Marketquest Group, Inc. v. Societe Bic, S.A. et al

Filing 403

ORDER denying Plaintiff's 199 Daubert Motion to Exclude the Survey and Testimony of Hal Poret. Signed by Judge Cynthia Bashant on 4/12/2018. (jah)

1
2
3
4
5
6
7
8
9
10

UNITED STATES DISTRICT COURT

11

SOUTHERN DISTRICT OF CALIFORNIA

12
13

MARKETQUEST GROUP, INC.,
Plaintiff,

14
15
16
17

Case No. 11-cv-618-BAS (JLB)

v.

ORDER DENYING
PLAINTIFF’S MOTION TO
EXCLUDE SURVEY AND
TESTIMONY OF HAL PORET
[ECF No. 199]

BIC CORPORATION, et al.,
Defendants.

18
19
20

AND RELATED COUNTERCLAIMS.

21
22

On March 28, 2001, Plaintiff Marketquest Group, Inc. (“Marketquest”)

23

filed this action for trademark infringement and unfair competition against

24

Defendants BIC Corp., BIC USA, and Norwood Promotional Products (“BIC”).

25

(ECF No. 1.) On May 5, 2011, Marketquest filed the operative First Amended

26

Complaint (“FAC”). (FAC, ECF No. 14.) On May 13, 2011, BIC filed its Answer

27

and Counterclaims. (ECF No. 17.)

28

Presently before the Court is Marketquest’s Motion to Exclude the Survey and
–1–

11cv618

1

Testimony of Hal Poret. (Marketquest’s Mot., ECF No. 199.) Marketquest argues

2

that Mr. Poret’s survey is irrelevant and unreliable because it surveyed the wrong

3

universe and failed to adequately replicate market conditions. BIC opposes. (Opp’n,

4

ECF No. 257.)

5

The Court finds this motion suitable for determination on the papers submitted

6

and without oral argument. See Fed. R. Civ. P. 78(b); Civ. L.R. 7.1(d)(1). For the

7

following reasons, the Court DENIES Marketquest’s motion.

8

I.

BACKGROUND

9

Plaintiff Marketquest is a California “supplier” that produces, advertises, and

10

sells customizable promotional products using the registered trademarks “ALL-IN-

11

ONE” and “The Write Choice.” (FAC ¶¶ 10-12.) Defendant BIC represents one of

12

the largest suppliers in the promotional products industry. (Marketquest’s Mot.

13

Partial Summ. J. 1:12-2:7, ECF No. 205.) The promotional products industry

14

supports the manufacture and distribution of promotional products to companies

15

(“end consumers”) who want to brand themselves with customers. (Marketquest’s

16

Mot. 1:16-18.) When a company wants to brand itself with its customers it

17

approaches a promotional products “distributor.” (Id. 1:19-21.) The distributor then

18

approaches a supplier (such as Marketquest or BIC) who either manufactures or

19

imports a promotional product on which the end consumer’s brand is imprinted. (Id.

20

1:21-23.) According to Marketquest, such segregation of suppliers, distributors, and

21

end consumers is an integral part of the industry and never bypassed. (Id. 2:1-3.)

22

Marketquest alleges BIC began advertising and selling products using marks

23

similar to Marketquest’s. (FAC ¶¶ 21-25.) Specifically, Marketquest claims BIC used

24

the phrase “The Write Pen Choice” in an online advertising campaign for writing

25

instruments beginning in October, 2010. (Id. ¶ 23.) Around the same time, Norwood

26

Promotional Products, LLC, a subsidiary of BIC USA, printed a 2011 catalog entitled

27

the “NORWOOD All in ONE” catalog. (Id. ¶ 24.) Alleging a form of reverse

28

confusion, Marketquest argues that BIC’s use of its marks to compete with it in
–2–

11cv618

1

selling the same products—to the same customers in the same advertising channels—

2

in combination with BIC’s reputation as a brand aggregator, caused confusion

3

amongst customers. (Marketquest’s Mot. Partial Summ. J. 4:14-18.)

4

BIC retained Mr. Poret as an expert witness to conduct a survey to determine

5

the extent to which use of the phrase “All in ONE” on the cover of the 2011 Norwood

6

Catalog is likely to have caused confusion with respect to Marketquest’s ALL-IN-

7

ONE mark. (Poret Report 1, ECF No. 256-3, Ex. 1.) Mr. Poret conducted a Sequential

8

Lineup Survey wherein end consumers were called on the phone and asked to look

9

at materials on their computer screen while they were being interviewed. (Id. 3.)

10

Respondents were first shown the Marketquest website, then parts of four catalogs

11

offering customizable promotional products, one of which was the 2011 Norwood

12

Catalog. (Id.) They were then asked if they connected any of the catalogs to the

13

Marketquest website and, if so, why. (Id. 4.) The survey was conducted under the

14

assumption that distributors often show catalogs to potential end consumers. (Id. 22.)

15

The results of the survey indicate a rate of 4.7% potential trademark confusion, a rate

16

typically considered de minimis. (Id. 21.) Based on the survey results, Mr. Poret is

17

of the opinion that there is no likelihood of confusion caused by the use of the words

18

“All in ONE” on the 2011 Norwood Catalog. (Id.)

19

II.

LEGAL STANDARD

20

Federal Rule of Evidence 702 establishes several requirements for

21

admissibility of expert opinion evidence: (1) the witness must be sufficiently

22

qualified as an expert by knowledge, skill, experience, training, or education; (2) the

23

scientific, technical, or other specialized knowledge must “assist the trier of fact”

24

either “to understand the evidence” or “to determine a fact in issue”; (3) the testimony

25

must be “based on sufficient facts and data”; (4) the testimony must be “the product

26

of reliable principles and methods”; and (5) the expert must reliably apply the

27

principles and methods to the facts of the case. Fed. R. Evid. 702.

28

Under Daubert and its progeny, the trial court is tasked with assuring that
–3–

11cv618

1

expert testimony “both rests on a reliable foundation and is relevant to the task at

2

hand.” Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 597 (1993). “Expert

3

opinion testimony is relevant if the knowledge underlying it has a valid connection

4

to the pertinent inquiry. And it is reliable if the knowledge underlying it has a reliable

5

basis in the knowledge and experience of the relevant discipline.” Primiano v. Cook,

6

598 F.3d 558, 565 (9th Cir. 2010) (citation and quotation marks omitted). Shaky but

7

admissible evidence is to be attacked by cross-examination, contrary evidence, and

8

careful instruction on the burden of proof, not exclusion. Daubert, 509 U.S. at 596.

9

The judge is “to screen the jury from unreliable nonsense opinions, but not exclude

10

opinions merely because they are impeachable.” Alaska Rent–A–Car, Inc. v. Avis

11

Budget Grp., Inc., 738 F.3d 960, 969 (9th Cir. 2013). In its role as gatekeeper, the

12

trial court “is not tasked with deciding whether the expert is right or wrong, just

13

whether his [or her] testimony has substance such that it would be helpful to a jury.”

14

Id. at 969-70.

15

The tests for admissibility in general, and reliability in particular, are flexible.

16

Primiano, 598 F.3d at 564. The Supreme Court has provided several factors to

17

determine reliability: (1) whether a theory or technique is testable; (2) whether it has

18

been published in peer reviewed literature; (3) the error rate of the theory or

19

technique; and (4) whether it has been generally accepted in the relevant scientific

20

community. Mukhtar v. Cal. State Univ., 299 F.3d 1053, 1064 (9th Cir. 2002)

21

(summarizing Daubert, 509 U.S. at 592-94), overruled on other grounds by Estate

22

of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 460 (9th Cir. 2014). These factors

23

are meant to be “helpful, not definitive.” Kumho Tire Co. v. Carmichael, 526 U.S.

24

137, 151 (1999). The court “has discretion to decide how to test an expert’s reliability

25

as well as whether the testimony is reliable, based on the particular circumstances of

26

the particular case.” Primiano, 598 F.3d at 564 (citations and quotation marks

27

omitted). “[T]he test under Daubert is not the correctness of [experts’] conclusions

28

but the soundness of [their] methodology.” Daubert v. Merrell Dow
–4–

11cv618

1

Pharmaceuticals, Inc., 43 F.3d 1311, 1318 (9th Cir. 1995). Once the threshold

2

established by Rule 702 is met, the expert may testify and the fact finder decides how

3

much weight to give that testimony. Primiano, 598 F.3d at 565.

4

After admissibility is established to the court’s satisfaction, attacks aimed at

5

the weight of the evidence are the province of the fact finder, not the judge. Pyramid

6

Techs., Inc. v. Hartford Cas. Ins. Co., 752 F.3d 807, 814 (9th Cir. 2014). The court

7

should not make credibility determinations that are reserved for the jury. Id.

8

III.

DISCUSSION

9

Marketquest argues Mr. Poret’s survey is inadmissible because (i) it surveyed

10

the wrong universe of respondents and (ii) does not adequately replicate market

11

conditions. (Marketquest’s Mot. 3:7-14.)

12

The Ninth Circuit has held that evidence from a professionally conducted

13

survey should generally be found sufficiently reliable and admissible under the

14

gatekeeping test of Daubert. Southland Farms v. Stover Seed Co., 108 F.3d 1134,

15

1143 n.8 (9th Cir. 1997) (explaining that survey evidence is ordinarily admissible

16

because “[u]nlike novel scientific theories, a jury should be able to determine whether

17

asserted technical deficiencies undermine a survey’s probative value”); E. & J. Gallo

18

Winery v. Gallo Cattle Co., 967 F.2d 1280, 1292 (9th Cir. 1992) (“[I]t is routine to

19

admit a relevant survey; any technical unreliability goes to weight, not

20

admissibility.”). A survey can be admitted into evidence once it has passed the

21

threshold conditions of having a proper foundation, being relevant and having been

22

conducted according to accepted principles. Clicks Billiards, Inc v. Sixshooters Inc.,

23

251 F.3d 1252, 1263 (9th Cir. 2001). “Once the survey is admitted . . . follow-on

24

issues of methodology, survey design, reliability, the experience and reputation of

25

the expert, critique of conclusions, and the like go to the weight of the survey rather

26

than its admissibility.” Id. However, serious flaws may make reliance on a survey

27

unreasonable such that a court ought to exercise its gatekeeping role and exclude the

28

survey. See M2 Software, Inc. v. Madacy Entm’t, 421 F.3d 1073, 1087 (9th Cir. 2005)
–5–

11cv618

1

(finding no abuse of discretion in excluding survey evidence where “it was not

2

created or conducted in a manner that complied with appropriate standards”).

3

Marketquest contends that is the case here.

4

A.

Universe of Respondents

5

Marketquest argues that Mr. Poret selected the wrong universe of respondents.

6

In particular, Marketquest claims Mr. Poret’s survey cannot assess the likelihood of

7

confusion in this case because it polled the reactions of end consumers of promotional

8

products rather than Marketquest’s actual customers, the distributors. (Marketquest’s

9

Mot. 3:7-14.) Although the decision to limit the survey’s population to end

10

consumers reduces its probative value, the Court is unwilling to exclude it on that

11

basis alone.

12

“The selection of an inappropriate universe generally affects the weight of the

13

resulting survey data, not its admissibility.” 6 J. Thomas McCarthy, McCarthy on

14

Trademarks and Unfair Competition § 32:162 (5th ed. 2018). Even if a survey does

15

not select the optimal universe, the results are often still probative of the fact it was

16

intended to prove. See Vision Sports, Inc. v. Melville Corp., 888 F.2d 609, 615 (9th

17

Cir. 1989) (holding that an underinclusive survey’s results were so strong that it still

18

supported a finding of secondary meaning). Indeed, courts within the Ninth Circuit

19

are largely unwilling to exclude survey evidence on the basis of an overinclusive or

20

underinclusive target population. See Icon Enters. Int’l v. Am. Prods. Co., No. CV

21

04-1240, 2004 WL 5644805, at *25-26 (C.D. Cal. Oct. 7, 2004) (summarizing cases

22

addressing improper survey universe). If the responses of the surveyed universe are

23

irrelevant to the opinions of the universe at issue, however, the court has the authority

24

to declare the survey inadmissible. McCarthy, supra, § 32:162; see also Shari

25

Seidman Diamond, Reference Guide on Survey Research, Reference Manual on

26

Scientific Evidence 359, 377 (Federal Judicial Center ed., 3d ed. 2011) (“A survey

27

that provides information about a wholly irrelevant universe of respondents is itself

28

irrelevant.”).
–6–

11cv618

1

In a reverse confusion case such as this, the proper universe to survey is the

2

senior user’s customer base. McCarthy, supra, § 32:159. Professor McCarthy

3

provides a useful juxtaposition of forward and reverse confusion:

4
5
6
7
8
9
10
11

The traditional pattern of classic “forward confusion” occurs when
customers mistakenly think that the junior user’s goods or services are
from the same source as or are connected with the senior user’s goods or
services. Customers want to buy the senior user’s product and because of
the similarity of marks, mistakenly buy the junior user’s product instead.
In “reverse confusion,” customers purchase the senior user’s goods under
the mistaken impression that they are getting the goods of the junior user.
That is, reverse confusion occurs when the junior user’s advertising and
promotion so swamps the senior user’s reputation in the market that
customers are likely to be confused into thinking that the senior user’s
goods are those of the junior user: the reverse of traditional confusion.

12
13

Id. § 23:10.

14

The question is not whether BIC had an intent to trade on Marketquest’s

15

goodwill and reputation, but rather whether a reasonable consumer might erroneously

16

believe that Marketquest’s goods are BIC’s. See Dreamwerks Prod. Grp., Inc. v. SKG

17

Studio, 142 F.3d 1127, 1130 (9th Cir. 1998) (holding in a reverse confusion case that

18

the proper question “is whether a reasonable consumer attending a Dreamwerks-

19

sponsored convention might do so believing that it is a convention sponsored by

20

DreamWorks”). The injury is that customers come to assume that the senior user’s

21

products are really the junior user’s or that the former has somehow become

22

connected to the latter. See Big O Tire Dealers, Inc. v. Goodyear Tire & Rubber Co.,

23

561 F.2d 1365, 1372 (10th Cir. 1977) (finding potential likelihood of reverse

24

confusion where “Big O presented more than a dozen witnesses who testified to

25

actual confusion as to the source of Big O’s ‘Big Foot’ tires after watching a

26

Goodyear ‘Bigfoot’ commercial”). This results in the senior user losing the value of

27

the trademark, i.e. control over its goodwill and reputation, its product identity,

28

corporate identity, and ability to move into new markets. Sands, Taylor & Wood Co.
–7–

11cv618

1

v. Quaker Oats Co., 978 F.2d 947, 957 (7th Cir. 1992), reversed on other grounds by

2

Sands, Taylor & Wood v. Quaker Oats Co., 34 F.3d 1340 (7th Cir. 1994). Therefore,

3

a proper survey of likelihood of confusion in this case should assess whether those

4

familiar with BIC’s stronger mark, and who encounter Marketquest’s less well-

5

known mark, affiliate it with BIC’s mark. See Dreamwerks, 142 F.3d at 1130.

6

The stated purpose of Mr. Poret’s survey was “to determine the likelihood . . .

7

that prospective purchasers of customizable promotional products would be confused

8

by the use of the words ‘All in ONE’ on the [2011 Norwood Catalog].” (Poret Report

9

1.) “[T]he survey universe consisted of prospective end purchasers of customizable

10

promotional products – i.e., individuals who are involved in decisions about ordering

11

customizable promotional products for their business or organization.” (Id. 3.)

12

Marketquest contends that because both parties are suppliers, and suppliers never

13

bypass distributors to sell directly to end consumers, the likelihood of end consumer

14

confusion is irrelevant in this case. (Marketquest’s Mot. 7:17-9:5.) Rather,

15

considering these supply chain distinctions, Mr. Poret’s survey should have tested

16

the likelihood of distributor confusion over BIC’s use of the words “All in ONE.”

17

(Id. 8:13-15.)

18

BIC maintains that potential end consumers are relevant because they are

19

exposed to Marketquest’s mark in the context of making decisions about purchasing

20

its products. (Opp’n 11:5-10.) Further, Mr. Poret argues, “it made more sense to

21

survey end users, because if end users are not confused, it would be unlikely that the

22

more sophisticated distributors would be confused.” (Poret Report 22.) McCarthy

23

agrees, stating that “[s]ince dealers are usually more difficult to confuse by similar

24

marks, while evidence of confusion of dealers should tend to prove confusion of

25

consumers, evidence of non-confusion among dealers would not necessarily prove

26

consumer non-confusion.” McCarthy, supra, § 32:161. Marketquest believes this

27

ignores the angle of its argument, that is, that end consumers never had the chance to

28

be confused because distributors upstream in the supply chain were confused due to
–8–

11cv618

1

knowledge of BIC’s reputation as a “brand aggregator.” (Marketquest’s Mot. 13:9-

2

15.) According to Marketquest, BIC has a reputation within the industry of acquiring

3

small suppliers and featuring them as “sub-brands” on the front covers of their annual

4

catalogs. (Id. 13:17-14:2.) Such knowledge of BIC’s reputation as a “brand

5

aggregator” might make distributors more likely to be confused than end consumers.

6

Thus, Marketquest contends, a survey disproving end consumer confusion not only

7

fails to survey either parties’ actual customers, but also evades distributors’ industry

8

knowledge, thereby making an impermissibly large analytical gap between what the

9

survey proves (likelihood of end consumer confusion) and what it is proffered to

10

prove (likelihood of distributor confusion). (Id. 15:1-15.)

11

In the Court’s estimation, the failure to include distributors reduces the

12

survey’s probative value, but not so completely as to vitiate its relevance altogether.

13

Although Marketquest may not sell directly to end consumers, it directly promotes

14

its mark and products to end consumers through its websites, online catalogs, and

15

catalogs that distributors show to customers to facilitate their orders. (See Opp’n

16

11:5-14:5.) Marketquest itself admits distributors “were one of two groups of

17

potential customers who could be exposed to catalogs of customizable promotional

18

products.” (Marketquest’s Mot. 8:15-18.) A universe comprised solely of end

19

customers retains relevancy in this context because, at least to some degree,

20

Marketquest promotes itself to end consumers and end consumers are routinely

21

exposed to the marks and products when considering purchases. Ultimately, end

22

consumers decide what Marketquest products to purchase; that they do so through

23

intermediaries does not render their perception of the products and marks irrelevant.

24

Marketquest certainly has an argument that in a reverse confusion case such as

25

this, end consumers may be less likely than distributors with a previous

26

understanding of BIC as a “brand aggregator” to believe Marketquest and BIC are

27

now one and the same. As discussed earlier, a proper survey of the likelihood of

28

confusion in this case should assess whether those familiar with BIC’s stronger mark,
–9–

11cv618

1

and who encounter Marketquest’s less well-known mark, affiliate it with BIC’s mark.

2

However, such arguments affect the weight of the survey and should be presented to

3

the jury on cross-examination. See Southland Farms, 108 F.3d at 1143 n.8 (“Unlike

4

novel scientific theories, a jury should be able to determine whether asserted

5

technical deficiencies undermine a survey’s probative value.”). Indeed, a jury could

6

very well understand and conclude that end consumers are in fact not familiar with

7

BIC’s stronger mark, that distributors would have been, and that such sophisticated

8

industry knowledge makes it more likely that distributors would be confused and

9

consequently afford Mr. Poret’s survey little to no weight.

10

Because the “selection of an inappropriate universe generally affects the

11

weight of the resulting survey data, not its admissibility,” McCarthy, supra, § 32:162,

12

the Court finds that the universe selected by Mr. Poret, standing alone, is not so

13

flawed as to render the survey evidence inadmissible.

14

B.

Replication of Market Conditions

15

Marketquest makes three arguments that Mr. Poret’s survey did not adequately

16

replicate the way customers encounter and perceive either party’s marks in the actual

17

marketplace: (i) respondents were shown static screenshots of Marketquest’s

18

website, (ii) Mr. Poret does not know whether end consumers search supplier

19

websites, and (iii) showing Marketquest’s website and then BIC’s and other catalogs

20

improperly enhanced the distinctions between the stimuli. (Marketquest’s Mot. 16:2-

21

19:25.)

22

The Court notes that no survey is perfect: by necessity they are imperfect

23

reflections of how customers act in reality. Surveys are evidence, as opposed to proof,

24

in the sense that they only provide information about a controlled and artificial world

25

from which one can draw inferences about the real world. See Harvey S. Perlman,

26

The Restatement of the Law of Unfair Competition: A Work in Progress, 80

27

Trademark Rep. 461, 472 (1990). In other words, helpful inferences might be drawn

28

from a survey questioning randomly selected shoppers at a mall about pictures of two
– 10 –

11cv618

1

products, but it would go too far to claim their responses are direct proof of actual

2

consumers as they make purchasing decisions. Id. The appropriate approach is to

3

view such evidence with an understanding of the difficulty of developing and

4

implementing a survey and to use any technical defects only to lessen evidentiary

5

weight, not to reject the results out-of-hand. McCarthy, supra, § 32:178; see Clicks

6

Billiards, 251 F.3d at 1263.

7

In light of this approach, the Court finds Mr. Poret’s survey sufficiently

8

replicates market conditions for admission purposes. As explained earlier, Mr. Poret

9

conducted a Sequential Lineup Survey wherein end purchasers were called on the

10

phone and asked to look at materials on their computer screen while they were being

11

interviewed. (Poret Report 3.) Respondents were first shown static screenshots of the

12

Marketquest website, then parts of four catalogs offering customizable promotional

13

products, one of which was the 2011 Norwood Catalog. (Id.) They were then asked

14

if they connected any of the catalogs to the Marketquest website and, if so, why. (Id.

15

4.) Thus, the respondents were shown numerous uses of Marketquest’s mark and had

16

ample opportunity to connect the similar mark used on the BIC catalog to

17

Marketquest.

18

The defects pointed out by Marketquest, if defects at all, are technical and

19

merely lessen the survey’s evidentiary weight. A survey using static webpages allows

20

for a carefully controlled procedure in which all respondents receive a standardized

21

exposure to the mark at issue. The ability to freely browse the live site was not so

22

essential in this case as to override the need for a controlled test.

23

Marketquest’s second argument is similarly unavailing. BIC has provided

24

numerous examples in which Marketquest’s website appears to speak directly to end

25

consumers. (See Opp’n 11:5-14:5.) That the website informs end consumers they

26

should contact “your distributor” does not mean that end consumers are not availing

27

themselves of the website; in fact, it suggests just the opposite. (See id. 12:10-13.)

28

Marketquest’s final argument is that showing Marketquest’s website and then
– 11 –

11cv618

1

BIC’s and other’s catalogs only enhanced the differences between the stimuli. As

2

evidence, Marketquest highlights a respondent’s explanation for why he believed a

3

catalog was from Marketquest’s website:

4

Well, I’m not sure if I understand the question (I repeated it verbatim to
help him).Well, my initial thought is that since this is obviously a scanned
in cover, this wasn’t the same company because online companies are
online and catalog companies are catalog. But aside from that I would say
they are the same company.

5
6
7
8

(Marketquest’s Mot. 19:11-19.) Such an argument cuts both ways, however. While

9

it does suggest that perhaps using a website followed by catalogs may have created

10

an artificial distinction between “online companies” and “catalog companies,” it also

11

highlights McCarthy’s opinion that “[s]ometimes, the most illuminating and

12

probative parts of a survey are not the numbers and percentages generated by the

13

responses, but the verbatim accounts of the responses.” McCarthy, supra, § 32:178.

14

With the understanding that surveys are clarifying evidence, rather than proof,

15

verbatim responses such as this provide a nuanced view into end consumer thought

16

processes the jury might otherwise not have. Thus, Marketquest’s final, double-edged

17

attack neatly encapsulates why professionally conducted surveys should usually be

18

found sufficiently reliable and admissible under the gatekeeping test of Daubert:

19

generally speaking, even flawed surveys can and do clarify the issues in a case.

20

Based on the foregoing, the Court finds neither the survey’s universe of

21

respondents nor its replication of market conditions, standing alone or in conjunction,

22

sufficiently flawed to permit exclusion.

23

IV.

CONCLUSION & ORDER

24

The Court DENIES Plaintiff’s Daubert Motion to Exclude the Survey and

25

Testimony of Hal Poret. (ECF No. 199.) The survey and testimony are relevant and

26

offered with sufficient foundation by one qualified to give it.

27
28

IT IS SO ORDERED.
DATED: April 12, 2018
– 12 –

11cv618