Schellenbach et al v. Godaddy Incorporated

Filing 141

ORDER denying 134 Motion for Reconsideration. Signed by Judge David G Campbell on 8/29/2017.(DGC, nvo)

1

WO

2
3
4
5
6

IN THE UNITED STATES DISTRICT COURT

7

FOR THE DISTRICT OF ARIZONA

8
9

Mark Schellenbach, et al.,

No. CV16-0746 PHX DGC

Plaintiffs,

10

ORDER

11

v.

12

Godaddy Incorporated, a Delaware
corporation.

13

Defendants.

14
15
16

Plaintiffs ask the Court to reconsider its order denying their motion for class

17

certification. Doc. 134. Defendants oppose the motion. Doc. 140. For the reasons that

18

follow, the Court will deny the motion for reconsideration.

19

I.

Legal Standard.

20

Courts in this district have identified four circumstances where a motion for

21

reconsideration will be granted: (1) the moving party has discovered material differences

22

in fact or law from those presented to the Court at the time of its initial decision, and the

23

party could not previously have known of the factual or legal differences through the

24

exercise of reasonable diligence; (2) material factual events have occurred since the

25

Court’s initial decision; (3) there has been a material change in the law since the Court’s

26

initial decision; or (4) the moving party makes a convincing showing that the Court failed

27

to consider material facts that were presented to the Court at the time of its initial

28

decision. See, e.g., Motorola, Inc. v. J.B. Rodgers Mech. Contractors, Inc., 215 F.R.D.

1

581, 586 (D. Ariz. 2003). Motions for reconsideration are disfavored, and are not the

2

place for parties to make new arguments not raised in their original briefs. See Northwest

3

Acceptance Corp. v. Lynnwood Equip., Inc., 841 F.2d 918, 925-26 (9th Cir. 1988).

4

II.

Discussion.

5

The Court denied Plaintiffs’ motion to certify a proposed class and a California

6

subclass under Rule 23(b)(2) and (3) of the Federal Rules of Civil Procedure. The

7

primary focus of the parties’ briefing and argument was Rule 23(b)(3). The Court found

8

that individual issues would predominate over common issues in the class and subclass,

9

making certification inappropriate under Rule 23(b)(3). Doc. 130 at 2-21. The Court

10

denied certification under Rule 23(b)(2) because the class was not suitable for injunctive

11

relief. Id. at 21-22.

12

A.

Plaintiffs’ First Objection.

13

Plaintiffs’ first argument focuses on the following sentence from section III.B of

14

the Court’s order:

“Given this target market, it is likely that the class includes

15

sophisticated computer users, and an individualized inquiry would be required to

16

determine whether class members had the sophistication to understand that VM meant

17

virtualized machine even if they did not visit the /servers webpage.” Doc. 130 at 7.

18

Plaintiffs argue that this language “sets forth the wrong standard for determining if an

19

advertisement was likely to deceive” under the Arizona Consumer Fraud Act (“ACFA”).

20

Doc. 134 at 4. Plaintiffs argue that Arizona law requires a much lower standard, one

21

looking to “whether the ‘least sophisticated consumer’ would be misled [by the

22

advertising in question] – not the most sophisticated or target consumer.” Id. at 5 (citing

23

Madsen v. W. Am. Mort. Co., 694 P.2d 1228, 1232 (Ariz. Ct. App. 1985)).

24

The Court first notes that Plaintiffs did not make this argument in their motion for

25

class certification or their reply memorandum. Docs. 109, 123. They never cited the

26

Madsen case and never argued that the “least sophisticated consumer” standard applied to

27

their ACFA claim. Id. As noted above, a motion for reconsideration is not the place for

28

-2-

1

parties to make arguments not raised in their original briefs. This is a sufficient basis for

2

denying Plaintiffs’ first ground for relief.

3

But the Court also concludes that it would have denied this ground for relief even

4

if it was timely raised. It is true that some Arizona cases have said that the ACFA

5

employs a “least sophisticated consumer” test for determining whether a communication

6

has a tendency to deceive. Madsen, the primary case relied on by Plaintiffs, cited a

7

Second Circuit case as support for this standard. 694 P.2d at 1232 (citing Exposition

8

Press, Inc. v. FTC, 295 F.2d 869 (2d Cir. 1961)). Exposition Press, in turn, was an

9

enforcement action brought by the FTC alleging deceptive advertising. In such an action,

10

the FTC must show that an advertisement has a “tendency to deceive” the public. 295

11

F.2d at 872. And in making such a showing, the FTC “should look not to the most

12

sophisticated readers but rather to the least.” Id.

13

The ACFA has the same public protection purpose.

The statute expressly

14

authorizes the Arizona Attorney General to bring enforcement actions against deceptive

15

advertisers. A.R.S. § 44-1524. The ACFA has been used in this manner. See, e.g., State

16

ex rel. Horne v. Autozone, Inc., 258 P.3d 289 (Ariz. Ct. App. 2011), opinion vacated in

17

part, 275 P.3d 1278 (Ariz. 2012).

18

When an agency such as the FTC or the Arizona Attorney General brings an

19

enforcement action alleging deceptive advertising, it makes no sense to require that the

20

agency prove it was misled by the advertising. The action is brought on behalf of

21

consumers generally, not the agency itself. As a result, the agency must show only that

22

the advertising had a “tendency” to mislead the public. Exposition Press, 295 F.2d at 872

23

(FTC must show that the advertising had “the tendency and capacity to deceive a

24

substantial portion of the purchasing public”). Indeed, enforcement agencies need not

25

present evidence from consumers. Id. (“Actual consumer testimony is in fact not needed

26

to support an inference of deceptiveness by the [FTC].”).

27
28

The language relied on by Plaintiffs’ motion for reconsideration concerns the
tendency of an advertisement to mislead. This is the key language from Madsen:

-3-

1
2
3
4
5
6
7
8
9
10

The term “deceptive” has been interpreted to include representations that
have a “tendency and capacity” to convey misleading impressions to
consumers even though interpretations that would not be misleading also
are possible. Chrysler Corp. v. FTC, 561 F.2d 357, 363 (D.C. Cir. 1977);
Goodman v. FTC, 244 F.2d 584 (9th Cir.1957), reaffirmed in Simeon
Management Corp. v. FTC, 579 F.2d 1137 (9th Cir.1978). The meaning
and impression are to be taken from all that is reasonably implied, not just
from what is said, Spiegel, Inc. v. FTC, 411 F.2d 481 (7th Cir.1969), and in
evaluating the representations, the test is whether the least sophisticated
reader would be misled. Exposition Press, Inc. v. FTC, 295 F.2d 869 (2d
Cir.1961), cert. denied 370 U.S. 917, 82 S.Ct. 1554, 8 L.Ed.2d 497 (1962).
694 P.2d at 1232. Significantly, every case cited in this language from Madsen involves
the FTC.

11

As Plaintiffs correctly note, the “least sophisticated consumer” language has been

12

repeated in Arizona cases addressing private rights of action implied under the ACFA,

13

rather than government enforcement actions.

14

Cheatam v. ADT Corp., 161 F. Supp. 3d 815, 828 (D. Ariz. 2016) (quoting Madsen). But

15

these private-plaintiff cases have not explained how the test is to be applied in a private

16

action brought by an individual claiming to have been misled by the defendants’

17

misrepresentations or omissions.

Madsen was such a case.

See also

18

One thing is clear, however. Arizona cases citing this language do not mean, as

19

Plaintiffs seem to suggest, that a person who was not misled by a misrepresentation or

20

omission because she was sophisticated enough to know the truth can nonetheless assert a

21

claim for misrepresentation or omission under the ACFA because a less sophisticated

22

consumer would have been misled. This is not the law in Arizona. As the Arizona Court

23

of Appeals has explained in applying the ACFA:

24
25
26
27
28

If appellants actually knew that statements in the ad regarding gross sales
and net profit were false, then they could not have relied on the truth of the
representations and would not have been damaged by them. The court
would then have been correct in entering judgment for appellee on the
counterclaim since under either theory of fraud, common law or [the
ACFA], a private claimant must show reliance.

-4-

1

Peery v. Hansen, 585 P.2d 574, 577-78 (Ariz. Ct. App. 1978); see also Shupe v. Cricket

2

Comm’s, Inc., No. CV 13-1052, 2014 WL 6983245 at *7 (D. Ariz. Dec. 10, 2014)

3

(applying ACFA and explaining: “Here, as in Peery, if Mr. Shupe knew that the ‘no

4

contracts’ advertisements were false, it is impossible that Mr. Shupe relied on the

5

veracity of those statements and thus Mr. Shupe could not have been damaged by that

6

misrepresentation.”); Gerber v. Wells Fargo Bank, N.A., No. 11-01083, 2011 WL

7

5007921 at *5 (D. Ariz. Oct. 20, 2011) (applying ACFA and explaining: “Accordingly, if

8

Gerber actually knew about Wells Fargo’s senior lien, he cannot establish reliance and

9

his cause of action fails.”). Thus, if a plaintiff was not misled by a misrepresentation or

10

omission because she knew the truth, she cannot bring a claim under the ACFA.

11

That principle applies here. If class members knew that the servers they purchased

12

were virtualized because they were sophisticated enough to understand that the acronym

13

“VM” meant “virtualized machine,” they were not injured by GoDaddy’s alleged failure

14

to disclose the virtualization and they have no claim under the ACFA. As a result, in this

15

case, individualized inquiries will be required to determine which class members saw the

16

“VM” acronym and understood what it meant.

17

This is the point the Court was making in the portion of the class certification

18

order now challenged by Plaintiffs. The order noted – and Plaintiffs do not dispute – that

19

the first element Plaintiffs must prove under the ACFA is that GoDaddy made a

20

misrepresentation or omission in its representations to Plaintiffs. Doc. 130 at 4 (citing

21

Parks v. Macro-Dynamics, Inc., 591 P.2d 1005, 1008 (Ariz. Ct. App. 1979)). Plaintiffs’

22

ACFA claim rests on one key omission – GoDaddy’s alleged failure to disclose that the

23

servers were virtualized. The discussion in section III.B of the order considers whether

24

Plaintiffs can prove this omission with class-wide evidence, or whether individual issues

25

will predominate. Indeed, section III.B is titled “Plaintiffs’ Key Omission and the Need

26

for Individual Inquiries.” Doc. 130 at 4.

27

The first portion of section III.B notes that many class members – up to 30% of

28

the class – viewed the “/servers webpage” that clearly disclosed the virtualized nature of

-5-

1

the servers. Id. at 4-6. Because these class members were not exposed to Plaintiffs’ key

2

omission and therefore could not prevail on the ACFA claim, a class-member by class-

3

member inquiry would be needed to identify them.

4

The language of the order on which Plaintiffs now focus concerns class members

5

who have no claim because they saw the acronym “VM” on the “/pro/dedicated-server

6

webpage” and were sophisticated enough to understand that it meant “virtualized

7

machine.” These class members are not an insignificant percentage of the class. As the

8

Court noted in its order, and Plaintiffs do not dispute, GoDaddy marketed these servers to

9

sophisticated computer users. Id. at 6-7. Sophisticated class members who understood

10

the meaning of “VM” were not subjected to Plaintiffs’ key omission and cannot establish

11

the first element of an ACFA claim. Individual inquiries would be needed to identify

12

them.

13

The opening sentence of the next paragraph makes clear that this remains the

14

focus of the Court’s discussion in section III.B: “And these are not the only ways class

15

members could have learned that the servers were virtualized.” Doc. 130 at 7. The order

16

goes on to note that some class members were told the servers were virtualized in phone

17

conversations with GoDaddy representatives. Id.

18

The conclusion of section III.B is stated this way: “Consequently, on the very first

19

element of Plaintiffs’ claims – the existence of a material omission – individual issues

20

would predominate if the class were certified.” Id. at 8. The Court finds no error in the

21

section III.B discussion or this conclusion.

22

B.

23

Plaintiffs next argue that the Court erred when it rejected their suggestion that

24

questionnaires could be used to eliminate individual issues. This is the paragraph from

25

the Court’s order to which Plaintiffs object:

26
27
28

Plaintiffs’ Second Objection.

Plaintiff suggested at oral argument that these individual inquiries
could be made through a simple two-question questionnaire sent to class
members, but the Court is not persuaded that individual issues could be
eliminated so easily. The Court could not require GoDaddy to accept a class

-6-

1
2
3
4
5
6
7

member’s simple assertion in a questionnaire that she did not visit the
/servers webpage or receive information regarding the virtual nature of the
server through other means. This issue lies at the heart of Plaintiffs’ liability
claim, and GoDaddy certainly would be entitled to test a class member’s
assertion on such a central fact, conducting discovery and presenting
evidence regarding the webpages visited by the class member and other
information the class member received regarding the Dedicated Server
before purchase.
Doc. 130 at 14.

8

Plaintiffs argue that this conclusion overlooks the Ninth Circuit’s approval of

9

questionnaires in Briseno v. ConAgra, 844 F.3d 1121, 1130 (9th Cir. 2017). But the

10

questionnaires in Briseno were to be used in the claims-administration process at the end

11

of a small-claim consumer class action. Id. As the court explained: “courts ‘can rely, as

12

they have for decades, on claim administrators, various auditing processes, sampling for

13

fraud detection, follow-up notices to explain the claims process, and other techniques

14

tailored by the parties and the court’ to avoid or minimize fraudulent claims.” Id.

15

(quoting Mullins v. Direct Digital, LLC, 795 F.3d 654, 667 (7th Cir. 2015)).

16

Using questionnaires or other standardized forms to pay claims at the end of a

17

class action is a far cry from determining liability at trial. As the Court noted in the

18

language quoted above, the issue of whether class members were exposed to the allegedly

19

fraudulent conduct in this case concerns far more than claims administration – it “lies at

20

the heart of Plaintiffs’ liability claim.” Id. The Court remains unpersuaded that the

21

individualized inquiries needed to resolve this central liability issue can be eliminated

22

through the use of questionnaires.

23

C.

24

Plaintiffs argue in their third objection that the Court “erroneously applied the

25

materiality standard under Mazza v. Am Honda Motor Co with regard to the presumption

26

of reliance for the California sub-class.” Doc. 134 at 9-10. The Court simply disagrees

27

with Plaintiffs’ reading of Mazza. That case found that “the misrepresentations at issue

28

. . . do not justify a presumption of reliance. This is so primarily because it is likely that

Plaintiffs’ Third Objection.

-7-

1

many class members were never exposed to the allegedly misleading advertisements[.]”

2

666 F.3d at 595. The same is true here – many class members were never exposed to

3

Plaintiffs’ key omission. And Plaintiffs’ motion for reconsideration makes no mention of

4

the other cases cited by the Court in support of its conclusion that a presumption of

5

reliance would not be appropriate. See Doc. 130 at 10-11 (citing Davis-Miller v. Auto.

6

Club of S. California, 201 Cal. App. 4th 106, 125 (Cal. Ct. App. 2011); Knapp v. AT & T

7

Wireless Servs., Inc., 195 Cal. App. 4th 932, 942-43 (Cal. Ct. App. 2011); Cohen v.

8

DIRECTV, Inc., 178 Cal. App. 4th 966, 973 (Cal. Ct. App. 2009); Stearns v, Ticketmaster

9

Corp, 655 F.3d 1013, 1022 (9th Cir. 2011), abrogated on other grounds by Comcast

10

Corp. v. Behrend, 133 S.Ct. 1426 (2013); Plascencia v. Lending 1st Mortg., No. C 07-

11

4485 CW, 2011 WL 5914278, at *2 (N.D. Cal. Nov. 28, 2011)).

12

D.

Plaintiffs’ Fourth Objection.

13

Plaintiffs’ fourth objection argues that the Court’s “findings that 30% of the

14

putative class likely viewed the /servers webpage or could have learned of the virtual

15

nature through word of mouth or trade publications are unsupported by evidence.”

16

Doc. 134 at 10-11. The Court relied on evidence from the declaration of Paul Bindel in

17

conducting this analysis. See Doc. 130 at 6; Doc. 116-2, ¶ 11.

18

Plaintiffs argue that there is no basis for concluding that the 30% of visitors who

19

saw the /servers webpages actually purchased the servers at issue in this case, and note

20

that perhaps only 10% made such purchases. The Court noted in its order that the 30%

21

figure for visits to the /servers webpage was conservative because there was likely

22

overlap between the nearly 400,000 people who visited the /servers webpage and the

23

almost 900,000 who visited the /pro/dedicated-server webpage. Doc. 130 at 6. Indeed, if

24

everyone who visited the /servers webpage also visited the /pro/dedicated-serve webpage,

25

then 44% of the total would have seen the /servers webpage where the virtualized nature

26

of the servers was disclosed. The Court chose to use the more conservative 30%. Id.

27

This percentage may be high when applied to the class, because some people who

28

saw the /servers webpage would have purchased through that webpage and would be

-8-

1

excluded from the class definition. But Plaintiffs have the burden of proof on class

2

certification, and they provided no information regarding the number of persons who

3

purchased through the /servers webpage or through other means. Thus, the Court has no

4

basis for discounting the 30% figure. But even if Plaintiffs are correct and the percentage

5

of class members who saw the /servers webpage and purchased the servers at issue in this

6

case is closer to 10% (see Doc. 134 at 11), this still amounts to approximately 1,000 of

7

the 10,000 potential class members, and it does not include those who learned of the

8

virtualized nature of the servers through phone conversations with GoDaddy

9

representatives or through the acronym “VM” that appeared on the /pro/dedicated-server

10

webpage. This case still would present the prospect of many class members who knew

11

the servers were virtualized. Individualized inquiries would be required.

12

Plaintiffs also argue that no evidence has been produced showing that class

13

members learned the virtualized nature of the servers through trade publications or word

14

of mouth. Doc. 134 at 11. True, but the Court stated only that its analysis “does not

15

account for other means by which the class could have learned that the servers were

16

virtualized, such as word of mouth or trade publications.” Doc. 130 at 8 (emphasis

17

added). The Court did not attempt to quantify those class members, but cited them as one

18

more illustration of the most basic problem with this class – that each class member had a

19

unique collection of information at the point of purchase, a reality that would necessitate

20

individualized inquiries before liability could be established.

21

E.

Plaintiffs’ Fifth Objection.

22

Plaintiffs’ final objection is that the Court failed to narrowly define the class in

23

order to make it acceptable for class certification. Doc. 134 at 12-13. Specifically,

24

Plaintiffs argue that “denial of class certification without attempting to narrow the class

25

to exclude such purchasers was premature. The Ninth Circuit has explained that when

26

overbreadth of a class definition is the issue, ‘the better course is not to deny class

27

certification entirely but to amend the class definition as needed to correct the

28

-9-

1

overbreadth.’” Id. at 12 (citing Torres v. Mercer Canyons, Inc., 835 F.3d 1125, 1139 (9th

2

Cir. 2016)).

3

But the Court does not agree that it bore the obligation of properly defining the

4

proposed class. That was Plaintiffs’ burden. Plaintiffs proposed the class and subclass

5

after more than a year of litigation and discovery. The Court addressed the class as

6

Plaintiffs presented it.

7

Plaintiffs contend that their “class definition may be cured by simply excluding

8

from the Class, ‘all purchasers who have viewed the ‘/server’ webpage prior to making

9

their purchase,” and that, “[i]n doing so, the Plaintiffs’ class definition will be precisely

10

limited to only purchasers who were not exposed to the purported discloser on the /server

11

webpage – thus, resolving the Court’s concerns regarding the purported inclusion of

12

‘many’ class members who were not subject to the key omission.” Id. at 13.

13

But even this proposed definition would not account for class members who

14

learned the servers were virtualized through phone conversations with GoDaddy

15

representatives or who saw the servers described as “VM” machines and had the

16

experience to know that this meant virtualized machine. And it would not address the

17

other problems the Court found with class certification. See Doc. 130 at 9-13.

18

What is more, Plaintiffs’ proposed new definition might well be an improper fail-

19

safe class. See Torres, 835 F.3d at 1138 n.7 (“Indeed, defining the class to include only

20

those individuals who were ‘injured’ by non-disclosure threatens to create a ‘fail safe’

21

class, one that is defined so narrowly as to ‘preclude[] membership unless the liability of

22

the defendant is established.’” (citing Kamar v. RadioShack Corp., 375 F. App’x. 734,

23

736 (9th Cir. 2010))).

24

IT IS ORDERED: Plaintiffs’ motion for reconsideration (Doc. 134) is denied.

25

Dated this 29th day of August, 2017.

26
27
28

- 10 -