In re Oracle Corporation Derivative Litigation

• Court: Supreme Court of Delaware
• Decided: January 21, 2025
• Docket: 139, 2024
• Status: Published

Opinion

IN THE SUPREME COURT OF THE STATE OF DELAWARE

IN RE ORACLE § CORPORATION § No. 139, 2024 DERIVATIVE LITIGATION § § Court Below: Court of Chancery § of the State of Delaware § § C.A. No. 2017-0337 § CONSOLIDATED

Submitted: October 23, 2024 Decided: January 21, 2025

Before SEITZ, Chief Justice; VALIHURA, TRAYNOR, LEGROW, and GRIFFITHS, Justices, constituting the Court en Banc.

Upon appeal from the Court of Chancery. AFFIRMED.

Joel Friedlander, Esquire (argued), Jeffrey M. Gorris, Esquire, David Hahn, Esquire, FRIEDLANDER & GORRIS, P.A., Wilmington, Delaware; Christopher H. Lyons, Esquire, Tayler D. Bolton, Esquire, ROBBINS GELLER RUDMAN & DOWD LLP, Wilmington, Delaware; Randall J. Baron, Esquire, David A. Knotts, Esquire, ROBBINS GELLER RUDMAN & DOWD LLP, San Diego, California; Gregory Del Gaizo, Esquire, ROBBINS LLP, San Diego, California for Plaintiffs Below/Appellants.

Elena C. Norman, Esquire, Richard J. Thomas, Esquire, Alberto E. Chávez, Esquire, YOUNG CONAWAY STARGATT & TAYLOR, LLP, Wilmington, Delaware; Peter A. Wald, Esquire (argued), LATHAM & WATKINS LLP, San Francisco, California; Blair Connelly, Esquire, LATHAM & WATKINS LLP, New York, New York; Melissa Arbus Sherry, Esquire, Christopher S. Turner, Esquire, Blake E. Stafford, Esquire, LATHAM & WATKINS LLP, Washington, D.C. for Defendants Below/Appellees Safra A. Catz and Lawrence J. Ellison.

Kevin R. Shannon, Esquire (argued), Berton W. Ashman, Jr., Esquire, POTTER ANDERSON & CORROON LLP, Wilmington, Delaware; Arthur H. Aufses III, Esquire, Jonathan M. Wagner, Esquire, KRAMER LEVIN NAFTALIS & FRANKEL LLP, New York, New York for Non-Party-Below/Appellee Special Litigation Committee of the Board of Directors of Oracle Corporation. SEITZ, Chief Justice:

Oracle Corporation acquired NetSuite Inc. in 2016. Following the acquisition,

Oracle stockholders filed a derivative suit against the Oracle directors and others.

They alleged that Lawrence Ellison, a co-founder of and substantial equity holder in

both companies, forced Oracle to overpay for NetSuite. After the Court of Chancery

denied the defendants’ motion to dismiss, the Oracle board formed a special

litigation committee (“SLC”) to review the plaintiffs’ derivative claims. The SLC

investigated and tried to settle the suit, but it eventually returned the case to the

plaintiffs to pursue. The parties litigated over five years and through the COVID-

19 pandemic. The Court of Chancery issued six pre-trial decisions and held a ten-

day trial. In its post-trial opinion, the court entered judgment for the remaining

defendants after concluding that the special committee negotiated the NetSuite

transaction untainted by Ellison’s or Oracle management’s influence.

On appeal, the stockholders contend that the court erred by: (1) allowing the

SLC to withhold its interview memos from the plaintiffs; (2) applying business

judgment review to a transaction involving an alleged controlling stockholder; (3)

employing the wrong legal standard when evaluating whether Ellison misled the

special committee by allegedly concealing his future NetSuite plans; and (4) finding

that Ellison’s alleged undisclosed future operational plans were immaterial to the

2 special committee’s evaluation and negotiation of the transaction. After careful

review, we affirm the Court of Chancery’s judgment.

I.

A.

We rely on the facts as found after trial.1 Oracle is a technology company

offering software, hardware, and cloud computing technologies. Its founder,

Lawrence Ellison, has served on its board of directors since 1977 and was Chief

Executive Officer (“CEO”) until September 2014. At that time, he became Chief

Technology Officer and Executive Chairman of the Board. Safra Catz and Mark

Hurd succeeded Ellison as co-CEOs. Hurd died in late 2019, at which point Catz

became the sole CEO.

In the 2000s, Oracle accelerated its growth strategy through acquisitions.

When Doug Kehring became Oracle’s Head of Corporate Development in 2006, he

implemented a standard framework for assessing potential acquisition targets, which

included a regularly updated dossier on select companies of interest.

NetSuite was one of these companies. Before the Oracle acquisition, NetSuite

was a technology company offering cloud-based enterprise resource planning

1 In re Oracle Corp. Deriv. Litig., 2023 WL 3408772 (Del. Ch. May 12, 2023) [hereinafter Post- Trial Opinion]. Except for their disclosure claim, the plaintiffs raise only legal errors on appeal. Thus, the facts are drawn from the Post-Trial Opinion, documents cited by the Court of Chancery, and the Court of Chancery record.

3 (“ERP”) and commerce software suites. Unlike Oracle, which primarily sold

customizable on-premises products to large customers, NetSuite for the most part

sold off-the-shelf cloud-based products to smaller customers. NetSuite’s co-

founder, Evan Goldberg, was a former Oracle employee. At the time of the

transaction, he served as NetSuite’s Chief Technology Officer and Chairman of the

Board.

Oracle’s interest in NetSuite started with Ellison. Ellison had long eyed

NetSuite as an Oracle acquisition target. He regularly made his views known to

“anyone who would listen” and “even to people who wouldn’t.”2 In February 2015,

Ellison met with Catz and Hurd to discuss a potential acquisition. Although Hurd

was supportive, Ellison was not convinced that the timing was right, a sentiment

echoed by Catz. Ellison was concerned that NetSuite was trading at such a high

premium that the acquisition would be dilutive to Oracle’s earnings. Ellison was

also concerned that the acquisition would distract Oracle management and confuse

the technology marketplace as Oracle transitioned its product offerings from on-

premises to cloud-based software. Unlike on-premises software, which is installed

and maintained “on the premises” of the customer, cloud-based software is hosted

and maintained off-premises by a third-party. And Oracle’s own cloud-based ERP

2 App. to Appellants’ Opening Br. at A1266 [hereinafter A__] (Tr. 1664:6–24); A1345 (Tr. 1980:4– 15).

4 product, Fusion, was just beginning to gain traction in the market after a decade of

development. To avoid upsetting the delicate transition period, Oracle did not

pursue an acquisition of NetSuite in early 2015.

Oracle did not, however, lose interest. Later that year, NetSuite failed to meet

its bookings growth rate projections. NetSuite attributed the flattening growth to its

pursuit of customers who required significant software customization, which

produced non-recurring and low-margin revenue and slowed down implementation

time. These customers were often larger in size and required new functionalities to

service their scale.

Ellison believed that NetSuite could not compete against Oracle, whose

primary customer base consisted of large enterprise customers. In October 2015,

Ellison met with NetSuite leadership – including Goldberg, CEO Zachary Nelson,

and President Jim McGeever – to discuss his concerns. During the meeting, Ellison

advocated for a new growth strategy focused on designing software functionalities

for specific industries and subindustries in the small and medium business (“SMB”)

market. This resulted in Project Atlas, later renamed SuiteSuccess. SuiteSuccess is

a pre-built software solution that leaves room for customization only during the “last

mile” of implementation. Its target customers were cost-conscious businesses with

little need for extensive customization. With SuiteSuccess, NetSuite would shift

5 away from low-margin implementation fees, reduce implementation time, and boost

customer satisfaction.

While NetSuite developed SuiteSuccess, it continued to develop upmarket

financial functionalities.3 McGeever continued to push for these functionalities

because he believed “the fundamental thing of [NetSuite’s] value proposition was

that we needed to be very strong in pure financials . . . .”4 The Court of Chancery

nevertheless found that, despite NetSuite’s continued development of these up-

market financial functionalities, NetSuite had “tempered its indiscriminate move

upmarket when it began developing [SuiteSuccess].”5

Shortly before Oracle’s January 14–15, 2016, board retreat, Ellison told Catz

that he believed “the time is now” for Oracle to buy NetSuite.6 At the board retreat,

Kehring presented three potential acquisition targets, one of which was NetSuite. It

was the first time that Oracle management discussed a potential acquisition of

NetSuite with the Oracle board. Before the NetSuite presentation, Ellison left the

room and recused himself from the discussion. After the presentation, the board

3 A1030 (Tr. 724:6–725:2). 4 A848 (Tr. 79:6–8). 5 Post-Trial Opinion at *30. 6 A1345 (Tr. 1982:10–11).

6 directed Catz and Hurd to gauge NetSuite’s interest in a potential acquisition. The

board instructed the co-CEOs not to discuss price.

Four days later, on January 19, 2016, Catz had dinner with NetSuite’s CEO

Zachary Nelson. Catz asked Nelson whether NetSuite would be open to being

acquired by Oracle. Nelson replied that any acquisition would have to involve a

“Concur-type multiple,” referring to the high revenue multiple from SAP’s Concur

acquisition in 2014.7 The Court of Chancery found as a factual matter that Catz did

not discuss price or make an offer during the dinner meeting.8

On January 27, 2016, Goldberg called Ellison and asked whether Oracle’s

pursuit of NetSuite was meant as punishment, which Ellison denied. Instead, Ellison

framed the transaction as strategically advantageous to both companies. But

Goldberg remained skeptical and was worried about NetSuite’s and his own

independence post-acquisition. Ellison assured Goldberg that, if acquired by Oracle,

NetSuite would become a Global Business Unit, and Goldberg would report directly

to Hurd or Ellison. Ellison also shared that he “was going to stay neutral and out of

the discussions and out of the voting.”9 This was Ellison’s last conversation with

7 A1115 (Tr. 1062:5–1063:17). 8 Post-Trial Opinion at *6. 9 A1308 (Tr. 1835:23–24).

7 Goldberg until the transaction closed in November 2016. Ellison did not disclose

this phone call to the board or the later-formed special committee.

B.

On March 18, 2016, the Oracle board, with Ellison recused, formed a special

committee (“Special Committee”) to negotiate a potential transaction with NetSuite.

The Special Committee was composed of Renee James, Leon Panetta, and George

Conrades, all of whom reported no conflicts and were recommended by counsel.

The Special Committee was fully empowered to control the transaction.

Over the course of the next seven months, the Special Committee met 15 times

to consider the transaction. On April 8, 2016, the Special Committee held its first

meeting, during which it elected James as the chair and engaged Skadden, Arps,

Slate, Meagher and Flom LLP as counsel. During this meeting, members of Oracle

management, including Catz and Kehring, discussed the strategic rationale for the

transaction. On April 19, 2016, the Special Committee engaged Moelis & Company

as financial advisor. The Special Committee chose Moelis over the other finalist,

Evercore, in large part because Moelis raised alternatives to the NetSuite transaction

and demonstrated its ability to challenge Oracle management.

On May 5, 2016, Moelis held an all-day diligence session attended by

Skadden, James, Kehring, and two other Oracle employees. Moelis presented its

findings that, although Fusion was successful with enterprise and near-enterprise

8 customers, it was not as well received by SMB customers. Moelis found that, in

contrast, NetSuite was strong where Fusion was weak. Although NetSuite did sell

to larger customers, NetSuite’s target customer base was the middle market, which

has “affectionately been referred to as the Fortune 5 million.”10

James found the presentation persuasive. On May 13, 2016, the Special

Committee met with Skadden without Oracle management. James reported that,

“after listening to the presentation [from the May 5 due diligence session], I found

[Oracle and NetSuite] to be very complementary.”11

A week later, the Special Committee met with Skadden, Moelis, and Oracle

management (including Catz and Kehring). Catz and Kehring highlighted the

necessity for Oracle to invest further in ERP software and NetSuite’s ability to fill

that niche. Oracle management recommended that the Special Committee move

forward with the acquisition and left the meeting. Subsequently, Moelis confirmed

the view that NetSuite could boost Oracle’s ability to compete in the cloud ERP

space. The Special Committee determined that (1) acquiring NetSuite would be

highly beneficial to Oracle, and (2) it was the right time to make an initial offer.

On May 23, 2016, Oracle representatives met with Moelis to discuss a

preliminary financial model for NetSuite. Three days later, Skadden sent the Special

10 A1022 (Tr. 692:5–6). 11 A1138 (Tr. 1155:9–15).

9 Committee-approved recusal rules to Oracle management, who forwarded them to

Ellison. On the same day, Catz spoke with Goldberg, who, despite expressing his

unwillingness to sell NetSuite, understood his responsibilities to NetSuite and

continued to push Oracle for a high price. Goldberg reported this phone

conversation to the NetSuite board, but Catz did not report it to the Oracle Special

Committee.

On May 27, 2016, the Special Committee met with Skadden, Moelis, and

Oracle management (including Catz and Kehring) to discuss NetSuite valuations.

Oracle management presented first. They shared discounted cash flow analyses,

precedent transaction multiples, and a home-grown incremental model designed by

Oracle’s corporate development team. The incremental model “reflect[ed] the

incremental revenue and expenses as a result of owning the target by

which[,] . . . over a five-year-horizon, [Oracle could] accomplish.”12 It did not

consider post-acquisition operating plans or overhead costs that would be allocated

to NetSuite for accounting and budgeting purposes. Based on all the NetSuite

valuations, Oracle management recommended an opening bid of $100 per share.

Moelis representatives reviewed the models and raised questions with Oracle

management. They ultimately concluded that the models were reasonable. After

12 A986 (Tr. 548:12–15).

10 Oracle management left the meeting, Moelis presented their valuations and shared

public market price targets, revenue multiples, and precedent transactions. Moelis

advised that the opening bid should be no lower than $100 per share to secure

NetSuite’s interest. The Special Committee decided to submit an opening bid of

$100 per share, which Moelis communicated to NetSuite’s financial advisor on June

1, 2016.

On June 7, 2016, NetSuite responded with a counterproposal of $125 per

share. The next day, the Special Committee raised its bid to $106 per share, which

left room for Oracle to negotiate below its $110 ceiling. On June 11, 2016, NetSuite

lowered its price to $120 per share but messaged that a lower price was unlikely.

Two days later, the Special Committee met with Skadden, Moelis, and Oracle

management (including Catz and Kehring), during which Oracle management

recommended against countering. At this point, the Special Committee, was

“prepared to let the possibility of acquiring NetSuite die.”13

A few weeks later, in late June 2016, NetSuite’s financial advisor, Qatalyst

Partners, called Oracle’s CFO, Stuart Goldstein, to share that NetSuite may be more

flexible on price and to move the deal forward. The Special Committee met with

Skadden, Moelis, and Oracle management (including Catz and Kehring). Catz and

13 A1142 (Tr. 1173:6–9).

11 Kehring wanted more due diligence to assess NetSuite’s forthcoming Q2 financial

results. After Oracle management left the meeting, the Special Committee decided

to conduct additional due diligence, which “signaled toughness on price and a lack

of anxiety to re-start negotiations.”14

On July 6, 2016, NetSuite’s CFO, Ron Gill, presented NetSuite’s Q2 financial

results to Oracle management and Special Committee Chair James. While NetSuite

had exceeded its earnings projections, its SaaS bookings subscription revenue was

low. Catz believed that the Special Committee could use the decreased subscription

revenue to “sow some negative thoughts,” which “would help Oracle negotiate a

better price.”15 After Catz reported the diligence call to the Special Committee, and

following a series of additional diligence meetings, the Special Committee decided

to remain firm at $106 per share.

On July 12, 2016, NetSuite lowered its price to $111 per share. Catz

recommended that Oracle split the difference and counter at $108.50 per share. The

next day, the Special Committee communicated its best and final offer of $109 per

share, which NetSuite accepted the same day. On July 15, 2016, Oracle and NetSuite

entered an exclusivity period. The transaction cleared additional due diligence by

Oracle and antitrust review by the United States Department of Justice.

14 Post-Trial Opinion at *13. 15 A1218 (Tr. 1477:19–21).

12 The parties structured the transaction as a tender offer in which a majority of

NetSuite shares unaffiliated with Ellison, NetSuite’s officers, and its directors were

required to support the transaction. The parties set the tender deadline for September

15, 2016, but extended it twice, largely due to uncertainty about the vote by

NetSuite’s largest unaffiliated stockholder, T. Rowe Price. T. Rowe Price believed

that $109 per share was too low to tender and pushed Oracle to increase its offer to

$133 per share. Nevertheless, the Special Committee refused to budge on price.

When the twice-extended deadline expired on November 4, 2016, 53.2% of

NetSuite’s unaffiliated shares tendered. The acquisition closed three days later.

C.

On May 3, 2017, an Oracle stockholder filed a derivative suit in the Court of

Chancery against Ellison, Catz, members of the Special Committee, and Oracle. The

plaintiff alleged that “Ellison took advantage of Oracle’s need for [a] cloud-based

acquisition and used Oracle’s money to overpay for NetSuite for the benefit of

himself and his family, receiving nearly $4 billion from the Transaction – a massive

return on Ellison’s initial $125 million investment in NetSuite.”16 The plaintiff also

alleged that Catz was Ellison’s “hand-selected consigliere” and that the Special

16 Verified S’holder Deriv. Compl. ¶ 1, Post-Trial Opinion, Docket No. 1 [hereinafter Ch. Dkt. __].

13 Committee was “flanked by Oracle’s senior management, to whom the Special

Committee and its advisors deferred.”17

After consolidating a related action, the Court of Chancery denied Ellison’s

and Catz’s motion to dismiss.18 The plaintiffs then voluntarily dismissed claims

against all defendants except Ellison and Catz.

On May 4, 2018, the Oracle board formed an SLC to investigate the plaintiffs’

claims. The SLC was fully empowered to control the litigation.19 It was chaired by

William Parrett with members Charles Moorman and Leon Panetta.20 It retained its

own independent counsel and financial advisor. The Court of Chancery stayed the

litigation for thirteen months while the SLC investigated the claims and explored

whether a settlement was feasible through non-binding mediation.

By mid-July 2019, the plaintiffs became frustrated that the SLC “ha[d] not yet

submitted its report or provided notice to Lead Plaintiff of its position respecting the

17 Id. ¶ 3. 18 In re Oracle Corp. Deriv. Litig., 2018 WL 1381331, at *23 (Del. Ch. Mar. 19, 2018). 19 In re Oracle Corp. Deriv. Litig., 2019 WL 6522297, at *7 (Del. Ch. Dec. 4, 2019) [hereinafter December 2019 Opinion]. The SLC was authorized to “(i) take all actions necessary to investigate, analyze and evaluate all matters relating to this lawsuit and the claims made in the action, and (ii) take any actions that the SLC deems to be in the best interests of the Company in connection with this lawsuit and any related matters.” Id. (citation omitted). 20 Id.; see also J.A. to Appellees’ Answering Br. at B3491–93 [hereinafter B__] (Mot. to Stay by the Special Litig. Comm. of the Bd. of Dirs. of Nominal Def. Oracle Corp. ¶¶ 12–18). Although Panetta also served on the Special Committee, the plaintiffs did not challenge his independence or fitness to serve on the SLC. Appellees’ Answering Br. at 7 (SLC).

14 derivative claims” and “ha[d] not communicated with the Lead Plaintiff ever since

the hearing on June 7, 2019.”21 The Court of Chancery permitted the plaintiffs to

file an amended complaint (“First Amended Complaint”). The First Amended

Complaint named Ellison, Catz, Hurd, the Oracle board, the Special Committee

(including Panetta, who now served on the SLC), Oracle (as nominal defendant),

Goldberg, and Nelson. Count One alleged breach of fiduciary duty against the

Oracle defendants. Count Two alleged aiding and abetting against the NetSuite

defendants.

On August 15, 2019, the SLC’s counsel informed the Court of Chancery that

“it appears unlikely that a settlement can be reached in the near future” and that “the

SLC has determined that the Lead Plaintiff should be allowed to proceed with the

derivative litigation on behalf of Oracle.”22 According to counsel, the SLC believed

that “the critical legal issue of whether the challenged NetSuite acquisition will be

reviewed under the entire fairness standard would not be resolved prior to trial,

thereby posing risks to both plaintiff and defendants.”23 As such, the SLC decided

to return the case to the plaintiffs.

21 Lead Pl.’s Mot. to Lift the Stay for the Limited Purpose of Filing Mot. for Leave to File Verified Am. Deriv. Compl. ¶ 7 (Del. Ch. July 18, 2019), Ch. Dkt. 134. 22 A1916 (Letter from Potter Anderson & Corroon LLP to The Honorable Sam Glasscock III, Aug. 15, 2019). 23 A1917.

15 D.

After the plaintiffs took over the case, they subpoenaed the SLC and its

counsel. They requested virtually all documents and communications concerning

the SLC’s investigation, including any draft or final report prepared by the SLC.

The plaintiffs argued, in essence, that all documents held by the SLC are relevant

and must be produced. In resisting the subpoena, the SLC and its counsel argued

that, because the SLC returned the litigation to the lead plaintiff, neither the court

nor the parties needed to “address or evaluate the SLC’s independence,

investigation, or determination,” rendering discovery “inappropriate and

unnecessary.”24 They also argued that the subpoena improperly sought the

production of privileged material. The SLC and its counsel also claimed that they

could not produce any documents received from third parties without the latter’s

authorization and that, in any case, the plaintiffs could directly request these

documents from the third parties. Finally, the SLC and counsel argued that the SLC

files contained no relevant non-privileged documents that the plaintiffs could not

obtain through usual discovery.

The Court of Chancery ordered the SLC to produce all relevant documents

not subject to any valid privileges and objections raised by the SLC and the

24 Lead Pl.’s Mot. to Enforce Subpoenas and Certificate of Serv., Ex. E at 1–2 (Del. Ch. Oct. 7, 2019), Ch. Dkt. 203.

16 individual defendants.25 It ruled that, as a threshold matter, “Zapata-style discovery

is unnecessary,”26 meaning that the court was not reviewing an SLC’s decision on

terminating the litigation. The Court of Chancery then ruled that the plaintiffs were

presumptively entitled to the production of all documents and communications actually reviewed and relied upon by the SLC or its counsel in forming its conclusions that (i) it would not be in Oracle’s best interests to seek to dismiss the derivative claims and (ii) it was in Oracle’s best interests to allow the Lead Plaintiff (rather than the SLC) to proceed with the litigation on behalf of Oracle.27

The “universe of documents” was, however, subject to, and limited by, the SLC’s

privilege objections.28 The Court of Chancery held that, with limited exceptions, the

plaintiffs were not entitled to documents exchanged in the mediation proceedings.

The SLC complied with the Court of Chancery’s December 2019 Opinion,

produced the required documents, and generated a privilege log for the documents

withheld on attorney-client privilege and work product grounds. The plaintiffs

moved to compel further production, including memoranda of SLC interviews, a

PowerPoint prepared by the SLC’s counsel summarizing the evidence, tables

summarizing NetSuite’s financial performance, and a draft report of the SLC.

25 December 2019 Opinion at *19–20. 26 Id. at *14. 27 Id. at *18. 28 Id.

17 The Court of Chancery denied the plaintiffs’ motion to compel further

production of documents.29 First, the court determined that the SLC properly

invoked work product protection and that the SLC did not waive this protection. It

found that the plaintiffs failed to demonstrate that they could not obtain the

substantial equivalent of the interview memos and other documents without undue

hardship.30 The court was not persuaded by the plaintiffs’ argument that, even if the

items sought were protected, the SLC had waived the protection by exchanging

mediation statements that purportedly disclosed the contents with Ellison and Catz.31

Instead, the court decided that the parties had an expectation of privacy in a

confidential mediation, supported by Delaware’s “strong public policy favoring

confidentiality in all mediation proceedings.”32

The Court of Chancery also denied the motion on fiduciary duty grounds.

The court rejected the plaintiffs’ argument that the court should have applied a

heightened review standard because the SLC’s assertion of work product protection

impeded the plaintiffs’ prosecution of the case. As the court explained, the SLC “is

composed of fiduciaries for Oracle, who may well have good faith reasons to keep

29 In re Oracle Corp. Deriv. Litig., 2020 WL 3867407, at *4, *12 (Del. Ch. July 9, 2020) [hereinafter July 2020 Opinion]. 30 Id. at *7–9. 31 Id. at *9–10. 32 Id. at *10.

18 the work product done on the SLC’s behalf confidential.”33 And because the

plaintiffs did not plead breach of duty claims against members of the SLC, the court

declined to question the SLC’s business judgment in withholding the documents.

E.

The Court of Chancery held a ten-day trial in July and August 2022. In its

post-trial opinion, the Court of Chancery concluded that the “transaction was

negotiated at arm’s length by a fully empowered Special Committee.”34 The court

first determined that, although Ellison would receive a non-ratable benefit from the

sale of his stock in NetSuite and was therefore conflicted, he removed himself from

the process and left the decision-making to an independent and disinterested special

committee.35

Next, the court examined whether Ellison exercised general control over the

Oracle board despite insulating himself from the process. The court found that

Ellison held less than 30% of Oracle voting power and therefore did not have hard

control. Also, the court found that Ellison did not control Oracle’s day-to-day

functions nor the board’s decisions over the company’s operations. And it found

33 Id. at *11. 34 Post-Trial Opinion at *36. 35 Id. at *18.

19 that the Oracle board “was not afraid to stand opposed to Ellison.”36 The court

concluded that, although “Ellison had clout,” he “did not exercise general control.”37

The Court of Chancery also found that, although Ellison “had the potential to

influence the transaction, [he] did not attempt to do so.”38 After examining the

relationship between Ellison and the Special Committee, as well as its negotiations,

the court decided that “the Special Committee completed the transaction unmolested

by [Ellison’s] influence.”39 The court did not accept the plaintiffs’ factual claims

that Ellison (1) proposed the transaction; (2) controlled the transaction through his

January 27 call with Goldberg; and (3) controlled the transaction through Catz. The

court was also not persuaded that Ellison’s potential to control a transaction was so

inherently coercive that it rendered him a controlling stockholder. Accordingly, the

court found that Ellison did not exercise transactional control over the acquisition.40

Finally, the court addressed the plaintiffs’ allegations that Ellison and Catz

defrauded the Oracle board and the Special Committee by failing to disclose material

36 Id. at *20. 37 Id. 38 Id. at *21. 39 Id. 40 Id. at *25–27.

20 facts relating to NetSuite’s valuation and their interactions with NetSuite.41

According to the court, whether the standard of review should be elevated from

business judgment review under the plaintiffs’ “fraud on the board” theory turned

on a five-part test: whether (1) the fiduciary was materially interested; (2) the board

was inattentive or ineffective; (3) the fiduciary deceived or manipulated the board;

(4) the deception was material; and (5) the deception tainted the board’s decision-

making process.42 Applying this test, the court found that neither Ellison nor Catz

withheld material information or misled the Oracle board and Special Committee.43

Thus, the Court of Chancery rejected the plaintiffs’ disclosure claim and found that

neither Ellison nor Catz defrauded the Oracle board and the Special Committee.44

II.

On appeal, the plaintiffs argue first that the Court of Chancery erred by

permitting the SLC to withhold its interview memos. According to the plaintiffs,

the court: ignored precedent supporting access to non-opinion work product;

incorrectly applied business judgment review to the SLC’s decision to withhold the

interview memos; failed to find waiver after the SLC used the memos or disclosed

41 Id. at *27. 42 Id. 43 Id. at *27, *34. 44 Id. at *27, *34.

21 the contents during mediation with Ellison and Catz; and incorrectly found no

substantial need or undue hardship by the plaintiffs to use the interview memos for

impeachment. We review discovery rulings to determine whether the trial court

exceeded its discretion.45 We review questions of law, including those concerning

attorney-client privilege and work product immunity, de novo.46

In Zapata Corp. v. Maldonado, we held that an SLC’s decision to dismiss a

derivative lawsuit is not subject to the business judgment rule standard of review.47

Instead, we applied a higher standard of review because the SLC’s motion “is

addressed necessarily to the reasonableness of dismissing the complaint prior to trial

without any concession of liability on the part of the defendants and without

adjudicating the merits of the cause of action itself.”48 To meet its burden, “the SLC

must show . . . that no disputed issues of material fact exist about the independence,

good faith, and reasonableness of the SLC’s investigation and whether the SLC had

45 Coleman v. PricewaterhouseCoopers, LLC, 902 A.2d 1102, 1106 (Del. 2006) (reviewing pretrial discovery rulings for abuse of discretion). 46 Espinoza v. Hewlett-Packard Co., 32 A.3d 365, 371 (Del. 2011) (citation omitted). 47 Zapata Corp. v. Maldonado, 430 A.2d 779, 787 (Del. 1981). 48 Diep ex rel. El Pollo Loco Hldgs., Inc. v. Trimaran Pollo P’rs, L.L.C., 280 A.3d 133, 151 (Del. 2022) (quoting Kaplan v. Wyatt, 484 A.2d 501, 507 (Del. Ch. 1984), aff’d, 499 A.2d 1184 (Del. 1985)).

22 reasonable bases for its conclusions.”49 And, as a second discretionary step, the

court can apply its own business judgment and decide whether dismissal or

settlement is in the corporation’s best interests.50

The plaintiffs make a two-part argument. First, according to the plaintiffs,

under Zapata, they are typically permitted discovery into the SLC’s process, which

would include access to interview memos; and second, the court should extend

Zapata’s heightened scrutiny to the SLC’s decision to withhold information based

on work product protection.

It is true that the Court of Chancery routinely allows limited discovery into an

SLC’s investigation.51 A filing plaintiff uses discovery to evaluate whether material

issues of disputed fact exist about an SLC’s independence and good faith, the

reasonableness of its investigation, and whether the SLC had reasonable bases for

its conclusions.52 But here, the court is not considering whether it should grant the

Oracle SLC’s motion to terminate litigation. Rather, the SLC decided to return the

litigation to the plaintiffs. They accepted. The plaintiffs neither challenge the

49 Id. 50 Id. 51 See, e.g., In re Baker Hughes, a GE Co., Deriv. Litig., 2023 WL 2967780, at *9 (Del. Ch. Apr. 17, 2023), aff’d, 312 A.3d 1154 (Del. 2024) (TABLE); London v. Tyrrell, 2010 WL 877528, at *13 (Del. Ch. Mar. 11, 2010); Sutherland v. Sutherland, 2007 WL 1954444, at *3–4 (Del. Ch. July 2, 2007). 52 Zapata, 430 A.2d at 788.

23 independence, good faith, and reasonableness of the SLC’s investigation nor

whether the SLC should have allowed the plaintiffs to assume control of the

litigation. The plaintiffs did not need discovery to evaluate the SLC’s decision.

Zapata review did not apply.

Perhaps recognizing the disconnect between Zapata review and the SLC’s

decision here, the plaintiffs ask us to extend Zapata enhanced review to “any SLC

determination that impairs [the] prosecution of a corporate claim . . . .” 53 The

plaintiffs claim that, by withholding the interview memos, the SLC “did not

demonstrate good faith or reasonableness in hindering Plaintiffs’ prosecution of

Oracle’s claims . . . .”54 As they argue, “Zapata supplies an appropriate framework

for testing the SLC’s good faith.”55

We decline the invitation. Zapata review is tailored to meet specific concerns

in a distinct area of Delaware corporate law – reviewing an SLC’s decision to

terminate litigation, either through dismissal or settlement.56 Although the plaintiffs

53 Appellants’ Opening Br. at 33 [hereinafter Opening Br.]. 54 Id. at 34. 55 Appellants’ Reply Br. at 13 [hereinafter Reply Br.]. 56 See Baker Hughes, 2023 WL 2967780, at *10 (observing that Zapata addresses “the tension between the board’s responsibility under Section 141 to control a corporation’s litigation assets and the risk that a conflicted board would seek to terminate a beneficial derivative action”); London, 2010 WL 877528, at *11 (discussing Zapata’s application in “demand-excused derivative cases in which the board sets up an SLC that investigates whether a derivative suit should proceed and recommends dismissal after its investigation”).

24 are correct that they and the SLC are somewhat aligned in their mission – to evaluate

and, if warranted, pursue claims against the defendants – we agree with the Court of

Chancery that creating another layer of review involving a discovery issue could

impinge on the SLC’s effectiveness when investigating derivative claims. As the

Court of Chancery held, “[a]llowing complete discovery of all documents provided

to or created by a special litigation committee in situations such as these, as requested

by the Subpoenas, could chill candor and access and limit the effectiveness of special

litigation committees going forward.”57 Instead of fiduciary review, we conclude

that the discovery rules – in particular Rule 26(b)(3) and decisions interpreting those

rules when fiduciaries are involved – are better suited to address privilege and work

product issues in situations where the SLC has turned over the litigation to the

plaintiffs.

The plaintiffs agree that non-opinion work product protection applies to

attorney interview summaries, that the mediation between the SLC and defendants

was protected by a signed mediation confidentiality agreement, and that the parties

conducted the mediation confidentially. Nonetheless, they contend that the SLC

waived any non-opinion work product protection when it allegedly shared interview

57 December 2019 Opinion at *17.

25 memoranda or referred to their contents in mediation submissions.58 A party can

waive work product protection by disclosing work product to an adversary or by

“intentionally disclos[ing] or consent[ing] to disclosure of any significant part of the

privileged or protected communication or information.”59 And under some

circumstances, a party can waive work product protection by disclosing the

documents with “either the intention or practical result that the opposing party may

see the documents.”60 But “[t]here is no waiver of privileged information to third

parties if a disclosing party had a reasonable expectancy of privacy when it made an

earlier disclosure[,]” and such expectancy of privacy was sanctioned by the law.61

The record does not support the plaintiffs’ contention that interview

memoranda were disclosed during the mediation. And even if the SLC relied on

interview summary content during the mediation, we agree with the Court of

Chancery that “[t]he SLC had a strong expectancy of privacy” in the mediation

proceedings.62 This is consistent with Court of Chancery Rule 174, which provides

that “[a]ll communications made in or in connection with the mediation that relate

58 Opening Br. at 36–37. 59 D.R.E. 510(a). 60 Saito v. McKesson HBOC, Inc., 2002 WL 31657622, at *4 (Del. Ch. Nov. 13, 2002) (quoting Wolhar v. Gen. Motors Corp., 712 A.2d 457, 462–63 (Del. Super. Ct. 1997)). 61 Id. 62 July 2020 Opinion at *10.

26 to the controversy being mediated, whether with the mediator or a party during the

mediation, are confidential”63 and “not subject to discovery.”64 Waiver occurs only

when privileged information is disclosed outside the confidential proceedings.65

According to the record, that was not the case here.

Holding otherwise would deter mediating parties from engaging in frank

exchanges to resolve a dispute. As the court explained in Wilmington Hospitality,

L.L.C. v. New Castle County, “[c]onfidentiality of all communications between the

parties or among them and the mediator serves the important public policy of

promoting a broad discussion of potential resolutions to the matters being

mediated.”66 The court warned that, “[w]ithout the expectation of confidentiality,

parties would hesitate to propose compromise solutions out of the concern that they

would later be prejudiced by their disclosure.”67

63 Ct. Ch. R. 174(g)(3) (emphasis added). 64 Id. at 174(h)(1), 174(h)(3). 65 Saito, 2002 WL 31657622, at *3–4. 66 788 A.2d 536, 541 (Del. Ch. 2001). 67 Id.; see also Princeton Ins. Co. v. Vergano, 883 A.2d 44, 62–63 (Del. Ch. 2005) (observing that Delaware’s approach to mediation aligns with that of the Uniform Mediation Act). The Act highlights the importance of confidentiality in instilling public confidence in mediation as a dispute resolution modality, stating that “frank exchange can be achieved only if the participants know that what is said in the mediation will not be used to their detriment through later court proceedings and other adjudicatory processes.” Nat’l Conference of Comm’rs on Unif. State L., Uniform Mediation Act, at Prefatory Note (2003) (citations omitted).

27 The plaintiffs point to Ryan v. Gifford and Tackett v. State Farm Fire &

Casualty Insurance Company.68 In Ryan, the Court of Chancery found that a special

committee waived its attorney-client privilege when it disclosed its detailed findings

in a meeting attended by the full board of directors, the director defendants (in their

individual capacity), and the director defendants’ individual, outside counsel.69 But

here, the SLC shared the mediation statements with Ellison and Catz as part of

confidential mediation proceedings. And the plaintiffs have not shown that the SLC

shared its detailed findings with Ellison and Catz.70 Instead, they simply speculated

that the protected materials were disclosed through mediation statements.71

Tackett is similarly unpersuasive. In Tackett, we held that a party waived

attorney-client privilege when it asserted an affirmative defense that implicated

privileged material.72 To support its argument that routine handling of a claim did

not contribute to the delay in payments to an insured, an insurer alleged

68 Ryan v. Gifford, 2008 WL 43699 (Del. Ch. Jan. 2, 2008); Tackett v. State Farm Fire & Cas. Ins. Co., 653 A.2d 254 (Del. 1995). 69 Ryan, 2008 WL 43699, at *6; see also Ryan v. Gifford, 2007 WL 4259557, at *3 (Del. Ch. Nov. 30, 2007) (“On January 18 and 19, 2007, the Special Committee presented its final oral report to Maxim’s board of directors. This report appears to be more than a mere acknowledgement of the existence of the report and instead disclosed such details that, for example, attendees were directed to turn in any notes taken during the presentation at the end of the meeting.”). 70 July 2020 Opinion at *9. 71 Id. 72 Tackett, 653 A.2d at 259–60.

28 particularized facts that implicitly relied upon communications with its counsel.73

Like the special committee’s disclosure in Ryan, however, the insurer’s disclosure

in Tackett was not made as part of a confidential proceeding. Here, the SLC’s

disclosure was made as part of confidential mediation proceedings, and the SLC did

not waive privilege.

Finally, the plaintiffs argue that, under Court of Chancery Rule 26(b)(3) and

Garner v. Wolfinbarger,74 the SLC’s non-opinion work product protection should

yield to the plaintiffs’ substantial need and undue hardship concerns for the SLC

memos.75 They claim a substantial need for the interview memos for two reasons:

first, Hurd passed away before he could be deposed, meaning that the interview

memos were needed for his recollection of events; and second, access to the memos

would allow the plaintiffs to test whether Catz and Nelson discussed price during

their January 2016 phone call when Catz had been instructed not to discuss price.

73 Id. 74 Ct. Ch. R. 26(b)(3); 430 F.2d 1093 (5th Cir. 1970). In Wal-Mart Stores, Inc. v. Ind. Elec. Workers Pension Tr. Fund IBEW, we held that the Garner factors applied to attorney client privilege only. 95 A.3d 1264, 1280 (Del. 2014). But we also held that these factors overlap with the required showing of “good cause” under Court of Chancery Rule 26(b)(3). Id. at 1280–81; see Zirn v. VLI Corp., 621 A.2d 773, 782 (Del. 1993) (applying the Garner Factors when analyzing whether a plaintiff has demonstrated “good cause” for the production of work product). 75 Opening Br. at 38–39. On appeal, the plaintiffs’ “substantial need and good cause” argument is limited to non-opinion work product. Id.

29 We will not consider the substantial-need argument related to Hurd’s

interview with the SLC. The plaintiffs did not directly raise the argument below. It

is waived on appeal.76 Whether the plaintiffs demonstrated substantial need or good

cause for the interview memos for impeachment related to price discussions is a

closer call. On appeal, however, we defer to the Court of Chancery unless it

exceeded its discretion.77

Under Rule 26(b)(3), to gain access to non-opinion work product, the

plaintiffs must show that they have a “substantial need of the materials in the

preparation of their case and that the party is unable without undue hardship to obtain

the substantial equivalent of the materials by other means.”78 The Court of Chancery

did not exceed its discretion when it found that the plaintiffs:

76 Supr. Ct. R. 8. In its July 2020 Opinion, the Court of Chancery observed that, although “[t]he Lead Plaintiff does note that two interview subjects – Hurd and former Oracle director Hector Garcia-Molina – have since died[,] . . . the Lead Plaintiff has failed to argue substantial need and undue hardship specifically regarding Hurd’s and Garcia-Molina’s interview memoranda.” July 2020 Opinion at *7 n.61. Confusingly, the plaintiffs claim that the court’s waiver finding “actually reflects that the significance of Hurd’s unavailability was argued below.” Reply Br. at 5. The Court will generally decline to review questions on appeal “unless they were first fairly presented to the trial court for consideration.” Ravindran v. GLAS Tr. Co. LLC, 2024 WL 4258889, at *12 (Del. Sept. 23, 2024) (quoting Russell v. State, 5 A.3d 622, 627 (Del. 2010)). The plaintiffs are correct that they brought Hurd’s passing to the court’s attention. They also referred to the difficulty caused by the passing. See Reply Br. at 5–7. But we agree with the defendants that the plaintiffs did not present a separate argument relating to substantial need caused by Hurd’s death. Referring to his death and potential complications caused by it was not the same as directly raising an argument under Rule 26(b)(3) so that the court understands it is an argument that must be considered and decided. 77 Alaska Elec. Pension Fund v. Brown, 988 A.2d 412, 419 (Del. 2010). 78 Ct. Ch. R. 26(b)(3).

30  were provided most of the contemporaneous documents and testimonial evidence generated by the SLC;

 with the exception of Hurd, had the opportunity to depose the witnesses interviewed by the SLC;

 could take witness testimony under oath while the interviews were presumably not and might lead to a mini-trial over what was actually said versus what was recorded in the memos; and

 only speculated that the witness interviews might offer impeachment opportunities.79

In addition, we note that the plaintiffs were not without impeachment evidence on

what they claim was a central issue in the case – whether Catz and Nelson discussed

price during their initial call. For example, the plaintiffs had the September 2016

letter from T. Rowe Price to the independent members of the NetSuite board.80 The

letter alleged that Catz and Nelson discussed a price range of $100–$125 during their

initial contact, which, T. Rowe Price alleges, “may have anchored the subsequent

79 July 2020 Opinion at *6–9. 80 In a letter written by NetSuite’s largest unaffiliated shareholder, T. Rowe Price, to the independent members of the NetSuite board, T. Rowe Price offered a different view regarding Catz’s silence on price:

In our recent meeting, Mr. Nelson described the initial contact with Oracle as a loose, pre-due-diligence, exploratory conversation where a price range of $100–$125 was discussed. We don’t think it’s a coincidence that the final agreement ended up very close to the midpoint of that range. We are concerned that this initial conversation . . . may have anchored the subsequent discussions. This anchoring effect . . . may have prevented full price discovery.

A1861 (Letter from T. Rowe Price Assocs., Inc. to the Indep. Members of the Bd. of NetSuite, Inc., Sept. 6, 2016).

31 discussions” and “prevented full price discovery.”81 Nonetheless, the Vice

Chancellor observed the witnesses at trial, including Catz and Nelson who testified

about their conversations,82 assessed their credibility, and found against the plaintiffs

on the issue.83 On appeal, we cannot say that the court exceeded its discretion.

III.

The Court of Chancery applied business judgment review – not entire

fairness – to the Oracle/NetSuite transaction. Although Ellison held a substantial

block of Oracle stock and was its visionary co-founder, the court decided that the

plaintiffs failed to prove Ellison wielded either general control over Oracle or

transaction-specific control or that fiduciaries misled the Special Committee while

considering the transaction.84 We affirm its decision to apply business judgment

review to the transaction.

81 Id. 82 See A1115 (Tr. 1062:5–1063:17) (Nelson’s testimony regarding the dinner conversation); A1208–09 (Tr. 1436:1–1438:14) (Catz’s testimony regarding the dinner conversation). 83 Post-Trial Opinion at *6. 84 Id. at *36.

32 A.

As a general rule, stockholders do not owe fiduciary duties to the corporation

or its stockholders and are free to act in their self-interest.85 But a stockholder who

owns or controls over 50% of a Delaware corporation’s stock is presumed to exercise

“hard” control and assumes fiduciary duties in certain circumstances.86 This is

because a majority stockholder controls the levers of power within the corporation.

As we have said before, “[i]n addition to the election of directors, many of the most

fundamental corporate changes also require approval by a majority vote of the

stockholders, e.g., mergers, consolidations, sales of all or substantially all of the

assets of a corporation and dissolutions.”87

85 Bershad v. Curtiss-Wright Corp., 535 A.2d 840, 845 (Del. 1987) (“Stockholders in Delaware corporations have a right to control and vote their shares in their own interest. They are limited only by any fiduciary duty owed to other stockholders. It is not objectionable that their motives may be for personal profit, or determined by whim or caprice, so long as they violate no duty owed other shareholders.”); see also Williams v. Geier, 671 A.2d 1368, 1384 (Del. 1996) (holding that a presumptive controlling block was free to vote their shares as they saw fit as long as the underlying act did not entail “waste, fraud, or manipulative or other inequitable conduct”). 86 Weinstein Enters., Inc. v. Orloff, 870 A.2d 499, 507 (Del. 2005) (first citing Paramount Commc’ns, Inc. v. QVC Network, Inc., 637 A.2d 34 (Del. 1994); then citing Citron v. Fairchild Camera & Instrument Corp., 569 A.2d 53, 70 (Del. 1989); and then citing Gilbert v. El Paso Co., 490 A.2d 1050, 1055 (Del. Ch. 1984)); see In re Match Grp., Inc. Deriv. Litig., 315 A.3d 446, 460 (Del. 2024) (“Controlling stockholders are at times free to act in their own self-interest. But a controlling stockholder is a fiduciary and must be fair to the corporation and its minority stockholders when it stands on both sides of a transaction and receives a non-ratable benefit. In such cases, the controlling stockholder bears the burden of demonstrating ‘the most scrupulous inherent fairness of the bargain.’” (citations omitted)). 87 Weinstein, 870 A.2d at 507.

33 Conversely, a stockholder who owns or controls less than 50% of a

corporation’s voting power is not presumed to be a controlling stockholder with

fiduciary duties.88 Even so, a minority stockholder can be a controlling stockholder

by exercising actual control over the corporation’s business and affairs or by

exercising actual control over a specific transaction.89

The test for actual control by a minority stockholder “is not an easy one to

satisfy.”90 The minority stockholder must have “a combination of potent voting

power and management control such that the stockholder could be deemed to have

effective control of the board without actually owning a majority of stock.” 91 To

prove actual control over a specific transaction, a plaintiff must prove that the

88 Id.; see Gilbert, 490 A.2d at 1055 (“[A] shareholder who owns less than 50% of a corporation’s outstanding stocks does not, without more, become a controlling shareholder of that corporation, with a concomitant fiduciary status.”). 89 Kahn v. Lynch Commc’n Sys., Inc., 638 A.2d 1110, 1113–14 (Del. 1994); Ivanhoe P’rs v. Newmont Mining Corp., 535 A.2d 1334, 1344 (Del. 1987); see also Basho Techs. Holdco B, LLC v. Georgetown Basho Invs., LLC, 2018 WL 3326693, at *25 (Del. Ch. July 6, 2018) (“The requisite degree of control can be shown to exist generally or with regard to the particular transaction that is being challenged.” (citation omitted)), aff’d sub nom. Davenport v. Basho Techs. Holdco B, LLC, 221 A.3d 100 (Del. 2019) (TABLE). 90 In re PNB Hldg. Co. S’holders Litig., 2006 WL 2403999, at *9 (Del. Ch. Aug. 18, 2006). 91 Corwin v. KKR Fin. Hldgs. LLC, 125 A.3d 304, 307 (Del. 2015); see also Lynch, 638 A.2d at 1113 (“[A] shareholder owes a fiduciary duty only if it owns a majority interest in or exercises control over the business affairs of the corporation.” (emphasis and citation omitted)).

34 minority stockholder “exercised actual control over the board of directors during the

course of a particular transaction.”92

The plaintiffs argue that Ellison exercised both general control over Oracle

and its board and specific control over the NetSuite transaction. According to the

plaintiffs, over time, Ellison held between a 20% to 43% ownership interest in

Oracle. Combined with his managerial authority, they claim he had general control

over Oracle.93 They also claim that Ellison’s status as a “visionary leader” to

promote the NetSuite acquisition and implement his allegedly concealed post-

acquisition plans support a general control finding.94

This appeal is not, however, from an early-stage dismissal decision. Even

though the Court of Chancery denied a pleading-stage dismissal regarding Ellison’s

controlling stockholder status, the plaintiffs’ general and transaction-specific control

arguments were fully vetted during a ten-day trial. On appeal, the plaintiffs cite facts

and testimony favorable to their arguments. But we do not weigh evidence on

92 In re W. Nat’l Corp. S’holders Litig., 2000 WL 710192, at *20 (Del. Ch. May 22, 2000) (holding that, where the significant stockholder exercises actual control over the transaction, even though not the general affairs of the corporation, she assumes fiduciary duties for purposes of that transaction); Basho, 2018 WL 3326693, at *27 (“Broader indicia of effective control . . . play a role in evaluating whether a defendant exercised actual control over a decision.”); In re Rouse Props., Inc., 2018 WL 1226015, at *12 (Del. Ch. Mar. 9, 2018) (holding that a minority block holder may be deemed a controlling stockholder if she exercised actual control over the “deciding committee with respect to the challenged transaction” (citation omitted)). 93 Opening Br. at 44–46. 94 Id.

35 appeal.95 Equally important, the plaintiffs have not argued that the Vice

Chancellor’s contrary factual findings on general and transactional control are

clearly wrong.96

The “control” question is “a judicial conclusion that is reached after a fact

specific analysis.”97 The Vice Chancellor found the following unchallenged facts to

conclude that Ellison, as a minority stockholder, did not exercise actual control over

the Oracle board:

 the Oracle board and management were not afraid to disagree with Ellison;

 Ellison neither controlled Oracle’s day-to-day functions nor dictated Oracle’s operations to the Oracle board;

 Ellison “scrupulously avoided” discussing the transaction with the Special Committee;

 Ellison neither proposed the transaction nor indirectly controlled the merger negotiations through his January 27, 2016, phone call with Goldberg; and

 although Ellison could have controlled the transaction, he did not interfere with or actually exercise control over the transaction.98 95 Arrants v. Home Depot, 65 A.3d 601, 605 (Del. 2013) (“Appellate courts do not weigh the evidence, determine questions of credibility, or make factual findings.” (citation omitted)). 96 Klaassen v. Allegro Dev. Corp., 106 A.3d 1035, 1043 (Del. 2014) (“We will not overturn the Court of Chancery’s factual findings unless they are clearly erroneous.” (citation omitted)). 97 Weinstein, 870 A.2d at 506–07. 98 Post-Trial Opinion at *20–27. The plaintiffs claim that “instead of analyzing Ellison’s ‘combination of stock voting power and managerial authority,’ . . . the Vice Chancellor erroneously looked at each indicia of control in isolation.” Reply Br. at 20. They argue that the

36 The plaintiffs rely heavily on In re Cysive, Inc. Shareholders’ Litigation.99

They argue that, as in Cysive, “[i]f the answer respecting Ellison’s ‘capability’ to

exercise control is yes, then ‘it cannot be that the mere fact that [Ellison] did not

interfere with the special committee is a reason to conclude that he is not a

controlling stockholder.’”100

In Cysive, the Court of Chancery found after trial that Cysive’s CEO, who

proposed a management buyout and controlled a 40% voting block, was not an

ordinary CEO, but “a hands-on one, to boot” who was “involved in all aspects of the

company’s business.”101 He “was the company’s creator” and “inspirational

force[,]” and had two close family members in executive positions as well as a

formerly-employed sister.102 He could also install a “new slate more to his liking

without having to attract much, if any, support from public stockholders.” 103

Court of Chancery should have conducted a “holistic evaluation of the sources of influence” in its analysis on “control.” Reply Br. at 24. Although we agree with the plaintiffs that a holistic evaluation is important, the Vice Chancellor did consider all the facts at trial in the aggregate and concluded that, “in light of these facts,” this was not a controlled transaction. Post-Trial Opinion at *27. 99 836 A.2d 531 (Del. Ch. 2003). 100 Reply Br. at 22 (alteration in original) (quoting Cysive, 836 A.2d at 552). 101 Cysive, 836 A.2d at 552. 102 Id. 103 Id.

37 Here, like in Cysive, Ellison was a co-founder and visionary leader of Oracle.

And, as the Court of Chancery found, a visionary leader might have, in combination

with other factors, the potential to exercise general control over the corporation. But

as the court said in Cysive, “the question of whether a large block holder is so

powerful as to have obtained the status of a ‘controlling stockholder’ is intensely

factual.”104 Although the controlling stockholder question is not a license to sue on

every transaction involving a corporation with a founder/visionary leader, “[i]n cases

when the determination of whether control exists turns on disputed facts, it is

impossible to determine whether a large block holder is a controlling stockholder

until an evidentiary hearing is held.”105 Here, as explained previously, the Court of

Chancery found after trial as a factual matter that Ellison, as a 28.8% stockholder,

could not exert the type of control found in Cysive.106 The Vice Chancellor came to

this conclusion after a ten-day trial, and the factual findings underpinning the

conclusion have not been challenged on appeal.107

104 Id. at 550–51. 105 Id. at 551. 106 Post-Trial Opinion at *26–27. 107 The plaintiffs also contend that the court “misapplied the legal significance of [its] factual finding that Ellison ‘likely had the potential to control the transaction at issue.’” Reply Br. at 18 (quoting Post-Trial Opinion at *27). They argue that Ellison’s potential to control Oracle transactions is the same as the observation in Cysive that the minority stockholder could influence transactions “if he so wishes.” See Cysive, 836 A.2d at 553. As we see it, however, the Vice Chancellor found that Ellison, as a minority stockholder, did not exercise general control over the

38 B.

We address together the last two issues on appeal – whether the Court of

Chancery used the incorrect “test” to decide if entire fairness review should have

applied and whether Ellison failed to disclose the material information about his

future NetSuite integration plans. The first issue raises a question of law; the second

raises a mixed question of fact and law. We review both issues de novo.108

The Court of Chancery applied a multi-factor test to assess the plaintiffs’

“fraud on the board” theory of liability directed at a fiduciary.109 We find, however,

that when it comes to fiduciaries accused of disloyal conduct, the inquiry does not

require a multi-factor test. Instead, the court starts from familiar ground and decides

whether a conflicted fiduciary violated his fiduciary duty of loyalty. When

Oracle board. When discussing the “potential” to control transactions, the court was simply addressing settled law in the transactional control setting that “the potential ability to exercise control is not sufficient.” Williamson v. Cox Commc’ns, Inc., 2006 WL 1586375, at *4 (Del. Ch. June 5, 2006). Here, Ellison recused himself and therefore did not exercise actual control over the Oracle/NetSuite transaction. In any event, the potential to control transactions, without more, does not lead ineluctably to controlling stockholder status. 108 In re Tesla Motors, Inc. S’holder. Litig., 298 A.3d 667, 699 (Del. 2023) (“Our review of the formulation and application of legal principles . . . is plenary and requires no deference.” (quoting Kahn v. Lynch Commc’n Sys., Inc., 669 A.2d 79, 84 (Del. 1995))). Zirn v. VLI Corp., 681 A.2d 1050, 1055 (Del. 1996) (“[I]n an appropriate case, this Court may review de novo mixed questions of law and fact, such as determinations of materiality.” (citation omitted)). 109 According to the Court of Chancery, to shift the standard of review under a “fraud on the board” theory, the plaintiffs must prove that (1) the fiduciary was materially interested; (2) the board was inattentive or ineffective; (3) the fiduciary deceived or manipulated the board; (4) the deception was material; and (5) the deception tainted the decision-making process of the board. Post-Trial Opinion at *27.

39 interacting with the Board, the duty of loyalty requires a fiduciary to act in good

faith.110 Good faith requires candor with the board. “[F]iduciaries, corporate or

otherwise, may not use superior information or knowledge to mislead others in the

performance of their own fiduciary obligations.”111 When a fiduciary withholds

material information from the board, engages in deceptive conduct, or otherwise

misleads the board, he has failed to act in good faith and therefore acted disloyally.

The plaintiffs fault the court for supposedly requiring the board to be

ineffective to sustain a “fraud on the board” claim. Although the board need not be

ineffective for a plaintiff to prevail on a breach of the duty of loyalty claim, as

explained next, we affirm the Court of Chancery’s finding that the plaintiffs failed

to prove that Ellison withheld material information from the Committee.

In Haley, we held that a fiduciary must disclose information to the board that

is “relevant and of a magnitude to be important to directors in carrying out their

fiduciary duty of care in decisionmaking.”112 The touchstone of the materiality

inquiry is whether a reasonable board or special committee member would have

regarded the existence of the undisclosed facts as significant. 113

110 Stone v. Ritter, 911 A.2d 362, 369–70 (Del. 2006). 111 Mills Acq. Co. v. Macmillan, Inc., 559 A.2d 1261, 1283 (Del. 1989). 112 City of Fort Myers Gen. Emps.’ Pension Fund v. Haley, 235 A.3d 702, 718 (Del. 2020) (quoting Brehm v. Eisner, 746 A.2d 244, 259 n.49 (Del. 2000)). 113 Id. at 724.

40 The Court of Chancery found that Oracle followed its usual practice in M&A

transactions by holding off considering the effects of post-closing plans until after

the deal was signed.114 The plaintiffs are troubled by this factual finding because

they believe the Oracle/NetSuite transaction was not a usual transaction.115 They

argue that, as a co-founder and controlling stockholder, Ellison possessed unique

insight into both companies and formulated post-closing operating plans that should

have been disclosed to the Special Committee.116 Had Ellison disclosed these

important insights, the plaintiffs contend, the Special Committee would have paid

closer to $74.58 per share for NetSuite, significantly lower than the $109 per share

Oracle ended up paying. 117

We note with skepticism the argument that Ellison should have made his

post-closing views known during the Special Committee’s process. If that were so,

114 Post-Trial Opinion at *32. 115 Reply Br. at 32–33. 116 Id. 117 Opening Br. at 55. The plaintiffs also argue that the Court of Chancery made two erroneous factual findings. First, they allege the court erred by finding that the Special Committee was “apprised of the level of competition between NetSuite and Oracle.” Opening Br. at 59 (quoting Post-Trial Opinion at *35). But during both the May 5, 2016 diligence meeting attended by James and the May 13, 2016 Special Committee meeting, the participants discussed Oracle’s and NetSuite’s respective market positioning, competitive landscape, and the two companies’ “potentially complementary nature.” Post-Trial Opinion at *9–11. They also claim that the Vice Chancellor erred by finding that NetSuite had upmarket ambitions and was in the process of implementing Ellison’s critiques of its strategy. Opening Br. at 59–60. Although NetSuite did not completely end all upmarket initiatives, NetSuite focused on its verticalization initiatives through Atlas/SuiteSuccess. See A1036–37 (Tr. 751:6–754:17).

41 the plaintiffs likely would have quickly pivoted and claimed that he breached his

fiduciary duty by improperly exerting his influence over the Special Committee’s

work.118 In any event, the Court of Chancery found that Ellison’s post-acquisition

plans were not of a magnitude that the Special Committee would have viewed it as

important.119 The court held that “[a]lthough there was competition between Oracle

and NetSuite at the margins, . . . the two were not significant competitors.”120

Furthermore, at the time of the transaction, NetSuite was in the process of

implementing changes to address Ellison’s 2015 concerns, which was known to the

Special Committee.121 Ellison did not want to move NetSuite in a strategic direction

that was entirely unforeseeable by the Special Committee or that otherwise changed

the fundamental value proposition of the transaction.122 On the contrary, he

118 The plaintiffs claim that Ellison could have worked out protocols for engaging with the Special Committee. Reply Br. at 29–30 (citing Tesla, 298 A.3d at 709 n.191). Even though that might be true, Ellison had no duty to do so, and, in any event, the Special Committee approved its own recusal protocol for the transaction (“Rules of the Road”). See B1482 (Email from B. Higgins to S. Catz et al.: RE: Rules of the Road for Project Napa). It is apparent that both Ellison and the Special Committee were satisfied with Ellison’s commitment to recuse himself from the Special Committee’s work. 119 Post-Trial Opinion at *32–33. 120 Id. at *29. 121 Id. at *30. 122 The plaintiffs claim that “Ellison’s plan for NetSuite contemplated significant new costs to reach a large number of smaller customers” and “jettisoning NetSuite’s up-market sales force and the corresponding projected revenues.” Opening Br. at 57. But the Special Committee was aware that NetSuite’s verticalization and international expansion efforts would require investment post- acquisition. See A867 (Tr. 74:11–75:20); A1151–52 (Tr. 1208:2–1212:22). And, as discussed

42 confirmed NetSuite’s verticalization strategy, which was already put in motion

through SuiteSuccess.123

Furthermore, Ellison did not actually control Oracle. He had no ability

unilaterally to bring about drastic changes without management and board approval,

which, as the Court of Chancery found, “did not appear cowed or overawed” by

Ellison.124 On appeal, we will not overturn the court’s conclusion that Ellison’s

undisclosed post-closing plans for operating NetSuite were immaterial to the Special

Committee’s evaluation and negotiation of the transaction.

IV.

The judgment of the Court of Chancery is affirmed.

above, NetSuite was already in the process of implementing Ellison’s 2015 concerns. Post-Trial Opinion at *30. 123 Post-Trial Opinion at *30. 124 Id. at *20.