Teradata Corporation et al v. SAP SE et al

Filing 65

Order by Judge William H. Orrick granting in part and denying in part 39 Motion to Dismiss. SAP's motion to dismiss is GRANTED with regards to the trade secret misappropriation claims but DENIED with regards to all other claims. Teradata's first amended complaint is DISMISSED WITH LEAVE TO AMEND within ten days of the date of this Order. (jmdS, COURT STAFF) (Filed on 12/12/2018)

1
2
3
4

UNITED STATES DISTRICT COURT

5

NORTHERN DISTRICT OF CALIFORNIA

6
7

TERADATA CORPORATION, et al.,
Plaintiffs,

8
9
10

United States District Court
Northern District of California

11

Case No. 18-cv-03670-WHO
ORDER GRANTING IN PART AND
DENYING IN PART MOTION TO
DISMISS

v.
SAP SE, et al.,

Re: Dkt. No. 39

Defendants.

12
13
14

INTRODUCTION
Plaintiffs Teradata Corporation, Teradata US, Inc., and Teradata Operations, Inc.

15

(collectively “Teradata”) assert five causes of action in the first amended complaint against

16

defendants SAP SE, SAP America, Inc., and SAP Labs, LLC (collectively “SAP”). Teradata

17

alleges that SAP disingenuously entered a joint venture with it to steal its trade secrets in the

18

Enterprise Data Analytics and Warehousing (“EDAW”) market and develop a competing product,

19

misappropriating trade secrets, infringing Teradata’s copyrights, and violating antitrust laws in the

20

process. SAP moves to dismiss all these claims with prejudice. I agree with SAP that Teradata

21

must describe its trade secrets with greater specificity to demonstrate that they are not generally

22

known in the trade or by those who are skilled in the trade. None of SAP’s other challenges to the

23

pleadings has merit. For the reasons below, SAP’s motion to dismiss is GRANTED in part and

24

DENIED in part. Teradata’s first amended complaint is DISMISSED WITH LEAVE TO

25

AMEND within ten days of the date of this Order.

26

BACKGROUND

27

Teradata and its subsidiaries conduct research, development, engineering, and other

28

technical operations related to its EDAW products. See First Amended Complaint (“FAC”) ¶¶ 4–

1

6 (Dkt. No. 24). Its flagship product is Teradata Database, a relational database management

2

system designed for EDAW. FAC ¶ 16. EDAW products provide centralized data storage

3

collected from numerous sources across a business enterprise in its day-to-day operations to help

4

large companies analyze their business operations. See id. Teradata was the first commercial

5

EDAW vendor to utilize massively parallel processing (“MPP”) through Teradata Database to

6

execute high volumes of analytical queries on massive amounts of data for EDAW customers.

7

FAC ¶ 17. Teradata Database accomplishes this feat by linear-performance scalability, meaning

8

that it can accommodate a customer’s analytical demands by adding parallel processors and data-

9

storage devices as needed. FAC ¶ 18. Teradata released the first commercial system
incorporating its MPP architecture in the early 1980s and has improved upon the technology since

11
United States District Court
Northern District of California

10

then, developing other technologies and trade secrets in the EDAW market. FAC ¶¶ 20–21.

12

SAP and its subsidiaries also work on research, development and engineering activities in

13

the EDAW space. FAC ¶ 9. At the same time, SAP is the dominant provider in the separate

14

market for Enterprise Resource Planning (“ERP”) Applications, comprised of the world’s most

15

complex, large-scale business enterprises (“Top-Tier ERP Applications Market”). FAC ¶ 30.

16

ERP Applications allow companies to manage data required to conduct their day-to-day operations

17

across numerous aspects of the business enterprise and are typically designed around a relational

18

transactional database that can ensure users have access to a uniform and current set of data. FAC

19

¶ 30. SAP also offers a Business Warehouse reporting tool (“SAP BW”), which allows ERP users

20

to generate reports with their ERP-derived data. FAC ¶ 31. The Top-Tier ERP Applications

21

Market that SAP services includes customers with millions of transactions or data-generating

22

events daily, multiple business lines, diverse geographic operations, multiple sources of data, and

23

revenues typically exceeding $1 billion. FAC ¶ 58.

24

In 2009, SAP and Teradata entered into a partnership referred to as the “Bridge Project” to

25

combine SAP’s ERP Applications and SAP BW tool interface with Teradata’s MPP architecture

26

that it uses in Teradata Database for EDAW. FAC ¶¶ 1, 31. Before the agreement was finalized

27

the parties executed two mutual non-disclosure agreements restricting the use of confidential

28

information outside evaluating the potential transaction. FAC ¶ 33. Eventually SAP and Teradata
2

1

executed two more agreements to formalize the Bridge Project, the Software Development

2

Cooperation Agreement (“SDCA”) and the Technology Partner Agreement (“TPA”). Id. These

3

agreements restricted disclosures of each parties’ confidential information and prohibited reverse

4

engineering while working to connect the SAP and Teradata software. Id. Teradata also provided

5

SAP with access to its databases for Bridge Project purposes pursuant to its “standard end user

6

license.” FAC ¶ 36.
During the Bridge Project, SAP and Teradata jointly developed “Teradata Foundation”

8

which enabled SAP’s Top-Tier ERP Applications Market to use Teradata for the transactional

9

database and data-analytics for EDAW activities. FAC ¶¶ 32, 37. The parties brought Teradata

10

Foundation to market and installed Teradata Foundation on site for one major customer facility.

11
United States District Court
Northern District of California

7

FAC ¶ 37. However, while the Bridge Project was underway, SAP was developing a competing

12

EDAW product called SAP HANA. FAC ¶ 38. At SAP’s user conference in November 2010, the

13

SAP Chief Technology Officer at the time, Dr. Vishal Sikka, announced SAP had begun shipping

14

the SAP HANA product. FAC ¶ 39. At the next year’s user conference, a SAP customer

15

demonstrated a new product, HANA for SAP BW, described as incorporating a “massively

16

parallel” database like the architecture used in Teradata Database. Id. By June 2011, SAP HANA

17

was commercially available. Id.

18

After nearly three years in the Bridge Project, and two months after SAP HANA was made

19

available, SAP unilaterally terminated the joint venture and stopped supporting, selling, and

20

marketing Teradata Foundation. FAC ¶ 40. SAP positioned itself as a competitor to Teradata

21

with the SAP HANA product offering to perform the same tasks as Teradata Foundation was

22

intended to achieve. FAC ¶¶ 40, 42. SAP customers were still able to access the data created in

23

their SAP ERP Applications by using Teradata EDAW products, but in February 2015 SAP

24

launched its latest version of ERP Application, SAP S/4HANA, which was incompatible with

25

other EDAW products besides HANA. FAC ¶¶ 77, 86. SAP then combined its ERP Application

26

and EDAW products into a single sales offering and announced that it was ending support for

27

prior versions of its ERP Applications by 2025 – an effort Teradata believes was intended to force

28

customers to adopt SAP HANA. FAC ¶¶ 87, 89, 91. Teradata expects that given the costs of
3

1

licensing and maintaining EDAW products, most Top-Tier SAP ERP Applications customers who

2

use Teradata EDAW products will abandon them in response to SAP’s decision not to facilitate

3

products other than HANA. FAC ¶ 92. Teradata also believes SAP is forcing customers to

4

abandon Teradata by restricting customer ability to access SAP ERP data for use with Teradata’s

5

EDAW products. FAC ¶ 93.

6

LEGAL STANDARD

7

Under Federal Rule of Civil Procedure 12(b)(6), a district court must dismiss a complaint

8

if it fails to state a claim upon which relief can be granted. To survive a Rule 12(b)(6) motion to

9

dismiss, the plaintiff must allege “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). A claim is facially plausible when

11
United States District Court
Northern District of California

10

the plaintiff pleads facts that “allow the court to draw the reasonable inference that the defendant

12

is liable for the misconduct alleged.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (citation

13

omitted). There must be “more than a sheer possibility that a defendant has acted unlawfully.” Id.

14

While courts do not require “heightened fact pleading of specifics,” a plaintiff must allege facts

15

sufficient to “raise a right to relief above the speculative level.” Twombly, 550 U.S. at 555, 570.

16

In deciding whether the plaintiff has stated a claim upon which relief can be granted, the

17

court accepts the plaintiff’s allegations as true and draws all reasonable inferences in favor of the

18

plaintiff. See Usher v. City of Los Angeles, 828 F.2d 556, 561 (9th Cir. 1987). However, the court

19

is not required to accept as true “allegations that are merely conclusory, unwarranted deductions of

20

fact, or unreasonable inferences.” In re Gilead Scis. Sec. Litig., 536 F.3d 1049, 1055 (9th Cir.

21

2008).

22

If the court dismisses the complaint, it “should grant leave to amend even if no request to

23

amend the pleading was made, unless it determines that the pleading could not possibly be cured

24

by the allegation of other facts.” Lopez v. Smith, 203 F.3d 1122, 1127 (9th Cir. 2000). In making

25

this determination, the court should consider factors such as “the presence or absence of undue

26

delay, bad faith, dilatory motive, repeated failure to cure deficiencies by previous amendments,

27

undue prejudice to the opposing party and futility of the proposed amendment.” Moore v. Kayport

28

Package Express, 885 F.2d 531, 538 (9th Cir. 1989).
4

1
2
3

DISCUSSION
I.

MISAPPROPRIATION OF TRADE SECRETS (COUNTS I AND II)
Teradata alleges that SAP misappropriated its trade secrets under the federal Defend Trade

4

Secrets Act (“DTSA”), 18 U.S.C. § 1836 et seq., and California’s Uniform Trade Secrets Act

5

(“CUTSA”), Cal. Civ. Code § 3426 et seq. The elements of trade secret misappropriation are

6

substantially the same under the federal and state statutes. See Waymo LLC v. Uber Techs., Inc.,

7

No. 17–CV–00939–WHA, 2017 WL 2123560, at *7 (N.D. Cal. May 15, 2017) (stating that “the

8

California Uniform Trade Secrets Act and the federal Defend Trade Secrets Act…offer essentially

9

the same definitions for our purposes.”).

10

A trade secret is defined to include “all forms and types” of information that derives value

United States District Court
Northern District of California

11

from being secret and that the owner took reasonable measures to keep secret. 18 U.S.C. §

12

1839(3); see also Cal. Civ. Code § 3426.1(d). Misappropriation is the “acquisition of a trade

13

secret” by a person who knows or should know the secret was acquired by improper means. 18

14

U.S.C. § 1839(5); Cal. Civ. Code § 3426.1(b). The term “improper means” includes, among other

15

ways, misrepresentation, theft, or breach of a duty to maintain secrecy. 18 U.S.C. § 1839(6); Cal.

16

Civil Code § 3426.1(a).

17

SAP moves to dismiss both trade secret misappropriation claims on the basis that the FAC

18

fails to identify the trade secrets and to allege misappropriation prohibited by the Bridge Project

19

agreements. Further SAP argues that the claims are barred by applicable statutes of limitations

20

and that the DTSA does not have retroactive effect. I address each argument in turn.

21

A.

22

To identify a trade secret, Teradata “need not spell out the details of the trade secret” the

The FAC Does Not Sufficiently Allege Trade Secrets

23

claim is based on, see Space Data Corp. v. X, No. 16–CV–03260–BLF, 2017 WL 5013363, at *2

24

(N.D. Cal. Feb. 16, 2017) (internal quotation marks and citation omitted), but it must “describe the

25

subject matter of the trade secret with sufficient particularity to separate it from matters of general

26

knowledge in the trade or of special knowledge of those persons…skilled in the trade.” See Imax

27

Corp. v. Cinema Techs., Inc., 152 F.3d 1161, 1164–65 (9th Cir. 1998) (internal quotation, citation,

28

and emphasis omitted). The pleadings must give defendants “reasonable notice of the issues
5

1

which must be met at the time of trial and…provide reasonable guidance in ascertaining the scope

2

of appropriate discovery.” Diodes, Inc. v. Franzen, 260 Cal. App. 2d 244, 252–53 (1968).

3

The FAC alleges trade secret information related to massively parallel processing

4

(“MPP”). See FAC ¶¶ 1, 16–23. Teradata contends that its trade secrets include information “on

5

the design and optimization of Teradata’s MPP systems and the execution of analytical queries in

6

such systems.” FAC ¶ 34. These trade secrets, information and techniques for optimizing the

7

integration and analysis of massive data, are repeated several times. See FAC ¶¶ 24, 32, 34, 35.

8

In addition, Teradata asserts that with its trade secrets, SAP “optimize[d] the processing of certain

9

Open SQL queries for large volumes of data, enabling improved performance speed and

10

United States District Court
Northern District of California

11

opportunities for parallel processing and other enhancements on SAP’s HANA.” FAC ¶ 44.
At the hearing, Teradata repeated its reliance on the paragraphs of the complaint discussed

12

above, arguing that allegations concerning its proprietary methods related to MPP and executing

13

queries were enough to survive the motion to dismiss. However, several cases in this district

14

demonstrate what is currently lacking in Teradata’s FAC. In Farhang v. Indian Inst. of Tech.,

15

Kharagpur, the Hon. Ronald Whyte dismissed a trade secret claim defined as “specific business

16

models and implementations” related to a “core technology.” Farhang v. Indian Inst. of Tech.,

17

Kharagpur, No. 08–CV–02658–RMW, 2010 WL 2228936, at *14 (N.D. Cal. June 1, 2010).

18

Judge Whyte found defendants, on these allegations, could not determine what subject matter of

19

the trade secret was separate from matters of general knowledge in the trade. Id.

20

Similarly, in Synopsys, Inc. v. ATopTech, Inc., the Hon. Samuel Conti dismissed a trade

21

secret defined as information including “proprietary input and output formats, scripts, and

22

technical product documentation…” Synopsys, Inc. v. ATopTech, Inc., No. 13–CV–02965–SC,

23

2013 WL 5770542, at *6 (N.D. Cal. Oct. 24, 2013). Judge Conti found this too vague to

24

determine “where trade secret protection begins and ends as to any of this material.” Id.

25

Finally, in Vendavo, Inc. v. Price f(x) AG, the Hon. Richard Seeborg analyzed a trade

26

secret claim that allegedly included things like “…negative knowhow learned through the course

27

of research and development, and other information related to the development of its price-

28

optimization software, including ideas and plans for product enhancements.” Id. at *3 (internal
6

1

quotations omitted). Vendavo, Inc. v. Price f(x) AG, No. 17–CV–06930–RS, 2018 WL 1456697,

2

at *4 (N.D. Cal. Mar. 23, 2018). Judge Seeborg dismissed these allegations, finding they were too

3

broad. Id.

4

Teradata’s alleged trade secret information is too broad and vague. The repeated

5

allegations that its trade secrets include specific ways to optimize data with its MPP technology is

6

indistinguishable from the allegations in Farhang, Synopsis, and Vendavo. Even the most specific

7

allegation that the trade secret information included an optimization process for “certain Open

8

SQL queries” suffers the same flaw. FAC ¶ 44. As it stands, there are no allegations suggesting

9

what the proprietary information regarding optimization was, or how or why it is proprietary

10

United States District Court
Northern District of California

11

besides simply being labeled a trade secret by Teradata in its first amended complaint.
Teradata informed the court of its intention to disclose a trade secrets list to opposing

12

counsel. Although Teradata need not “spell out the details” of its trade secrets, Space Data Corp.,

13

2017 WL 5013363, at *2, more particularity is needed to separate Teradata’s alleged trade secret

14

information on design and optimization from other information that is “general knowledge in the

15

trade or of special knowledge of those persons who are skilled in the trade.” Diodes, 260 Cal.

16

App. at 253.

17

B.

18

SAP contends the claim must be dismissed because Teradata failed to comply with its

The FAC Sufficiently Alleges Misappropriation by Improper Means

19

contractual obligation to designate information as confidential when it is disclosed and failed to

20

plead that SAP breached the Bridge Project agreements. See Convolve Inc. v. Compaq Computer

21

Corp., 527 Fed. App’x 910, 924–25 (Fed. Cir. 2013) (finding “a duty to maintain the secrecy of

22

the disclosed information is dictated by the terms of the NDA.”).

23

SAP relies on cases finding a “written non-disclosure agreement supplants any implied

24

duty of confidentiality that may have existed between the parties.” Marketel Inter., Inc. v.

25

Priceline.com, Inc., 36 Fed. Appx. 423, 425 (Fed. Cir. 2002) (citing Union Pac. R.R. Co. v.

26

Mower, 219 F.3d 1069, 1076 (9th Cir. 2000)). In response, Teradata argues that whether it

27

complied with the Bridge Project agreements is a premature question at the motion to dismiss

28

stage because this requirement is not determinative of the claim. See PQ Labs, Inc. v. Yang Qi,
7

1

No. 12–CV–0450–CW, 2014 WL 334453, at *4 (N.D. Cal. Jan. 29, 2014) (finding the marking

2

requirement was irrelevant at summary judgment because plaintiff “has presented evidence that it

3

used other means to notify its employees and agents that its technological and customer

4

information was confidential.”).

5

As an initial matter, it is appropriate to consider the Bridge Project agreements on a motion

6

to dismiss where they have been incorporated by reference in the complaint. Incorporation by

7

reference is “a judicially created doctrine that treats certain documents as though they are part of

8

the complaint itself.” Khoja v. Orexigen Therapeutics, Inc., 899 F.3d 988, 1002 (9th Cir. 2018).

9

A defendant may seek to incorporate documents into the complaint “if the plaintiff refers
extensively to the document or the document forms the basis of the plaintiff’s claim.” Id. (internal

11
United States District Court
Northern District of California

10

citations and quotations omitted).

12

Here, SAP seeks to incorporate five documents for review in support of its motion to

13

dismiss. See Lanier Decl. Ex. A-E (Dkt. No. 38-7). There appears to be no dispute of the

14

authenticity of Exhibits A-C, respectively the SDCA, TPA, and a 2009 mutual non-disclosure

15

agreement. It is appropriate to incorporate these documents by reference in the complaint under

16

Khoja. However, Exhibits D and E are introduced on “information and belief” and are the subject

17

of dispute. For the reasons explained more fully in the forthcoming copyright infringement

18

analysis, Exhibits D and E will not be incorporated by reference. I now turn to SAP’s

19

misappropriation arguments.

20

1.

Designating Confidential Information

21

Teradata’s reliance on PQ Labs is persuasive. There, the court found that evidence

22

plaintiffs took other measures to protect information beyond a non-disclosure agreement made the

23

issue of marking documents confidential irrelevant. Id. The facts included explicitly warning

24

defendant not to disclose any product design concepts and placing other controls on access to

25

company research. Id.; see also Vesta Corp. v. Amdocs Mgmt., Ltd, No. 14–CV–01142–HZ, 2018

26

WL 4354301, at *15 (D. Or. Sept. 12, 2018) (finding the case law “unclear about the extent to

27

which compliance with an NDA is required to succeed on a claim for trade secret

28

misappropriation.”).
8

SAP counters that the facts here are more like Gemisys Corp. v. Phoenix Am., Inc., 186

1
2

F.R.D. 551 (N.D. Cal. 1999), in which the court dismissed a trade secret claim on summary

3

judgment for failing to mark the trade secrets as confidential, than like PQ Labs which did not

4

involve an NDA with a marking requirement. See Reply at 4 (Dkt. No. 47-4). This does not make

5

the dicta in PQ Labs, which stated the marking requirement was irrelevant, any less applicable. In

6

PQ Labs, the NDA marking requirement was irrelevant because there remained a question of fact

7

whether plaintiffs notified defendants of confidential information through other means. See 2014

8

WL 334453, at *4. Teradata is correct that at the motion to dismiss stage, the mutual non-

9

disclosure agreements could be non-determinative of the outcome.
The mutual non-disclosure agreements were only two of four contracts involved in the

10

United States District Court
Northern District of California

11

Bridge Project to ensure that Teradata’s proprietary information would not be misappropriated or

12

reverse engineered. The SDCA and TPA reference the NDA’s confidentiality terms but

13

independently also place restrictions on reverse engineering or creating derivative works of

14

software. See Lanier Decl. Ex. A §§ 10.7-8; Ex. B § 8.6. In addition, Teradata pleads specific

15

instances in which it provided SAP information subject to the “parties’ agreements” which refers

16

not only the NDAs, but to the SDCA, TPA, and EUL. See, e.g., FAC ¶ 34 (training sessions on

17

Teradata’s database); ¶ 35 (identifying solutions for SAP’s software based on confidential

18

solutions Teradata implements in its own products); ¶ 36 (installing Teradata Database at SAP’s

19

COIL facility in Palo Alto, California and its research center in Walldorf, Germany); ¶ 36

20

(providing access to Teradata Express subject to EUL). Given the other agreements Teradata

21

entered to protect its trade secrets, the NDA’s are not determinative of the claim at the pleading

22

stage.

23
24

2.

Bridge Project Agreements

Although PQ Labs recognized that an NDA may not be determinative of a trade secret

25

misappropriation claim, Teradata’s other measures to protect its information also lie in the Bridge

26

Project agreements. SAP contends the trade secrets, as alleged, are barred by the terms of the

27

SDCA in three respects.

28

First, the SDCA grants SAP a license to use “Input” in any SAP products, which is defined
9

1

as “suggestions, comments, and feedback.” Lanier Decl. Ex. A §§ 1.6, 9.4. In response, Teradata

2

argues that it has alleged it provided SAP with trade secret information beyond “Input.” See FAC

3

¶¶ 32-36. I agree with Teradata. The FAC, even without sufficiently detailing the trade secrets,

4

already states that Teradata provided SAP with “information, software, tools, and other materials,”

5

FAC ¶ 33, “conducted training sessions,” FAC ¶ 34, and conveyed “techniques for optimizing the

6

speed and efficiency” of the joint product. FAC ¶ 35. Taking these allegations as true, it is

7

plausible that SAP misappropriated information beyond the terms of the SDCA.

8
9

Second, the SDCA provides for Teradata to share “optimization” support with SAP. See
id. Ex. A, Appx. 2 (Task 5B). SAP argues that the alleged trade secrets were limited to
optimization support based on allegations that Teradata conveyed “techniques for optimizing”

11
United States District Court
Northern District of California

10

massive amounts of data, see FAC ¶¶ 32,35, and that it shared the “optimization of Teradata’s

12

MPP systems.” FAC ¶ 34. However, Teradata asserts again that the information it provided went

13

beyond “optimization” as identified above. Teradata also argues that regardless of the agreement,

14

it did not authorize SAP to use the information for its competing product.

15

Third, SAP contends that the SDCA grants it ownership of “Newly Developed Materials”

16

resulting from the Bridge Project. See id. §§ 1.8, 10.3. The agreement defined new materials as

17

those “developed by SAP and/or [Teradata] in connection with or as a result of a party’s

18

interaction with the other party within the context of this Agreement…” Id. § 1.8. As with the

19

argument that Teradata agreed to share “optimization” support to SAP, there is no indication from

20

the agreement that it contemplated using the information outside the context of the Bridge Project,

21

let alone for a competing product. SAP, in its reply, argues that it would own any intellectual

22

property “if SAP learned something proprietary from Teradata during the Bridge Project.” Reply

23

at 5. Even this argument, however, does not extend to Teradata’s existing trade secrets, which it

24

alleges it developed before the Bridge Project and which was taken from it under the pretense of

25

the agreement. See FAC ¶ 1 (“SAP then stole Teradata’s trade secrets (accumulated by Teradata

26

over the course of four decades in the EDAW space)…”); ¶ 34 (“Teradata provided to SAP

27

proprietary, confidential, and trade secret information acquired through decades of research and

28

development.”).
10

1
2

Accordingly, assuming the allegations in the complaint as true, the FAC contains plausible
allegations that there is misappropriation by improper means.

3

C.

4

SAP contends that regardless of the pleadings, Teradata’s trade secret misappropriation

5

claims are barred by the Bridge Project SDCA’s two-year limitations period, and the three-year

6

statutory limitations period. These arguments are not well taken.

7
8

The Trade Secret Misappropriation Claims Are Not Time Barred

1.

The SDCA Time Bar Is Not In Effect and Would Not Apply

It is a well-settled point of law that parties may contract around a limitations period shorter
than the default statute of limitations. See, e.g., W. Filter Corp. v. Argan, Inc., 540 F.3d 947, 952

10

(9th Cir. 2008). Section 8.5 of the SDCA provides “Any claims for damages” must be filed within

11
United States District Court
Northern District of California

9

two years after the party becomes “aware of the event giving rise to the claim.” Lanier Decl. Ex.

12

A § 8.5. The SDCA also contains a survival clause in Section 13.4 that provides “upon such

13

effective date of termination, each Party’s rights and obligations hereunder will terminate…” but

14

the terms in Articles 9 through 13 survive. Id. at § 13.4.

15

Teradata argues that the SDCA time bar in Section 8.5 no longer applies after SAP

16

terminated the Bridge Project. See id. § 13.4; FAC ¶ 40. SAP insists that there is a presumption

17

that dispute resolution provisions survive the termination of a contract. See Marcotte v. Micros

18

Sys., Inc., 2014 WL 4477349, at *9 (N.D. Cal. Sept. 11, 2014) (citing Litton Fin. Printing Div. v.

19

NLRB, 501 U.S. 190, 204 (1991) (finding “a presumption in favor of postexpiration arbitration of

20

matters unless negated expressly or by clear implication.”) (internal quotations omitted).

21

I find that the SDCA two-year limitations period is not in effect. The exception in Litton

22

concerned arbitration rights that “accrued or vested under the agreement, or where, under normal

23

principles of contract interpretation, the disputed contractual right survives expiration of the

24

remainder of the agreement.” Litton, 501 U.S. at 206. Litton’s holding has also applied to forum

25

selection clauses. See Marcotte v. Micros Sys., Inc., 2014 WL 4477349, at *9 (N.D. Cal. Sept. 11,

26

2014) (citing Saleemi v. Gosh Enterprises, Inc., 467 F. App'x 744 (9th Cir. 2012)). SAP seeks to

27

extend it further, but I am not aware of, and the parties have not cited, any cases in this jurisdiction

28

that have directly applied Litton to a contractual limitations provision.
11

1

The only case SAP provides applying the presumption of survivability to a contractual

2

limitations provision is a Massachusetts District Court case, Creative Playthings Franchising,

3

Corp. v. Reiser, 2011 WL 13250940, at *3 (D. Mass. Apr. 28, 2011), which found a liability

4

limitation provision was “in the nature” of other dispute resolution provisions. The Ninth Circuit

5

has found, however, that in California “contractual stipulations are not favored” and should be

6

“construed with strictness against the party invoking them” because they are in “derogation” of the

7

statutorily set limitation. W. Filter Corp., 540 F.3d at 952 (internal quotation and citation

8

omitted).

9

Construing the agreement strictly against SAP, the contract language of Section 13.4 is
clear that only Articles 9 through 13 survive after the agreement is terminated. Accordingly,

11
United States District Court
Northern District of California

10

absent clear language to the contrary, principles of contract construction require that I not give

12

Section 8.5’s contractual limitations clause post-termination effect.

13

Assuming the contractual limitation was in effect, SAP also argues that Teradata has not

14

alleged “willful misconduct,” which the SDCA identifies as an exception to the two-year

15

limitations period. Willful misconduct has been defined similarly in various contexts. See, e.g.,

16

Manuel v. Pac. Gas & Elec. Co., 173 Cal. App. 4th 927, 947 (2009) (defining willful misconduct

17

as involving a “positive intent actually to harm another or to do an act with a positive, active and

18

absolute disregard for its consequences.”) (quotation marks and citation omitted). In the CUTSA,

19

the statute states if “willful” misappropriation exists, the court can award exemplary damages. See

20

Cal. Civ. Code § 3426.3(c). In turn, the California Code of Civil Procedure defines “malice”

21

related to exemplary damages as conduct intended to cause injury or conduct “with a willful and

22

conscious disregard of the rights” of others. See id. at § 3294(c)(1).

23

Teradata has sufficiently pleaded willful and conscious disregard of its rights. The FAC

24

alleges its trade secrets were improperly misappropriated (as discussed above) when SAP

25

intentionally stole its information by entering the Bridge Project under false pretenses. See FAC ¶

26

52. Additionally, Teradata alleges that Dr. Sikka was the driving force within SAP behind the plot

27

to steal Teradata’s trade secret information and incorporate it into the HANA product. See id. ¶¶

28

1, 44, 45, 47, 52. At least at the pleading stage, Teradata has alleged facts that plausibly support
12

1

its theory of a willful and conscious plan by SAP and Dr. Sikka to steal its trade secrets in direct

2

disregard of the Bridge Project, to develop a competing product, and to engage in anticompetitive

3

practices. Therefore, even if the SDCA limitations period was in effect after the agreement was

4

terminated, Teradata’s allegations fit within the exception such that the SDCA’s time bar would

5

not apply.

6
7

2.

Teradata Pleads a Claim Within the Statutory Limitations Period

Both the DTSA and the CUTSA have a three-year statute of limitations. See 18 U.S.C. §

8

1836(d); Cal. Civ. Code § 3426.6. Teradata alleges that its trade secrets were improperly acquired

9

during the Bridge Project from 2009 to 2011 when SAP developed HANA, and that the
information continues to be misused in the present. FAC ¶¶ 40, 52. Yet it asserts that it learned of

11
United States District Court
Northern District of California

10

the theft in September 2015, only after a Der Spiegel article revealed SAP’s conduct. FAC ¶¶ 47,

12

51. According to the article, SAP concealed an internal investigation from a SAP auditor in 2012

13

who concluded that SAP had stolen Teradata’s intellectual property during the Bridge Project.

14

FAC ¶¶ 47, 48, 51.

15

SAP’s motion essentially contests the truthfulness of these allegations. It proposes that

16

Teradata should have been aware of potential misappropriation in 2011 when HANA was

17

developed, making the misappropriation claim time barred unless there are allegations that it

18

investigated its suspicions. See Fox v. Ethicon Endo-Surgery, Inc., 35 Cal. 4th 797, 808 (2005)

19

(“a potential plaintiff who suspects than injury has been wrongfully ceased must conduct a

20

reasonable investigation.”); see also Jolly v. Eli Lilly & Co., 44 Cal. 3d 1103, 1111 (1988) (finding

21

the plaintiff must go find the facts “[s]o long as a suspicion exists.”). At the hearing, SAP also

22

referred to paragraphs in the complaint purportedly showing Teradata was on notice of potential

23

misappropriation in 2011. See FAC ¶¶ 3, 38–40, 52.

24

For three reasons, however, Teradata’s FAC does not reveal a reasonable basis for

25

suspicion of misappropriation in 2011. First, the general claim early in the FAC that SAP “could

26

not have so quickly developed and marketed HANA…without its theft of Teradata’s trade secrets”

27

is not a veiled admission of some reasonable suspicion in 2011. FAC ¶ 3. The FAC dedicates an

28

entire section on SAP’s quick development of HANA, see FAC ¶¶ 38–46, and a section on the
13

1

discovery of the theft in which it pleads that based on the Der Spiegel article “it has become

2

clear…SAP was able to go to market so quickly only because SAP entered into an agreement with

3

Teradata…and then incorporated [trade secrets] into and used them to develop HANA.” FAC ¶

4

52 (emphasis added).

5

Second, SAP’s announcement of HANA for SAP BW “just days” after SAP terminated the

6

Bridge Project in 2011 also does not provide a reasonable suspicion of misappropriation because,

7

as the FAC alleges, the product did not perform as well as Teradata’s MPP database. See FAC ¶¶

8

41, 16–18. SAP argued at the hearing that the announcement alone should have raised suspicions

9

and a duty to investigate for misappropriation. Perhaps if the product performed similarly to

10

Teradata’s it would have raised suspicions, but that is not alleged in the FAC.

United States District Court
Northern District of California

11

Finally, the Teradata employees working on HANA during the Bridge Project and those

12

who left to work for SAP later were allegedly not known to Teradata until after the Der Spiegel

13

article. See FAC ¶ 45 (“Teradata was not aware of this cross-pollination between SAP’s Bridge

14

Project and HANA development teams.”). This is another allegation that suggests Teradata lacked

15

notice of misappropriation in 2011.

16

Teradata need not plead that it initiated an investigation sometime after 2011 when it also

17

pleads that it lacked any reasonable suspicion of misappropriation until 2015. Its trade secret

18

claim did not accrue until it discovered the infringement in September 2015. Because the

19

complaint was filed on June 19, 2018, it was within the three-year statute of limitation under the

20

DTSA and CUTSA. The misappropriation claims are not barred.

21

D.

22

The DTSA applies to “any misappropriation of a trade secret…for which any act occurs on

The DTSA Can be Based on a Continuing-Use Theory

23

or after [May 11, 2016,] the date of the enactment of [the] Act.” Defend Trade Secrets Act of

24

2016, Pub. L. No. 114–153, 130 Stat. 376, 381–82 (May 11, 2016). When bringing a claim under

25

the DTSA, the plaintiff may rely on one of three theories: “(1) acquisition, (2) disclosure, or (3)

26

use.” Cave Consulting Grp., Inc. v. Truven Health Analytics, Inc., No. 15–CV–02177–SI, 2017

27

WL 1436044, at *4 (N.D. Cal. Apr. 24, 2017). To state a claim under any of these theories,

28

plaintiffs need to allege “that acts of misappropriation occurred after DTSA came into effect.”
14

1

Avago Techs. U.S. Inc. v. Nanoprecision Prods, Inc., No. 16–CV–3737–JCS, 2017 WL 412524, at

2

*9 (N.D. Cal. Jan 31, 2017) (emphasis in original).

3

If the trade secret was “publicly disclosed before the effective date of the DTSA, a plaintiff

4

cannot rely on a theory ‘that the same information was disclosed again.’” Veronica Foods Co. v.

5

Ecklin, No. 16–CV–07223–JCS, 2017 WL 2806706, at *13 (N.D. Cal. June 29, 2017) (quoting

6

Avago, 2017 WL 412524, at *8–9) (emphasis in original). However, where the information was

7

never initially disclosed, “a DTSA claim based on continuous use after May 11, 2016 does not

8

require allegations of ‘new or different’ misappropriation, unlike a claim based on disclosure of

9

trade secrets.” Order Denying Defendants’ Motion to Dismiss, Space Data Corp. v. X, Alphabet,

10

United States District Court
Northern District of California

11

Inc., et al., No. 16–cv–03260–BLF (N.D. Cal. 2016) (Dkt. No. 176).
The parties dispute whether the misappropriation claim can have retroactive effect. As

12

discussed above, Teradata alleges a misappropriation that occurred during the Bridge Project but

13

was discovered in September 2015. FAC ¶¶ 47, 51. The misappropriation allegedly continues to

14

the present as SAP keeps selling HANA which, according to Teradata, was only possible because

15

of the trade secrets that were stolen from it. FAC ¶¶ 41, 54, 102–103. SAP argues that the DTSA

16

claim should be dismissed because the federal statute does not have retroactive effect for the

17

continued use of the same misappropriated information before DTSA was enacted.

18

SAP misapplies the law to the facts of this case. In Avago, the Hon. Joseph Spero held

19

there is “no authority suggesting that the DTSA allows a misappropriation claim to be asserted

20

based on the continued use of information that was disclosed prior to the effective date of the

21

statute.” Id. Next, SAP points to Space Data Corp., where the Hon. Beth Freeman found that

22

plaintiff failed to allege a continuing-use theory when the information was disclosed prior to the

23

DTSA and failed to allege any new post-enactment misappropriation. Space Data Corp. v. X, No.

24

16–CV–03260–BLF, 2017 WL 3007078, at *3 (N.D. Cal. July 14, 2017). Finally, SAP relies on

25

Cave Consulting, where the Hon. Susan Illston dismissed DTSA claims for failing to allege post-

26

enactment uses in the first instance. 2017 WL 1436044, at *5. Consistent with cases where the

27

trade secret is previously disclosed, Judge Illston found plaintiff failed to state a DTSA claim

28

without “new or somehow different [information] from the prior misappropriation.” Id.
15

1

Unlike those cases, however, the trade secret information alleged here has not been

2

publicly disclosed or “extinguished” at the time it was initially misappropriated. See, e.g., Avago,

3

2017 WL 412524, at *1-3 (quoting Ultimax Cement Mfg. Corp. v. CTS Cement Mfg. Corp., 587

4

F.3d 1339, 1355 (Fed. Cir. 2009)); Attia v. Google LLC, No. 17–CV–06037–BLF, 2018 WL

5

2971049, at *11 (N.D. Cal. June 13, 2018) (dismissing claim for lack of allegations differentiating

6

between secrets disclosed in patent application and secrets remaining that were not disclosed prior

7

to DTSA effective date); Veronica Foods, 2017 WL 2806706, at *13 (stating the rule in Avago

8

applied where the trade secret was allegedly “publicly disclosed before the effective date.”).
Teradata alleges continuing-use of its trade secrets; that is not disputed. However, there

10

are no factual allegations in this case that Teradata’s trade secret information was ever publicly

11
United States District Court
Northern District of California

9

disclosed pre- or post-enactment of the DTSA. Therefore, assuming Teradata can amend the

12

complaint to sufficiently identify its trade secret information as outlined above, it has adequately

13

brought a DTSA claim on a continuing-use theory.

14

II.

COPYRIGHT INFRINGEMENT (COUNT III)

15

The FAC alleges copyright infringement prohibited by Teradata’s end-user license

16

(“EUL”). SAP moves to dismiss the copyright claim because there is no breach of any obligations

17

under the parties’ agreements, and because the claim is time-barred under the Bridge Project

18

agreements and 17 U.S.C. § 507(b) of the Copyright Act.

19

SAP’s interpretation of the Bridge Project agreements that are incorporated by reference in

20

the complaint fall short when it comes to the EUL. To restate the law briefly, the Ninth Circuit in

21

Khoja v. Orexigen Therapeutics, Inc. clarified that incorporation by reference allows courts to

22

treat documents as if they are a part of the complaint “if the plaintiff refers extensively to the

23

document or the document forms the basis of the plaintiff’s claim.” 899 F.3d at 1002 (internal

24

citations and quotations omitted). However, “if the document merely creates a defense to the

25

well-pled allegations in the complaint, then that document did not necessarily form the basis of the

26

complaint.” Id.

27
28

Here, SAP asserts the Bridge Project agreements allowed it to reverse engineer Teradata’s
software within the terms of the agreement and that the EUL has its own two-year limitations
16

provision that is distinct from the SDCA terms addressed earlier. However, SAP’s request to

2

incorporate the EUL by reference is based on “information and belief” that the exhibit has the

3

same license language as the EUL referenced in the first amended complaint. Lanier Decl. ¶ 6,

4

Ex. E. I cannot know or assume on SAP’s information and belief that the attached EUL is the

5

same as the EUL plaintiff extensively refers to in the FAC merely because the FAC alleges the

6

EUL was a “standard” agreement. On the specific contract interpretation arguments raised here,

7

treating a different EUL like it contains the same terms as the one referenced in the FAC would

8

permit what the court in Khoja cautioned; it would allow SAP to create a defense to an otherwise

9

well-pleaded allegation. See 899 F.3d at 1002. Without knowing the terms of the relevant EUL,

10

SAP cannot prove that a contractual limitation provision bars the claim, that reverse engineering

11
United States District Court
Northern District of California

1

for the purpose of interoperability or circumvention was permitted under the EUL, or that no set of

12

facts exists for Teradata to raise a plausible copyright claim.

13

SAP’s remaining argument is that, regardless of the claim, the FAC is time-barred under

14

the three-year statute of limitations in the Copyright Act. This is incorrect. The statute of

15

limitations begins to run “when one has knowledge of a violation or is chargeable with such

16

knowledge.” Roley v. New World Pictures, Ltd., 19 F.3d 479, 481 (9th Cir. 1994). Each

17

infringing act is considered a distinct harm so the statute of limitations “bars infringement claims

18

that accrued more than three years before suit was filed, but does not preclude infringement claims

19

that accrued within the statutory period.” Oppenheimer v. Allvoices, Inc., No. 14–CV–00499–LB,

20

2014 WL 2604033, at *3 (N.D. Cal. June 10, 2014). The point when a plaintiff knew or should

21

have known about an infringement is a “question of fact,” Polar Bear Prods., Inc. v. Timex Corp.,

22

384 F.3d 700, 707 (9th Cir. 2004), and courts may dismiss a claim where there is no reasonable

23

basis to conclude a lack of knowledge under the circumstances. Goldberg v. Cameron, 482

24

F.Supp.2d 1136, 1148 (N.D. Cal. 2007).

25

SAP asserts that Teradata failed to plead that it could not reasonably discover the

26

infringement within three years of 2011 when HANA for SAP BW was launched. This is a

27

mischaracterization of the FAC and appears to incorrectly presume that the claim accrued when

28

infringement began. Teradata alleges that it learned of the theft of its proprietary information after
17

1

the September 2015 Der Spiegel article. See FAC ¶¶ 47, 51; see also Oppo. at 3:18, 12:11. The

2

article revealed an internal SAP auditor’s conclusion that SAP stole Teradata’s intellectual

3

property during the Bridge Project, and that SAP concealed the investigation from Teradata and

4

the public until the article exposed the infringement. Id.
In sum, Teradata’s copyright claim did not accrue until it discovered the infringement in

5
6

September 2015. The complaint was filed on June 19, 2018, within the three-year statute of

7

limitation under 17 U.S.C. § 507(b) of the Copyright Act. Accordingly, the motion to dismiss the

8

copyright claim is DENIED.

9

III.

10

United States District Court
Northern District of California

11

ANTITRUST VIOLATIONS (COUNTS IV AND V)
Teradata’s antitrust claims include unlawful tying under 15 U.S.C. §§ 1, 14, and attempted

monopolization under 15 U.S.C. § 2. SAP moves to dismiss each for failure to state a claim.

12

A.

13

In a tying arrangement the seller conditions one product, the tying product, on the buyer’s

Unlawful Tying

14

purchase of another product, the tied product, to extend its market power in a distinct product

15

market. See Cascade Health Sols. v. PeaceHealth, 515 F.3d 883, 912 (9th Cir. 2008). A tying

16

arrangement is “forbidden on the theory that, if the seller has market power over the tying product,

17

the seller can leverage this market power through tying arrangements to exclude other sellers of

18

the tied product.” Id.

19

Teradata appears to accuse SAP of a per se tying violation. See, e.g., Oppo. at 16:13–18.

20

To sufficiently bring a per se tying claim, Teradata must plead that: (i) SAP “tied together the sale

21

of two distinct products or services;” (ii) SAP has the “economic power in the tying product

22

market to coerce its customers into purchasing the tied product;” and (iii) the arrangement “affects

23

a not insubstantial volume of commerce in the tied product market.” Cascade Health Sols., 515

24

F.3d at 913 (internal citation and quotation omitted). SAP moves to dismiss, arguing that there are

25

deficient allegations of tied products, coercion, and market power, and that the claim is subject to

26

the rule of reason.

27
28

1.

The FAC Sufficiently Alleges SAP Tied Two Products

SAP contends that S/4HANA is one integrated product and therefore the first element of
18

1

the tying claim cannot be satisfied. See Int’l Mfg. Co. v. Landon, Inc., 336 F.2d 723, 730 (9th Cir.

2

1964) (“it is not an unlawful tying arrangement for a seller to include several items in a single

3

mandatory package when the items may be reasonably considered to constitute parts of a single

4

distinct product.”). At the hearing, SAP repeated that the allegations in paragraphs 69, 73, 86, 87,

5

88, and 129 of the FAC support that S/4HANA is an integrated product. See, e.g., FAC ¶ 73

6

(“SAP developed HANA to function as both a transactional database for managing ERP

7

Applications data and an analytical database with EDAW functionality.”). But Teradata asserts

8

that it sufficiently alleged a market for different products, the S/4HANA product in the ERP

9

Applications market and the HANA product in the EDAW market.

10

The existence of distinct products depends on “the character of the demand for the two

United States District Court
Northern District of California

11

items.” Jefferson Par. Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 19 (1984). This consideration,

12

known as the purchaser demand test, examines direct and indirect evidence of consumer demand

13

and whether defendants “foreclosed competition on the merits in a product market distinct from

14

the market for the tying item.” Id. at 21. Direct evidence of demand includes “whether, when

15

given a choice, consumers purchase the tied good from the tying good maker, or from other

16

firms.” Rick-Mik Enterprises, Inc. v. Equilon Enterprises LLC, 532 F.3d 963, 975 (9th Cir. 2008)

17

(internal citations and quotations omitted). Indirect evidence includes firm behaviors, for instance

18

a single product is apparent if “competitive firms always bundle the tying and tied goods”

19

together. Id.

20

Here, the FAC contains plausible allegations directly and indirectly evidencing demand in

21

distinct markets. Teradata alleges that EDAW and ERP Applications products perform different

22

functions. FAC ¶ 69. “ERP Applications allow companies to gather and manage the data required

23

to conduct their day-to-day operations across many aspects of the business enterprise,” and

24

“typically are designed around a relational database that acts as a common repository for all the

25

data used and managed by the ERP Applications…” FAC ¶ 30. An EDAW product, in contrast to

26

ERP Applications and transactional databases, “involves the centralized storage and integration of

27

vast amounts of data collected from numerous sources across an entire business enterprise in its

28

day-to-day operations…” FAC ¶ 16. Teradata further describes the EDAW market as products
19

1

enabling Top-Tier ERP Applications customers to retain their data, and “to perform complex

2

analytical operations on, vast amounts of data from a wide variety of data streams (i.e., the

3

companies’ ERP Applications and numerous other sources).” FAC ¶ 68. EDAW products copy

4

customer ERP Applications data from the transactional database to incorporate it into the EDAW

5

system, which allows customers to “run complex analytical functions against all the data”

6

collected from the ERP data. FAC ¶ 70.
The FAC also alleges that “EDAW products are indispensable for” Top-Tier ERP

8

Application customers, but selling the products together is not necessary. See FAC ¶ 72. In fact,

9

EDAW and ERP Application products were historically sold separately to the same customer

10

base. FAC ¶ 69. Both before and after the Bridge Project, customers had the ability to select

11
United States District Court
Northern District of California

7

different ERP Applications and EDAW products when given the freedom to choose. FAC ¶¶ 77–

12

78, 84. When HANA was first released, SAP continued to allow its ERP Application customers to

13

choose their own transactional database and their own EDAW products. FAC ¶ 81. This was

14

purportedly the case until SAP made its HANA products inoperable with other EDAW databases.

15

FAC ¶¶ 86–87. After HANA’s release, customers still approached Teradata and encouraged it to

16

develop integration for HANA. FAC ¶ 84.

17

Nonetheless, SAP argues that the product is bundled, which does not necessarily amount to

18

an unlawful tying arrangement. See Cascade Health Sols., 515 F.3d at 915 n. 27. According to

19

SAP, because the bundling “innovates,” “integrates,” and “improves” the product, it should be

20

treated as a single product for purposes of an unlawful tying claim. See X.P. Areeda, H.

21

Hovenkamp & E. Elhauge, Antitrust Law, ¶ 1746b, p. 208 (3d ed. 2007) (explaining a single

22

product should be found where defendant “integrate[s] previously unbundled inputs into a new

23

product design that results in better combined performance than could be obtained if the items

24

were offered unbundled and combined by purchasers or intermediaries.”). It relies on two cases to

25

make this argument.

26

First, SAP asserts that I cannot balance the benefits of “a product improvement against its

27

anticompetitive effects.” Allied Orthopedic Appliances Inc. v. Tyco Health Care Grp. LP, 592

28

F.3d 991, 1000 (9th Cir. 2010) (finding product improvement alone does not render unlawful a
20

1

monopoly under Section 2). Since the product is an improvement, that would be the end of the

2

matter. The Allied Orthopedic case, however, was not concerned with the sufficiency of a

3

plaintiff’s tying allegations. Rather, the court analyzed whether “a design change that improves a

4

product by providing a new benefit to consumers” could defeat summary judgment of a Section 2

5

monopoly claim. Id. at 998–999.

6

Second, SAP contends that because there is a plausible claim that its integrated product

7

offers advantages “unavailable if the functionalities are bought separately and combined by the

8

purchaser,” there is no viable tying claim. United States v. Microsoft Corp., 147 F.3d 935, 948

9

(D.C. Cir. 1998). However, in Microsoft Corp., the court was operating under its reading of an
antitrust consent decree, and left open “[w]hether or not this is the appropriate test for antitrust law

11
United States District Court
Northern District of California

10

generally…” Id. at 950. SAP does not provide any Ninth Circuit case law applying the

12

plausibility test since Microsoft Corp.

13

Even under the case law SAP relies on, Teradata satisfies its burden to plead tied products.

14

Assuming the court could apply the Microsoft Corp. analysis consistent with antitrust law

15

generally, it is arguable whether there are allegations that the S/4HANA and HANA products are

16

integrated. As SAP pointed out, Teradata’s FAC recognizes that EDAW and EPR Applications

17

products can operate more efficiently together – as was similarly the aim of the Bridge Project and

18

allegedly achieved in Teradata Foundation. See FAC ¶ 32 (“A key challenge of the Bridge Project

19

was to ensure fast and efficient interoperation between SAP’s front-end systems and Teradata’s

20

EDAW product.”). But the allegations refer only to HANA as it relates to integrating with

21

transactional databases, not integrating with the ERP Applications, S/4HANA. See, e.g., FAC ¶

22

42 (stating that like Teradata Foundation, “SAP’s HANA product combines a database solution

23

with integrated software to perform data analytics.”); ¶ 52 (“…integrating the two companies’

24

technologies” as a reference to Teradata Foundation solution, not HANA or S/4HANA); ¶ 83

25

(alleging HANA “was ill-suited for integration of enterprise data from third-party sources.”).

26

Applying Allied Orthopedic, even if a design change leads to improvements it is tolerated

27

in the antitrust context “unless the monopolist abuses or leverages its monopoly power in some

28

other way when introducing the product.” Allied Orthopedic Appliances Inc., 592 F.3d at 1000.
21

An abuse of leverage “in some other way” is exactly what the FAC alleges. Id. For instance,

2

Teradata asserts that SAP is a dominant player in the ERP Application market and made its newest

3

version of the ERP Application, S/4HANA, incompatible with other EDAW products and

4

transactional databases, FAC ¶¶ 86–87, while at the same time announcing that it is ending

5

support for prior versions of its ERP Applications by 2025. FAC ¶ 89. Teradata also alleges that,

6

in addition to bundling the sales of HANA and S/4HANA, SAP placed restrictive language in its

7

licensing agreements preventing customers from using other databases, FAC ¶ 87, and restricted

8

customers’ ability to extract their ERP-derived data stored within HANA. FAC ¶ 93. These are

9

all changes that Teradata alleges are specifically geared toward extending SAP’s market power

10

from the ERP Application market to the distinct EDAW market. See Rick-Mik Enterprises, 532

11
United States District Court
Northern District of California

1

F.3d at 971. Accordingly, the FAC contains enough allegations satisfying the first element that

12

SAP tied together the sale of two distinct products.

13
14

2.

The FAC Sufficiently Alleges Coercion

SAP contests the second element, that Teradata has not plausibly alleged coercion. The

15

coercion element requires Teradata to “present evidence that the defendant went beyond

16

persuasion and coerced or forced its customers to buy the tied product in order to obtain the tying

17

product.” Paladin Assocs., Inc. v. Montana Power Co., 328 F.3d 1145, 1159 (9th Cir. 2003). The

18

Ninth Circuit recognizes so called “implied” or “de facto” tying claims in which a seller “adopts a

19

policy that makes it unreasonably difficult or costly to buy the tying product…without buying the

20

tied product.” Aerotec Int'l, Inc. v. Honeywell Int'l, Inc., 836 F.3d 1171, 1179 (9th Cir. 2016)

21

(quoting Collins Inkjet Corp. v. Eastman Kodak Co., 781 F.3d 264, 272 (6th Cir. 2015)).

22

SAP asserts that “products are not tied unless the supplier refuses to accommodate those

23

who prefer one without the other.” IX Antitrust Law, ¶ 1700i, p. 9. It focuses on language in

24

Foremost Pro Color, Inc. v. Eastman Kodak Co. that requires plaintiffs plead “some modicum” of

25

facts “that the purchase of the alleged tied products was required as a condition of sale of the

26

alleged tying products.” 703 F.2d 534, 540, 542 (9th Cir. 1983). Teradata has done so. SAP’s

27

most recent ERP Applications product, S/4HANA, is allegedly now combined with the sale of

28

HANA in a single offering, which requires customers who buy S/4HANA to buy HANA as their
22

1

EDAW product. See FAC ¶ 87. This is a direct allegation that the purchase of the tied product is

2

conditioned on the sale of the tying product. See Eastman Kodak, 703 F.2d at 542.

3

The claim becomes stronger when considering the tying allegations that imply coercion.
For example, the costs of implementing ERP Applications are allegedly extremely high, as are the

5

costs of switching ERP Applications products. A single customer invests tens of millions of

6

dollars on its ERP Applications each year. FAC ¶¶ 62–66. These expenses include licensing,

7

development, and implementation of ERP Applications for their unique businesses, as well as the

8

needs of training employees to use the product, troubleshooting issues with the product, migrating

9

data to the new service, and other expenses. FAC ¶¶ 63–64. In this context, SAP placed a sunset

10

on its existing ERP Applications and is ending support of prior versions by 2025. FAC ¶ 89. On

11
United States District Court
Northern District of California

4

Teradata’s information and belief, SAP also limited the updates for prior versions of its ERP

12

Applications. FAC ¶ 91. Therefore, existing ERP Applications customers can upgrade to

13

S/4HANA before the 2025 sunset period, which is allegedly tied to HANA for EDAW services, or

14

they can suffer the switching costs. Teradata alleges that when faced with this choice, customers

15

will upgrade so they can continue to have “the latest features and functionality, most robust

16

support and most recent security and software updates.” FAC ¶ 59.

17

SAP contends there is no coercion because customers can still buy stand-alone versions of

18

SAP’s Top-Tier ERP Applications without purchasing HANA. However, that is not an allegation

19

in the FAC. Even if it was alleged that customers can still buy stand-alone ERP Applications from

20

SAP, this does not reconcile with the other allegations such as the high costs of implementing

21

these systems, the lack of updates offered for older versions, and the expected 2025 sunset date for

22

older versions of the product. Accordingly, at the pleading stage, Teradata has alleged coercion.

23
24
25
26
27
28

3.

The FAC Sufficiently Alleges SAP’s Market Power in a Relevant
Market

Third, also within the second element of a tying claim, SAP challenges the market power
allegations. The definition of a “relevant market” in which defendant has market power is
typically a factual rather than legal question. Newcal Indus., Inc. v. Ikon Office Sol., 513 F.3d
1038, 1045 (9th Cir. 2008). The relevant market inquiry need not be pleaded with specificity to
23

1

survive a motion to dismiss. Id. Still, claims can be dismissed if they fail to satisfy certain legal

2

principles. Id.

3

In Newcal, the Ninth Circuit stated the relevant market: (i) must be “a product market,”

4

not one defined by consumers; (ii) must encompass the product as well as “all economic

5

substitutes for the product;” and (iii) may include a submarket if it is “economically distinct from

6

the general product market.” Id. Submarkets are identified by indicia like “industry or public

7

recognition of the submarket as a separate economic entity, the product’s peculiar characteristics

8

and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price

9

changes, and specialized vendors.” Brown Shoe v. United States, 370 U.S. 294, 325 (1962).

10

Teradata alleges that there is a product market for ERP Applications in which the

United States District Court
Northern District of California

11

customers are large-scale, complex enterprises, referred to in the FAC as the “Top-Tier ERP

12

Applications Market.” FAC ¶¶ 56, 131. In the Top-Tier ERP Applications market, Teradata

13

alleges that SAP holds a dominant position in the market of about 60% to 90%, with Oracle as its

14

only significant competitor. FAC ¶ 67. An additional source of SAP’s power in the market for

15

Top-Tier ERP Applications is the allegedly locked-in customer base due to the severe costs of

16

switching vendors discussed above. See FAC ¶¶ 63–64. This suffices to plead SAP’s market

17

power, and it does not appear that SAP disputes this aspect of the complaint.

18

Instead, SAP argues that other portions of the FAC are “internally contradictory” and make

19

it unclear whether the relevant ERP market includes competitors to SAP or is limited to SAP’s

20

ERP Applications alone. See Apple Inc. v. Psystar Corp., 586 F. Supp. 2d 1190, 1200 (N.D. Cal.

21

2008) (granting motion to dismiss where antitrust claim did not plausibly allege an independent

22

market because the allegations were internally contradictory). If the relevant market includes only

23

SAP’s products, SAP contends this is fatal to the claim. See Datel Holdings, Ltd. v. Microsoft

24

Corp., 712 F. Supp. 2d 974, 986 (N.D. Cal. 2010) (“In general, single brand markets do not

25

constitute a relevant market.”). SAP cites three “internally contradictory” paragraphs of the FAC

26

to that end. See FAC ¶¶ 56, 60, 129. I disagree that there are internal contradictions in the

27

complaint.

28

First, in paragraph 56 of the FAC, Teradata identifies the product market for ERP
24

1

Applications, “such as SAP’s S/4HANA and SAP’s predecessor ERP programs.” FAC ¶ 56. This

2

refers to the relevant market discussed above, made up of SAP and Oracle products. Referring to

3

SAP’s HANA product line as an exemplary product in the ERP Applications market does not

4

introduce a contradiction when, elsewhere in the pleading, it is clear that the market is also alleged

5

to include Oracle and its products. FAC ¶ 67 (“Oracle is the only other significant competitor for

6

these Top-Tier customers.”).

7

Second, in paragraph 60, Teradata alleges “there are not reasonable or adequate economic

8

substitutes for upgrades of SAP ERP Applications for the vast majority of Top-Tier ERP

9

Applications customers because they are locked-in to their current ERP application provider.”
FAC ¶ 60. This appears to refer to a submarket of customers in the ERP Application market over

11
United States District Court
Northern District of California

10

whom SAP has particularly strong control due to their locked-in relationship. It is legally

12

permissible to plead a submarket which is “economically distinct” from the general product

13

market (in this case distinct customers who are locked-in to SAP as their ERP Applications

14

vendor). Newcal, 513 F.3d at 1045. An alleged submarket of SAP customers who use SAP ERP

15

Applications products is entirely consistent with the general product market for ERP Applications

16

dominated by SAP with competition from Oracle.

17

Finally, paragraph 129 asserts that SAP’s Top-Tier ERP Applications, products like

18

S/4HANA, are a “separate and distinct” market from the market for HANA and EDAW products.

19

FAC ¶ 129. SAP argues it is contradictory to allege a market for SAP’s Top-Tier ERP

20

Applications products as well as a market for S/4HANA. I do not see the contradiction. In the

21

context of the alleged submarket of locked-in SAP customers, some customers may use older

22

versions of the ERP Applications and some may use S/4HANA. This paragraph is not

23

contradictory.

24

The FAC’s allegations that SAP exerted market power in a relevant market survive the

25

motion to dismiss. Teradata ultimately must prove the validity of the relevant market factual

26

element “subject to factual testing at summary judgment or trial.” Newcal, 513 F.3d at 1045.

27
28

4.

The Tying Claim Assessed Under the Rule of Reason

Finally, SAP contends the tying claim should be dismissed because it must be assessed
25

1

under the rule of reason, which Teradata fails to do. See Mot. to Dismiss at 21. Teradata alleges

2

SAP’s tying arrangement is both a per se violation and a rule of reason violation of the Sherman

3

Act. I do not need to decide whether the per se or rule of reason analysis applies; that is more

4

appropriate on a motion for summary judgment. See, e.g., In re High-Tech Employee Antitrust

5

Litig., 856 F. Supp. 2d 1103, 1122 (N.D. Cal. 2012) (finding it inappropriate to determine whether

6

rule or reason analysis applies until summary judgment). If one of the theories is plausible, that is

7

sufficient at this stage.

8

Under the rule of reason test, Teradata must plead: (1) a contract, combination or

9

conspiracy among two or more persons or distinct business entities; (2) by which the persons or
entities intended to harm or restrain trade or commerce [ ]; (3) which actually injures

11
United States District Court
Northern District of California

10

competition.” Kendall v. Visa U.S.A., Inc., 518 F.3d 1042, 1047 (9th Cir. 2008). In addition,

12

Teradata must plead an antitrust injury: “(4) that they were harmed by the defendant’s anti-

13

competitive contract, combination, or conspiracy…” Brantley v. NBC Universal, Inc., 675 F.3d

14

1192, 1197 (9th Cir. 2012) (citations omitted). It satisfies the test.

15
16

a.

Delineating a Relevant Tied Market

Starting with the first element, Teradata “must delineate a relevant market” to plead a

17

restraint on trade that ultimately harms competition in the tied market. Bhan v. NME Hosps., Inc.,

18

929 F.2d 1404, 1413 (9th Cir. 1991). As with SAP’s argument concerning the tying market for

19

Top-Tier ERP Applications, it contends that the tied market allegations for EDAW products are

20

internally contradictory. See Psystar Corp., 586 F. Supp. 2d at 1200. Teradata argues not that the

21

relevant market is EDAW products for Top-Tier Applications customers, FAC ¶ 68, but that it is

22

limited to existing SAP customers within a derivative aftermarket, FAC ¶¶ 129, 133. See Oppo. at

23

23:20–21.

24

SAP contends again that limiting the tied market to a single-product is impermissible,

25

while Teradata claims that it has adequately alleged a tied market comprised of a single-product

26

market for EDAW products used with SAP Top-Tier ERP Applications. Generally, “single brand

27

markets do not constitute a relevant market.” Datel Holdings Ltd., 712 F. Supp. 2d at 986.

28

However, the Supreme Court, in Eastman Kodak, adopted a limited exception for a single-product
26

1

“aftermarket” in which customers do not agree on restrictions that were undisclosed at the time of

2

the purchase of the product from the primary market. See Eastman Kodak Co. v. Image Tech.

3

Servs., 504 U.S. 451, 464–78 (1992). The actual existence of an aftermarket is a factual question.

4

Newcal, 513 F.3d at 1051.
In Newcal, the Ninth Circuit determined that allegations fit into the exception of Eastman

6

Kodak to survive a motion to dismiss because of four relevant aspects of the complaint. First, the

7

court found the complaint sufficiently alleged “two separate but related markets in intrabrand”

8

products and services. Id. at 1049. Teradata alleges the first market is the initial market for ERP

9

Applications products for Top-Tier ERP Applications customers, in which SAP and Oracle are the

10

only major competitors. The second market is the derivative aftermarket for EDAW products that

11
United States District Court
Northern District of California

5

are “specifically designed for” SAP customer’s specific ERP Application and are not

12

“interchangeable” with EDAW products for another ERP Application. FAC ¶ 74.

13

The alleged aftermarket here has aspects of Eastman Kodak, but it also has aspects of an

14

impermissible contractual aftermarket depending on what ERP Application program the customer

15

has. See, e.g., Forsyth v. Humana, Inc., 114 F.3d 1467, 1476 (9th Cir.1997) (rejecting a

16

contractually based restriction on competition could not form the basis for an antitrust submarket).

17

On one hand, the market is derivative from the primary market because EDAW products for SAP

18

ERP Applications customers would exist regardless of SAP’s ERP Application products. On the

19

other hand, EDAW products that are “specifically designed” for SAP ERP Applications and that

20

are not “interchangeable” would only exist in a market for SAP ERP customers. Relatedly, SAP’s

21

customers who purchased S/4HANA after the alleged policy changes are not subjected to the same

22

information disparities that Teradata alleges with other SAP ERP Applications customers because

23

they would have purchased the product knowing that HANA was the only compatible choice for

24

EDAW services.

25

The three remaining “relevant” allegations identified in Newcal are present in Teradata’s

26

FAC as well. Newcal, 513 F.3d at 1050. Teradata contends that there is a restraint on trade in the

27

tied market for SAP customer’s EDAW products but does not allege a restraint in the initial

28

market for ERP Applications, which is competitive between Oracle and SAP. FAC ¶ 67 (alleging
27

1

that SAP holds a dominant position in the ERP Applications market of about 60% to 90%, with

2

Oracle as its only significant competitor.). Teradata claims that SAP’s market power in the tied

3

market for SAP customer EDAW products flows from its relationship with its customers in the

4

ERP Application market because there are extremely high implementation costs for ERP

5

Applications, and high switching costs needed to change ERP Applications providers. FAC ¶¶

6

62–66. Finally, Teradata alleges that SAP customers were subjected to a change in practice that is

7

not prevalent in the market for ERP Applications and could not be known at the time they decided

8

to purchase SAP’s product. FAC ¶ 90; see Newcal, 513 F.3d at 1050 (“The fourth relevant aspect

9

of the complaint is that it alleges that market imperfections…prevent consumers from realizing
that their choice in the initial market will impact their freedom to shop in the aftermarket.”).

11
United States District Court
Northern District of California

10

The relevant tied market allegations are similar to Eastman Kodak , though SAP has

12

identified some differences that will be explored in the litigation. Teradata’s allegations – taken as

13

true and drawing reasonable inferences in its favor at this stage – are enough to survive a motion

14

to dismiss.

15

b.

Allegations of Anticompetitive Harm

16

Next, SAP contends the third element of the rule of reason violation claim is not met since

17

there is no unreasonable anticompetitive harm alleged. SAP claims the S/4HANA product was an

18

innovation, as discussed previously, and it relies on Brantley v. NBC Universal, Inc., 675 F.3d

19

1192 (9th Cir. 2012), to argue that there was no actual anticompetitive effect from the S/4HANA

20

upgrade requiring SAP’s HANA product to be used for EDAW.

21

As an initial matter, the argument that SAP was simply innovating a product and is

22

therefore immune from unlawful tying antitrust claims before discovery is not convincing. The

23

FAC does not claim SAP/4HANA was innovative or integrated the EDAW and ERP Applications

24

products. To the contrary, the allegations concerning integration only refer to HANA as it

25

integrated with transactional databases. See, e.g., FAC ¶ 42 (stating that like Teradata Foundation,

26

“SAP’s HANA product combines a database solution with integrated software to perform data

27

analytics.”).

28

SAP is right that in Brantley, the Ninth Circuit found agreements that effectively reduce
28

1

customer choices or increase prices were not sufficient to allege an anticompetitive harm because

2

“[b]oth effects are fully consistent with a free, competitive market.” Brantley, 675 F.3d at 1202.

3

But in that case, the court found that the complaint did not allege injuries to competition that are

4

typically sufficient; such as selling product packages that “excludes other sellers… from the

5

market,” placing “barriers to entry” in the market, or causing customers to “forego the purchase of

6

substitutes for the tied product.” Id. at 1201.

7

Here, Teradata has alleged that SAP’s market behavior excludes Teradata from the market

8

as a direct result of SAP locking in its customer base. FAC ¶ 149. It asserts that SAP’s conduct

9

prevents Teradata’s EDAW product offerings for Top-Tier ERP Applications customers from
entering the market. FAC ¶ 75. It also alleges SAP’s customers have little reasonable choice but

11
United States District Court
Northern District of California

10

to adopt HANA for EDAW rather than Teradata’s substitute given the prohibitive costs of

12

switching services and the lack of information that customers have at the time of purchase. FAC ¶

13

92. This is more than enough to plead an anticompetitive injury.

14

B.

15

An attempted monopolization claim under Section 2 of the Sherman Act must first

Attempted Monopolization

16

“specify the market [SAP] targeted and [SAP’s] economic power within that market.” United

17

Energy Trading, LLC v. Pac. Gas & Elec. Co., 200 F. Supp. 3d 1012, 1020 (N.D. Cal. 2016).

18

Teradata must plead that SAP: (i) had “specific intent to control prices or destroy competition;”

19

(ii) engaged in “anticompetitive conduct directed at accomplishing that purpose;” (iii) has a

20

“dangerous probability of achieving monopoly power;” and (iv) a “causal antitrust injury.” Id.

21

(citing Rebel Oil Co. v. Atl. Richfield Co., 51 F.3d 1421, 1432–33 (9th Cir. 1995)). SAP moves to

22

dismiss the attempted monopolization claim on the ground that Teradata fails to plead a dangerous

23

probability of monopolization. See Mot. to Dismiss at 23.

24

Under the Sherman Act, “monopoly power” is the power to control prices or exclude

25

competition. United States v. Grinnell Corp., 384 U.S. 563, 571 (1966). Monopoly power can be

26

inferred “from the predominate share of the market.” Id. Similarly, a dangerous probability of

27

monopolization is based on “the relevant market and the defendant’s ability to lessen or destroy

28

competition in that market.” Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 456 (1993).
29

1

SAP argues again that there are no allegations about the competitive conditions in the

2

general EDAW market for Top-Tier ERP Applications customers other than to state there are

3

multiple competitors. FAC ¶ 75 (“EDAW products providers, such as Teradata,…”). It also

4

asserts that because the relevant market is only “SAP’s Top-Tier Applications customers,” ¶¶ 143,

5

147, the claim must be dismissed. See Todd v. Exxon Corp., 275 F.3d 191, 200 (2d Cir. 2001)

6

(finding cases were dismissal is frequently appropriate involve attempts “to limit a product market

7

to a single brand, franchise, institution, or comparable entity that competes with potential

8

substitutes.”). But as I explained earlier in this Order, Teradata has sufficiently alleged a relevant

9

derivative aftermarket of SAP’s Top-Tier ERP Applications customers.

10

The question remaining is whether Teradata has alleged that SAP has market power in the

United States District Court
Northern District of California

11

relevant EDAW market for Top-Tier Applications customers. It has. Teradata alleged that it has a

12

market-leading EDAW product generally and that SAP was a dominant player in the Top-Tier

13

ERP Applications market—with 60% to 90% market share on information and belief. FAC ¶¶ 31,

14

67. Teradata also alleges that 60% of SAP’s existing ERP Applications customers “are employing

15

or preparing to employ HANA” based on SAP’s recent anticompetitive conduct. FAC ¶ 148. The

16

FAC states there are prohibitively high switching costs combined with restrictions on customers’

17

ability to export their own ERP Applications data to use with other EDAW products. FAC ¶¶ 75,

18

94. Customers in the Top-Tier ERP Applications market also lack the information at the point of

19

purchase to perform detailed cost analyses, contributing to SAP’s ability to lock-in those

20

customers. FAC ¶¶ 60–62.

21

Considering these allegations regarding SAP’s market share and conduct that precludes

22

competitors from servicing its Top-Tier Applications customers in the EDAW market, I find

23

Teradata sufficiently states a claim. Accordingly, SAP’s motion to dismiss the attempted

24

monopolization claim is DENIED.

25
26

CONCLUSION
In accordance with the foregoing, SAP’s motion to dismiss is GRANTED with regards to

27

the trade secret misappropriation claims but DENIED with regards to all other claims. Teradata

28

does not sufficiently allege its trade secrets. Therefore, Teradata’s first amended complaint is
30

1
2
3

DISMISSED WITH LEAVE TO AMEND within ten days of the date of this Order.
IT IS SO ORDERED.
Dated: December 12, 2018

4
5
William H. Orrick
United States District Judge

6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
31