‘21. I certainly would expect us to be within that range. If you recall, in the past, I have said that we’re sort of biased towards reducing prepays. However, that’s not always inside our control because we have our customers’ demand as well. And so that sometimes ebbs and flows. So I think going into FY ‘21, my view is that it would be down from fiscal year ‘20. But certainly, I think it’s going to still be within the range that we have seen over the last several years, which is low 40s to high — or low 40s to low 50s. [ECF No. 47-1 at 1]. Gallenberger’s statement that Defendants were presently “biased towards reducing prepays” is, according to Plaintiff’s allegations, untrue, see [id.], and calls into question the surrounding projections which may reflect that bias, see also, e.g., [ECF No. 47-1 at 5 (statement referenced at ¶ 146 that, in the “current quarter, we have a strong pipeline,” is false accepting Plaintiff’s allegations as true and calls into question accompanying financial projections); see also City of Austin Police Ret. Sys. v. Kinross Gold Corp., 957 F. Supp. 2d 277, 305 (S.D.N.Y. 2013) (“Although [] statements by [defendants] about” a construction schedule were “statements of opinion about the course of future events, they [were] not protected as forward-looking” because, “even if there had been [] cautionary language, [plaintiff] alleged sufficient facts to show that defendants had to appreciate [] that a material delay of the prior schedule . . . was likely, and that that schedule was no longer realistic. And it is well-settled that cautionary language cannot protect against the ‘omission of present fact.’”) (quoting Iowa Pub. Emps.’ Ret. Sys. v. MF Global, Ltd., 620 F.3d 137, 142 (2d Cir. 2010) (internal citations omitted)). Accordingly, the statements at ¶¶ 137, 146, 151, 161 and 163–164 are not protected by the PSLRA safe harbor.14 2. Statements of Corporate Optimism “The corporate puffery rule applies to loose optimism about both a company’s current state of affairs and its future prospects.” Fitzer v. Sec. Dynamics Techs., Inc., 119 F. Supp. 2d

12, 23 (D. Mass. 2000). “[C]laims of puffery . . . require a court to consider (1) ‘whether the statement is so vague, so general, or so loosely optimistic that a reasonable investor would find it unimportant in the total mix of information’ and (2) ‘whether the statement was also considered unimportant to the total mix of information by the market as a whole.’” In re Boston Sci. Corp. Sec. Litig., No. 10-cv-10593, 2011 WL 4381889, at *11 (D. Mass. Sept. 19, 2011) (quoting Brumbaugh v. Wave Sys. Corp., 416 F. Supp. 2d 239, 250 (D. Mass. 2006)). The Court must therefore look to whether the allegedly misleading statement “offers . . . details on which a reasonable investor would rely . . . .” Fitzer, 119 F. Supp. 2d at 26. Here, Defendants argue that the statements discussed in ¶¶ 137, 139, 143–144, 146, 148– 149, 151, 161, 163–164, 170, and 182–18315 are “[v]ague statement[s] of corporate optimism,”

and thus non-actionable corporate puffery. See generally [ECF Nos. 40-1, 47-1]; see also [ECF No. 40 at 20 (“statements that the company’s prospects were ‘strong’ . . . and the like are not actionable”); ECF No 48 at 10 (“Courts in this District routinely hold vague language such as ‘strong,’ ‘bright future,’ and ‘progressing well’ to be nonactionable corporate optimism”)]. Plaintiff responds that the statements here were not vague, but instead “highly specific, providing

14 Because the Court finds that the statements referenced in ¶¶ 144, 148–149 and 170 are non- actionable statements of corporate puffery, see infra, the Court need not decide whether they are protected by the PSLRA safe harbor.

15 Paragraphs referring to statements that the Court found are non-actionable above are not repeated here. precise consumption timelines, quantifying ‘prepay’ revenues, (which were actually from minimum commitments) and insisting that the Company would remain within its historical prepay revenue range when the opposite was true.” [ECF No. 47 at 12 (emphasis in brief)]. General statements that Cerence was in a strong financial position based on market

factors (e.g., competitive position in the market, adoption of products, recovery in the auto market, the company’s strategic decisions, etc.) did not provide details on which a reasonable investor would rely and are therefore non-actionable corporate optimism. See Boston Sci., 2011 WL 4381889, at *11 (statements that “sales force was ‘stable, large, experienced’ and ‘very successful’” was “loose optimism about both a company’s current state of affairs and its future prospects” and was therefore non-actionable puffery); Fitzer, 119 F. Supp. 2d at 23 (finding a statement that a company was “well-positioned” was non-actionable puffery). For example, a reasonable investor would not rely on the vague statement discussed in ¶ 139, in which Dhawan said the following: In particular, non-GAAP earnings per share at $0.61 per share was 88% above the midpoint of our guidance of $0.33 per share. The outperformance was primarily driven by great adoption of our products and services by the auto OEMs, the strong recovery in the auto market, coupled with the prudent financial controls we have implemented in recent quarters. [ECF No. 47-1 at 2]. The same is true for the statements referenced in ¶¶ 143–144, 148–149, 170, 182–183, which along with ¶ 139, are non-actionable statements of corporate optimism. [Id. at 2–4, 6, 14, 19–20]. In contrast, the statements referenced in ¶¶ 137, 146, 151, 161 and 163–164 are not merely vague corporate optimism. See [ECF No. 47-1 at 1 (¶ 137 referencing specific revenue projections and purported “bias toward reducing prepays”), 5 (¶ 146 referencing the purported “strong pipeline” and specific sales expectations), 7 (similar for ¶ 151), 10–12 (similar for ¶¶ 161, 163, 164)]. Rather, they contain statements about present facts, specific sales and/or sales opportunities, as well as specific revenue targets, that an investor or analyst might rely on, and are thus not protected by the corporate puffery doctrine. See In re Smith & Wesson Holding Corp. Sec. Litig., 604 F. Supp. 2d 332, 342 (D. Mass. 2009) (finding that “statements regarding the strength of past demand and ‘backlog’ orders are definite and important to a potential

investor” and “only purely forward-looking statements are entitled to protection as ‘mere puffery.’ Statements of present or historical fact are not mere ‘puffery.’” (quoting Brumbaugh, 416 F. Supp. 2d at 250)). 3. Material Misrepresentation or Omission For nearly all of the remaining challenged statements, Defendants broadly argue that “Plaintiff[s] [] failed to plead any particularized facts plausibly suggesting that Defendants were engaged in a scheme to increase Cerence’s use of fixed contract arrangements.” [ECF No. 40 at 18]; see generally [ECF No. 47-1]. In other words, Defendants argue that the challenged statements were not material misrepresentations or omissions. As an initial matter, several of the allegations here come from a confidential source, FE1. See, e.g., [Am. Compl. ¶¶ 70–77]. When allegations depend in part on confidential sources, a

court “evaluat[es], inter alia, [] the level of detail provided by the confidential sources, the corroborative nature of the other facts alleged (including from other sources), the coherence and plausibility of the allegations, the number of sources, the reliability of the sources, and similar indicia.” N.J. Carpenters Pensions & Annuity Funds v. Biogen IDEC Inc., 537 F.3d 35, 51 (1st Cir. 2008) (quoting In re Cabletron Sys., Inc., 311 F.3d 11, 29–30 (1st Cir. 2002)). Plaintiff alleges that, among other things, FE1 reported that Dhawan “strongly pushed the prepayment deals” beginning in January 2020 and “every quarter” thereafter until he left Cerence in September 2021, [Am. Compl. ¶ 70]; that Dhawan “direct[ed] sales personnel to sell more prepays, including by converting variable deals to prepays,” [id. ¶ 72]; that pressure to “push customers to buy prepays” came directly from Dhawan or from his direct reports, [id. ¶¶ 72–73]; that he (FE1) received calls and emails directly from Dhawan putting pressure on him to increase prepaid contracts, [id. ¶ 74]; and that Dhawan “required all Sales [teams] to convert all (typical) license contracts to Prepayment to get the cash,” [id. ¶ 77].

Meanwhile, the revenue share from fixed license agreements increased consistently during the class period, see, e.g., [Am. Compl. ¶ 48 (November 2020), ¶ 58 (May 2021), ¶ 100 (February 2022)], despite statements from Defendants suggesting they wanted and expected the share of fixed contracts to remain steady, and that fixed contracts would be limited in the future, see, e.g., [id. ¶ 136 (November 2020), ¶¶ 152–153 (February 2021), ¶ 171 (August 2021), ¶ 174 (same), ¶ 184 (November 2021)]. The Court finds that FE1’s specific allegations are corroborated by Cerence’s actual sales results, and that they collectively support an inference that Defendants were engaged in a scheme to increase Cerence’s use of fixed contract arrangements. Thus, the Court will consider which of the remaining challenged statements or omissions, if any, were misleading under the PSLRA in light of this alleged scheme.

The remaining challenged statements can generally be grouped in the following categories: • Statement that Cerence was biased toward limiting the number of fixed license agreements.16 • Statements that Cerence expected revenue from fixed license deals to remain in its historical range.17

16 [Am. Compl. ¶ 137]; see also [ECF No. 47-1 at 1 (¶ 137)].

17 [Am. Compl. ¶¶ 161, 185]; see also [ECF No. 47-1 at 10 (¶ 161), 21 (¶ 185)]. • Statements that Cerence had a strong pipeline.18 • Statements that Cerence was in a strong financial position based on market factors.19 • Statements allegedly omitting the existence of minimum commitment deals.20 The Court considers each category in turn. Statement that Cerence was biased toward limiting the number of fixed license agreements. On the November 11, 2020 earnings call for Q4 2020, Gallenberger made the following challenged statement: So going into fiscal year ‘21. I certainly would expect us to be within that range. If you recall, in the past, I have said that we’re sort of biased towards reducing prepays. However, that’s not always inside our control because we have our customers’ demand as well. And so that sometimes ebbs and flows. So I think going into FY ‘21, my view is that it would be down from fiscal year ‘20. But certainly, I think it’s going to still be within the range that we have seen over the last several years, which is low 40s to high — or low 40s to low 50s.

18 [Am. Compl. ¶¶ 146, 151, 163–64]; see also [ECF No. 47-1 at 5 (¶ 146), 7 (¶ 151), 11 (¶ 163), 12 (¶ 164)].

19 [Am. Compl. ¶¶ 159, 168, 189]; see also [ECF No. 47-1 at 9 (¶ 159), 13 (¶ 168), 24 (¶ 189)].

20 [Am. Compl. ¶¶ 178, 190–191, 193–194, 196]; see also [ECF No. 47-1 at 18 (¶ 178), 25 (¶ 190), 26 (¶ 191), 27 (¶ 193), 28 (¶ 194), 29 (¶ 196)]. With regards to the one statement/omission that Defendants do not explicitly argue was misleading, see [ECF No. 47-1 at 18 (citing [Am. Compl. ¶ 178])], they instead argue that Plaintiff misstates the transcript from an investor forum, [id.]. According to Defendants, Gallenberger “did not represent that all of Cerence’s fixed license deals were prepays.” [Id.]. The key portion of the challenged statement (which is from the transcript), that “the other two-thirds [of revenue, other than variable contracts,] is a combination of fixed contracts – volume contracts, which some people call as prepays,” [id.], is ambiguous. Although stating a “combination of fixed contracts” implies more than one type of contract, the language “which some call as prepays” seemingly refers to the entire group, and thus could be interpreted as implying that all fixed licenses were “prepays.” Drawing inferences in favor of the Plaintiff, as the Court must, the Court will consider whether omitting a specific reference to minimum commitment agreements was misleading in the context of this statement. [ECF No. 47-1 at 1]. Plaintiff alleges that “[i]t was materially false and misleading for Gallenberger” to claim that Cerence “was ‘biased toward reducing prepays[,]’” that “the amount of prepays was ‘not always inside our control because we have our customers’ demand as well,’” and to claim based on those statements that revenue from prepaid agreements would be lower in

fiscal year 2021, in the “‘low 40s to low 50’” million range when, in fact, “Defendants were directing sales personnel to increase fixed license revenue dramatically.” [Id.]. An investor with the knowledge that Defendants were emphasizing fixed license agreements and converting variable deals to fixed deals, see [Am. Compl. ¶¶ 70–74], could have believed that Gallenberger’s statement was (1) untrue, and (2) highly relevant to the question of whether Defendants’ projections were reasonable, see Construction Indust. & Labs. Joint Pension Tr. v. Carbonite, Inc., 22 F.4th 1, 8 (1st Cir. 2021) (“A fact is material if it is substantially likely ‘that the disclosure of the omitted [or misrepresented] fact would have been viewed by the reasonable investor as having significantly altered the “total mix” of information made available.’”). Although the portion of the statement about bias standing alone could very

well be true and not actionable (e.g., the company may have preferred to reduce prepays), when combined with Gallenberger stating the number of prepays was “out of [their] control,” “ebbs and flows,” “[his] view [was] that [the number of prepays would be down,” and that they would be “within the range that we have seen over the last several years,” other parts of the statement, considered together with allegations in the Amended Complaint concerning Dhawan, (e.g. Dhawan “strongly pushed the prepayment deals,” [Am. Compl. ¶ 70], “all over the world,” [id. ¶ 71], beginning in January 2020, and “direct[ed] sales personnel to sell more prepays, including by converting variable deals to prepays,” [id. ¶ 72], “every quarter” thereafter at least until FE1 left Cerence in September 2021, [id. ¶ 70]; see also [id. ¶¶ 72–74 (the pressure to “push customers to buy prepays” came directly from Dhawan or from his direct reports, and in the second quarter of 2021, for example, FE1 received calls and emails directly from Dhawan putting pressure on him to increase prepaid contracts); id. ¶¶ 77–79 (FE1 reported that Dhawan “required all Sales [teams] to convert all (typical) license contracts to Prepayment to get the

cash,” and under this “pull-forward scheme,” “in the second month of each quarter, Dhawan added pressure” to convert variable contracts to prepay contracts “[d]espite knowing that prepays ‘crashed the future opportunity,’”)], narrowly cross the line into actionability. Thus, drawing all reasonable inferences in Plaintiff’s favor, and recognizing that this is a close call and would be scrutinized closely for purposes of any summary judgment motion, the Court finds for purposes of this motion to dismiss that the statement referenced in ¶ 137 is actionable under the PSLRA.21 Statements that Cerence expected revenue from fixed license deals to remain in its historical range. In the statements referenced in ¶¶ 161 and 185, Gallenberger suggested that the higher number of fixed license agreements was due to one or two specific and unusual deals, and

21 Defendants argue that their statements were not misleading because “Cerence disclosed its revenue from such contracts in each quarter, and also disclosed its use of minimum commitment contracts in the quarters in which they were introduced,” Q4 2021 and Q1 2022. [ECF No. 40 at 18]. That Cerence disclosed revenues from fixed contracts on a quarterly basis does not necessarily mean that the Defendants’ statements about those fixed contracts were not false or misleading. See Smith & Wesson, 669 F.3d at 74 (where plaintiffs alleged that “strong sales numbers resulted in part from discounts used to pull orders from future quarters,” the sales numbers provided by defendant, “however accurate,” do not necessarily render statements not misleading if the “company’s cheerleading commentary . . . carr[ied] . . . an [incorrect] implied message that they reflected strong demand”). For example, in light of the alleged scheme to increase fixed licenses, and the actual increase of the same, at least the following statements could have been misleading (whether actionable or not for other reasons), even if investors knew the revenues from fixed license agreements: that (1) “23% revenue growth . . . reflects our strong competitive position enabled by our continued focus on innovation and speed of execution,” [Am. Compl. ¶ 142]; (2) that increased fixed revenue resulted from “1 customer that we had [that] accounted for over 50% of the entire fixed amount,” [id. ¶ 160]; and (3) attributing revenue growth to “our breadth of customers, products, and services,” [id. ¶ 167]. that Defendants expected revenue from fixed agreements to be in the historical range going forward. See [ECF No. 47-1 at 10 (¶ 161) (emphasis in brief) (“We were above the run rate, really driven by 1 customer that we had accounted for over 50% of the entire fixed amount . . . [and] we’re probably now going to be flat to up from the prior year.”), 21 (¶ 185) (quoting ECF

No. 41-37 at 14 (emphasis added) (Analyst: “what drove the above-average growth in prepaid in the September quarter because it was quite significant”; Gallenberger: “it was driven by 2 larger than typical deals that we had closed in the quarter. . . . And so it was just that timing which drove it. I think if you look into fiscal ‘22, we do expect it to recede. We certainly don’t think it’s going to be a repeat of last year where we had a $71 million record. And if you look at our historical range, we’ve typically been in that low 40s to mid-50 type range.”))]. Plaintiff alleges that these statements were materially false and misleading because Defendants were “attribut[ing] the increase in prepays to one customer,” [ECF No 47-1 at 10 (¶ 161)], or to “‘two larger than typical deals’ and ‘timing,’” [id. at 21 (¶ 185), “when, in reality,” they were pushing fixed agreements, [id. at 10 (¶ 161), 21 (¶ 185)]. Although this is again a close call, the Court

agrees. The statements could suggest to a reasonable investor that the increase in fixed licenses was an aberration, but knowledge of the alleged scheme would have supported an equal or stronger inference that the increase in fixed license agreements was the result of Defendants’ intentional scheme to promote them. See Construction Indust., 22 F.4th at 8 (“A fact is material if it is substantially likely ‘that the disclosure of the omitted [or misrepresented] fact would have been viewed by the reasonable investor as having significantly altered the “total mix” of information made available.’”). Thus, the Court finds that the statements referenced in ¶¶ 161 and 185 are actionable under the PSLRA. Statements that Cerence had a strong pipeline. Plaintiff argues that the statements referenced at ¶¶ 146, 151, 163, and 164, in which Defendants said that Cerence had a “strong pipeline” for future business, [ECF No. 47-1 at 5 (¶ 146), 7 (¶ 151), 11 (¶ 163), 12 (¶ 164)], were materially false and misleading because, “by virtue of the pull-forward scheme . . . Defendants

were depleting the pipeline of revenue to a highly material degree,” [ECF No. 47-1 at 5 (¶ 146), 7 (¶ 151), 12 (¶ 164)], or because “the Company’s rate of bookings . . . was not . . . sustainable [as] . . . [Defendants’] scheme [] cannibalized the Company’s future revenue,” [id. at 11 (¶ 163)]. None of these challenged statements were specific to revenue from fixed and/or variable licenses. See [ECF No. 41-23 at 14, 16; ECF No. 41-28 at 6, 9]. Rather, they referenced Cerence’s overall sales pipeline and ability to meet its overall future revenue projections. See generally [ECF Nos. 41-23, 41-28]. It is entirely possible, for example, that Defendants believed the pipeline was strong because there were customers with whom they had not entered license agreements that provided future revenue opportunities. In short, Plaintiff has not pleaded sufficient facts to support an inference that the overall pipeline for long term revenue was not

strong, or at the very least, that Defendants did not reasonably believe that it was, and thus the Court finds that the statements referenced in ¶¶ 146, 151, 163, and 164 are not materially false and misleading statements and are therefore not actionable under the PSLRA. Statements that Cerence was in a strong financial position based on market factors. For largely the same reasons that statements regarding the pipeline are non-actionable, see supra, the Court finds that statements about Cerence’s general strategy and growth were not materially false or misleading. [Am. Compl. ¶¶ 159, 168, 189]; see also [ECF No. 47-1 at 9 (¶ 159), 13 (¶ 168), 24 (¶ 189)] Statements allegedly omitting the existence of minimum commitment deals. Finally, Plaintiff alleges that statements referenced in ¶¶ 178, 190–191, 193–194 and 196 were false and misleading because, in general, Defendants had not disclosed that they were using minimum commitment deals and/or had not explained the nature of those deals to investors. See [ECF No.

47-1 at 18 (¶ 178), 25 (¶ 190), 26 (¶ 191), 27 (¶ 193), 28 (¶ 194), 29 (¶ 196)]. Three of the “statements” refer to a footnote in an investor presentation that explained “fixed license revenue includes prepaid and minimum commitment deals.” [ECF No. 47-1 at 25 (¶ 190), 26 (¶ 191), 28 (¶ 194)]. The footnote is true on its face and Plaintiff does not connect the footnote to statements that would render any omission (for example, failure to describe the nature of minimum commitment deals in the context of expected earnings) material in the eyes of an investor. Similarly, the statements referred to in ¶¶ 193 and 196 were about fixed licenses in general, not minimum commitments or prepays specifically, and thus are true and not misleading on their face. [ECF No. 47-1 at 27 (¶ 193), 29 (¶ 196)]. Finally, the statement referenced in ¶ 178 was about Cerence’s general business and, even if it is ambiguous, the

context of the statement does not suggest that any ambiguity was material to an investor—i.e., this general statement about Cerence’s business would be unlikely to be “viewed by [a] reasonable investor as having significantly altered the “total mix” of information made available.’” Construction Indust., 22 F.4th at 8. Thus, the statements referenced at ¶¶ 178, 190– 191, 193–194 and 196 are not materially false and misleading statements and are therefore not actionable under the PSLRA. B. Scienter

“[T]he PSLRA requires that complaints ‘state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind.’” Kader, 887 F.3d at 57 (emphasis in original) (quoting 15 U.S.C. § 78u-4(b)(2)(A)). “[P]laintiff[s] must show either that the defendants consciously intended to defraud, or that they acted with a high degree of recklessness.” Aldridge v. A.T. Cross Corp., 284 F.3d 72, 82 (1st Cir. 2002). Defendants are reckless under the PSLRA when they make “a highly unreasonable omission, involving not

merely simple, or even inexcusable, negligence, but an extreme departure from the standards of ordinary care, and which presents a danger of misleading buyers or sellers that is either known to the defendant or is so obvious the actor must have been aware of it.” City of Dearborn Heights Act 345 Police & Fire Ret. Sys. v. Waters Corp., 632 F.3d 751, 757 (1st Cir. 2011) (quoting Greebel v. FTP Software, Inc., 194 F.3d 185, 198 (1st Cir. 1999)). In determining whether a plaintiff has adequately pleaded scienter, “the court’s job is not to scrutinize each allegation in isolation but to assess all the allegations holistically.” Tellabs, 551 U.S. at 326. Moreover, the Supreme Court has held that “an inference of scienter must be more than merely plausible or reasonable—it must be cogent and at least as compelling as any opposing inference of nonfraudulent intent.” Id. at 314.

Here, Defendants argue that there is no strong inference of scienter because “Cerence was voluntarily disclosing its fixed license revenue[,] . . . and accordingly no Defendant could reasonably have expected that investors would be misled.” [ECF No. 40 at 10; id. at 17 (citing Liu v. Intercept Pharms., Inc., No. 17-cv-07371, 2020 WL 5441345, at *7 & n.57 (S.D.N.Y. Sept. 9, 2020))]. Plaintiff, on the other hand, avers that while Cerence was disclosing fixed license revenue, Dhawan and Gallenberger were, in reality, (1) “personally direct[ing]” an undisclosed “scheme to pull forward revenue and hide its worsening financial condition,” including by “converting variable deals into less favorable prepaids and minimum commitments,” [ECF No. 47 at 14]; (2) aware of the revenue consequences of that scheme, [id. at 20]; and (3) omitting key facts from their disclosures about fixed contracts (i.e., that “‘fixed’ contracts increasingly consisted of undefined ‘minimum commitment’ deals”), [id. at 17]. Although public availability of data bearing on the alleged fraud may “undermine[] an inference of scienter” in some circumstances, see Liu, 2020 WL 5441345, at *7 n.57, where, as

here, the data was (1) used to support a narrative (i.e., that the company was growing and had room to grow in the future) that is undermined by Defendants’ own actions, and/or (2) is itself misleading (i.e., the company did not disclose facts, like their prioritization of fixed license deals), the Court finds that there is an equally if not more compelling inference that Defendants acted with scienter, see In re Allaire Corp. Sec. Litig., 224 F. Supp. 2d 319, 323, 331 (D. Mass. 2002) (strong inference of scienter where “in addition to not disclosing [] bad news, [the defendant] issued a continued series of press releases, SEC filings, and earnings reports that painted a rosy picture for the future—a future that all concerned knew depended on [the product at issue’s] sales”); see also In re Plantronics, Inc. Sec. Litig,, No. 19-cv-07481, 2022 WL 3653333, at *2 (N.D. Cal. Aug. 17, 2022) (finding a strong inference of scienter in part because

“positive revenue results that Defendants touted during the Class Period” were driven by a new and undisclosed “sales practice [that] was intended to boost revenues and enable the Company to meet revenue guidance at the end of each quarter, and it involved offering discounts and other incentives to distributors” and “resulted in essentially borrowing sales or revenues from future quarters and was, therefore, unsustainable.”). Second, Defendants aver that FE1 is a confidential witness whose statements are entitled to little weight because they are self-serving to Plaintiff and are merely criticisms of Cerence’s management. [ECF No. 40 at 10–12]. As explained above, FE1’s specific allegations are corroborated by Cerence’s actual sales results, and they collectively weigh in favor of an inference of scienter. See supra. Third, Defendants argue that Dhawan and Gallenberger’s stock sales were not suspicious or unusual because shares were sold pursuant to prearranged trading plans and were widely

distributed before and throughout the class period. [ECF No. 40 at 13]. Further, Dhawan held $20 million in stock when he left the company, and Gallenberger’s equity position remained relatively constant during the class period. [Id. at 14–15]. Plaintiff does not respond with any specific allegations regarding Gallenberger, and with respect to Dhawan, argues that he sold 5.7 times the number of shares in the class period than in the prior same length period, with the majority of shares he sold being part of a trading plan adopted during the class period, rather than before. [ECF No. 47 at 19]. Where, as here, Defendants’ trades were made pursuant to stock plans, were distributed before and throughout the period, and did not immediately precede the end of the class period, the Court finds that their trades are neutral as to whether they support an inference of scienter. See Cabletron, 311 F.3d at 40 (“The vitality of the inference to be drawn

[from insider stock sales] depends on the facts, and can range from marginal to strong” (quoting Greebel, 194 F.3d at 197–98); see also Greebel, 194 F.3d at 198 (“the trading must be in a context where defendants have incentives to withhold material, non-public information, and it must be unusual, well beyond the normal patterns of trading by those defendants”). Fourth, Defendants argue that Dhawan and Gallenberger’s incentive plans were commonplace and thus not evidence of scienter. [ECF No. 40 at 16]. That an executive’s “compensation depended on the company’s earnings . . . alone [does] not and cannot be enough to establish scienter,” but when, for example, “financial incentives to exaggerate earnings go far beyond the usual arrangements of compensation based on the company’s earnings, they may be considered among other facts to show scienter.” Aldridge, 284 F.3d at 83. Moreover, when a “target [revenue] value [is] barely met—by 0.06%—[it can] suggest[] that some human manipulation may have been involved.” Washtenaw Cnty. Emps. Ret. Sys. v. Celera Corp., No. 10-cv-02604, 2012 WL 3835078, at *4 (N.D. Cal. Sept. 4, 2012). Plaintiff alleges that Cerence

barely exceeded the revenue threshold for Dhawan and Gallenberger to receive additional compensation by 0.5% ($387.2 million in revenue and a $385 million target). [Am. Compl. ¶¶ 126–127]. The Court finds this fact, coupled with the allegation that Dhawan and others “direct[ed] sales personnel to sell more prepays, including by converting variable deals to prepays,” [id. ¶ 72], weighs in favor of an inference of scienter. Fifth, Defendants aver that Dhawan and Gallenberger’s departures, shortly before the end of the class period (Dhawan) and on the last day of the class period (Gallenberger), [Am. Compl. at 1, ¶¶ 97, 100–101], do not support an inference of scienter, [ECF No. 40 at 16]. Specifically, Defendants argue that Dhawan took another CEO position and Gallenberger continued to work with Cerence in an advisory position, and an inference of wrongdoing from their departures is

conclusory. [Id.]. Though it is true that an abrupt resignation can weigh in favor of finding scienter, see In re Fibrogen, 2022 WL 2793032, at *25 (N.D. Cal. July 15, 2022) (“a resignation by itself is an insufficient indication of scienter, . . . [but] may be considered together with other allegations . . . because scienter is reviewed holistically”), here, where the additional context argued by Defendants minimizes an inference in favor of scienter, the Court finds Dhawan and Gallenberger’s resignations are neutral with respect to the scienter analysis. Finally, Plaintiff argues that Cerence’s post-class period “admissions” regarding the “revenue acceleration scheme” support an inference of scienter. [ECF No. 40 at 18]. Specifically, Plaintiff argues that Cerence admitted that Defendants “knew about” and should have “anticipated” the fixed license issue when they gave guidance during the class period, [ECF No. 47 at 18 (citing Am. Compl. ¶ 107)]. They further assert that Dhawan approved fixed license deals and that Defendants knew “that fixed license[] revenue had largely come ‘from [Cerence’s] backlog’—i.e., from cannibalizing existing variable contracts,” [id. (citing Am.

Compl. ¶¶ 104, 112, 118)]. Defendants respond that the post-class period statements “actually show that the company ha[d] been consistent in its disclosures” by “explain[ing] the benefits of [fixed] contracts,” while also “not[ing] that a countervailing concern is that ‘you end up with very lumpy [revenue growth],’ . . . [and that b]alancing th[ese] concern[s] . . . is a matter of business judgment.” [ECF No. 48 at 6]. The comments highlighted by Defendants do not speak to whether, at the time the challenged statements were made, Defendants were misleading investors regarding the company’s use of fixed license deals, how those deals were structured, and how they impacted the business going forward. See Carpenters Pension Tr. Fund for N. Cal. v. Allstate Corp., No. 16-cv-10510, 2018 WL 1071442, at *6 (N.D. Ill. Feb. 27, 2018) (finding scienter adequately pleaded in part based on post-class period statements because those

statements “d[id] not merely indicate that [defendant’s] assurances to investors . . . were incorrect in retrospect; it suggests that they were incorrect when made.”). The Court finds that the post-class period statements highlighted by Plaintiff support an inference of scienter because they suggest that Defendants knew, at the time they were making the challenged statements, that they were pushing fixed license deals over variable license deals, and this practice could negatively impact revenue in the future. Thus, although the individual facts discussed above may not be enough on their own to plead scienter, viewed “holistically,” Tellabs, 551 U.S. at 326, the Court finds that Plaintiff’s allegations provide a cogent and compelling inference that Defendants acted with fraudulent intent or with a high degree of recklessness by increasing the use of fixed license agreements at the expense of future revenue stability and growth, see id. at 314. V. CONCLUSION Accordingly, the motion to dismiss, [ECF No. 39], is DENIED because Plaintiff has

pleaded sufficient facts to support claims under Section 10(b) and 20(a) for the statements referenced in ¶¶ 137, 161 and 185 of the Amended Complaint. SO ORDERED.

March 25, 2024 /s/ Allison D. Burroughs ALLISON D. BURROUGHS U.S. DISTRICT JUDGE