USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 1 of 33
[DO NOT PUBLISH]
IN THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
________________________
No. 19-10771
________________________
D.C. Docket No. 6:16-cv-02001-GAP-GJK
SUPERIOR CONSULTING SERVICES, INC.,
a Florida corporation doing business as
Your Future Health doing business as YFH,
Plaintiff-Counter Defendant-Appellant,
versus
SHAKLEE CORPORATION
Defendant-Appellee,
SHAKLEE U.S., LLC,
Defendant-Counter Claimant-Appellee.
________________________
Appeal from the United States District Court
for the Middle District of Florida
________________________
(September 28, 2021)
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 2 of 33
Before BRANCH, LUCK, and ED CARNES, Circuit Judges.
LUCK, Circuit Judge:
Superior Consulting Services, Inc. sued Shaklee Corporation and Shaklee
U.S., LLC for trademark infringement. We previously affirmed the district court’s
denial of Superior’s motion for a preliminary injunction and concluded that the
district court did not clearly err in finding that Superior failed to establish a
likelihood of trademark confusion. Superior Consulting Servs., Inc. v. Shaklee
Corp., 710 F. App’x 850, 859 (11th Cir. 2017). The district court then held a bench
trial on the merits of Superior’s trademark infringement claims and found for
Shaklee. We affirm.
FACTUAL BACKGROUND AND PROCEDURAL HISTORY
Superior describes itself as “[t]he premier blood testing and customized
nutrition analysis company since 1976.” In addition to blood tests, Superior offers
a questionnaire that helps give customers nutrition and dietary recommendations.
Superior recommends, but does not sell, nutritional supplements to customers based
on their answers to the questionnaire.
Superior owns two federal trademarks for the mark “Healthprint.” The first,
registered on November 5, 2002, covers “[n]utritional supplements for general
health maintenance,” “[p]rinted instructional and teaching material in the field of
health care,” and “consulting services in the field of health care.” The second,
2
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 3 of 33
registered on March 1, 2005, covers “blood testing services [and] consultation in the
fields of food nutrition, diet and health.” Both marks are “incontestable” under 15
U.S.C. section 1065, meaning that their validity is “presumed” and “cannot be
challenged on the ground[] that [they are] merely descriptive, even if the challenger
can show that the mark[s] [were] improperly registered initially.” Dieter v. B & H
Indus. of Sw. Fla., Inc., 880 F.2d 322, 328 (11th Cir. 1989).
Shaklee (we refer to Shaklee Corporation and Shaklee U.S., LLC as Shaklee)
is a California-based manufacturer and distributor of nutritional supplements and
other goods. In June and August of 2016, Shaklee filed trademark applications to
register two “Healthprint” marks for “[p]roviding information in the field of personal
development, namely, personal improvement, and specifically excluding healthcare
information.” Shaklee uses its Healthprint marks in connection with a free online
questionnaire designed to promote Shaklee’s products to distributors and customers.
Shaklee does not label its products with its Healthprint marks and it does not offer
blood-testing services.
Superior sued Shaklee in the Middle District of Florida for, among other
things, trademark infringement and trademark dilution under the Lanham Act and
Florida law, unfair competition under Florida law, and for violating Florida’s
Deceptive and Unfair Trade Practices Act. Superior’s complaint included a jury
demand and sought actual, general, statutory, and punitive damages.
3
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 4 of 33
Superior then moved for a preliminary injunction to enjoin Shaklee from
“using the Healthprint mark” because there had been actual confusion and because
there was a likelihood of trademark confusion. The district court denied Superior’s
preliminary injunction motion. Superior appealed, challenging the district court’s
findings as to the trademark confusion factors and its ultimate finding as to the
likelihood of confusion. See Superior, 710 F. App’x at 853–60. We affirmed,
concluding that, although the district court clearly erred in making some of its
subsidiary findings, “it was not clear error for the district court to find that Superior
ha[d] not established a likelihood of confusion.” Id.
After we affirmed the denial of the preliminary injunction, Shaklee moved to
exclude, under Federal Rule of Evidence 702, a survey conducted by Superior’s
expert, Kirk Martensen, which measured the likelihood that the parties’ trademarks
would be confused. The district court granted Shaklee’s motion to exclude
Martensen’s survey because it didn’t “comply with the basic [tenets] of a Squirt[1]
survey” and the methodology he used wasn’t “reliable.” The district court found
that, “instead of presenting respondents with the two Healthprint marks that [were]
in dispute, Martensen’s survey simply included questions that inquired about the
word ‘Healthprint,’ asking, for example, . . . whether respondents believed it was
1
This type of survey is named after a case in which it was originally used. See SquirtCo
v. Seven-Up Co., 628 F.2d 1086, 1089 n.4 (8th Cir. 1980).
4
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 5 of 33
from one company, more than one company, or no company at all.” The court also
faulted Martensen for failing to use a control group. Finally, the district court found
that the survey “fail[ed] to actually describe Shaklee’s Healthprint service.”
The parties later filed cross-motions for summary judgment. While the
summary judgment motions were still pending, the parties filed a joint pretrial
statement. In the section of the pretrial statement titled “statement of elements of
money damages, and amount being sought,” Superior said it sought “disgorgement
of [Shaklee’s] profits, pursuant to 15 U.S.C. [section] 1117(a)” and “recovery of
attorney’s fees and costs,” the total amounts of which were “unknown at the time.”
It didn’t list any other forms of relief, including statutory and punitive damages.
Under the heading, “Statement of Disputed Facts to be Litigated,” however, the
parties listed “Whether punitive damages are warranted” as a disputed fact that
remained to be litigated. The pretrial statement also attached proposed jury
instructions.
The district court denied Superior’s motion for summary judgment and
granted Shaklee’s motion for summary judgment only as to Superior’s claims for
trademark dilution and tortious interference. As a result, Superior’s remaining
claims were trademark infringement under the Lanham Act and Florida law,
violation of the Florida Deceptive and Unfair Trade Practices Act, unfair
competition under Florida law, and unfair competition under the Lanham Act.
5
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 6 of 33
Shaklee moved for a bench trial. Shaklee argued that Superior’s only
remaining remedy was disgorgement because Superior waived its claims for punitive
and statutory damages by not listing them in the damages section of the joint pretrial
statement. Therefore, Shaklee argued, Superior wasn’t entitled to a jury trial because
a party seeking only disgorgement in a trademark infringement case has no right to
a jury trial.
In support, Shaklee cited one of the district court’s local rules which said that
the pretrial statement limited the scope of issues at trial:
All pleadings filed by any party prior to filing of the pretrial statement
shall be deemed to be merged therein, or in any subsequent pretrial
order entered by the Court. The pretrial statement and the pretrial order,
if any, will control the course of the trial and may not be amended
except by order of the Court in the furtherance of justice. If new
evidence or witnesses are discovered after filing of the pretrial
statement, the party desiring to use the same shall immediately notify
opposing counsel and the Court, and such use shall be permitted only
by order of the Court in the furtherance of justice.
M.D. Fla. R. 3.06(e) (2009). Shaklee argued that Superior’s focus on equitable
disgorgement damages in the pretrial statement was consistent with its other recent
and repeated representations to the district court. And, Shaklee contended, it only
joined in the pretrial statement’s pronouncement “that the case [would] be tried by
a jury” because the district court had “yet to resolve certain ‘legal’ claims with a
right to a jury trial, including Superior’s count for tortious interference.”
6
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 7 of 33
The district court granted Shaklee’s bench trial motion, concluding that, “[t]o
the extent . . . Superior had claims for statutory or punitive damages prior to the
filing of the Joint Pretrial Statement, those claims were waived when it failed to
include them in its statement of damages sought.” The court explained that local
rule 3.06(e) was “clear that the Pretrial Statement [was] controlling.”
The district court then held a five-day bench trial. After the trial, the district
court found for Shaklee on all of Superior’s claims. The parties agreed, and the
district court concluded, that Superior’s claims boiled down to whether Shaklee’s
use of the Healthprint mark was “likely to cause confusion with Superior’s use of
the mark.” The district court then evaluated each of the trademark confusion factors.
After evaluating the factors, the district court considered the overall balance to
determine the likelihood of confusion:
Although Superior has a strong mark that is substantially similar to
Shaklee’s mark, there is no evidence of actual confusion, or malintent
on Shaklee’s part, and the other factors do nothing to tip the scale in
Superior’s favor. There is, however, evidence that strongly supports a
finding that there is no likelihood of customer confusion, and that is the
testimony of Shaklee’s expert witness, Hal Poret.
Mr. Poret, a recognized expert in the field of market research,
performed an online survey to test the likelihood of confusion between
the two Healthprint marks. Using a group of 400 potential customers,
Poret used a “squirt” survey method to compute a net rate of confusion
between the two Healthprint marks. Using screenshots from the
respective websites, the survey respondents were shown both
Superior’s and Shaklee’s Healthprint marks in close proximity and
asked questions as to whether they believed the two services come from
the same company or are otherwise related. Poret concluded that there
7
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 8 of 33
is a net rate of confusion of no more than 7% between the two marks.
A net rate of confusion at this level demonstrates that there is no
likelihood of confusion between these marks. The Court finds this
evidence to be credible and highly persuasive. Having considered the
seven factors, the Court concludes that Superior has failed to prove
infringement of its Healthprint mark.
Superior now appeals the district court’s judgment for Shaklee.
DISCUSSION
Superior raises three issues on appeal. First, Superior contends that it should
have received a jury trial because the district court abused its discretion by finding
that Superior waived its claims for statutory and punitive damages (which would
have entitled it to a jury trial). Second, Superior argues that the district court clearly
erred in finding no likelihood of trademark confusion. And third, Superior asserts
that the district court abused its discretion in excluding Martensen’s survey.
Jury Trial
The district court held a bench trial because it concluded that Superior waived
its claims for statutory and punitive damages by not listing them in the damages
section of the pretrial statement. Superior argues that the district court erred because
it ignored the pretrial statement “as a whole,” including the context in which it was
filed. We review the trial court’s interpretation of the pretrial statement for abuse of
discretion. Olmsted v. Taco Bell Corp., 141 F.3d 1457, 1461 (11th Cir. 1998).
Likewise, we give “great deference to a district court’s interpretation of its local
rules.” Clark v. Hous. Auth. of City of Alma, 971 F.2d 723, 727 (11th Cir. 1992).
8
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 9 of 33
Early in the case, the district court entered a case management order
instructing the parties to prepare a joint final pretrial statement “that strictly
conform[ed] to the requirements of Local Rule 3.06(c).” Local rule 3.06(c) required
that the pretrial statement include “a statement of the elements of each [claim for
money damages] and the amount being sought with respect to each such element.”
M.D. Fla. R. 3.06(c)(7) (2009). Consistent with local rule 3.06(e), the district court’s
order also said that the pretrial statement would “control the course of the trial.” See
id. R. 3.06(e) (“The pretrial statement and the pretrial order, if any, will control the
course of the trial . . . .”).
But the parties’ pretrial statement didn’t include a claim for statutory and
punitive damages in the damages section. Although the jury instructions attached to
the pretrial statement referenced both types of damages, the local rules made clear
that the pretrial statement controlled. See id.
Also, Shaklee explicitly told the district court that it was only seeking
disgorgement. At a hearing to determine whether Superior could introduce evidence
to show Shaklee’s profits from the alleged infringement, the district court asked
Superior: “What are [Superior’s] damages, other than a disgorgement remedy?”
Superior responded that its “damages as currently in the report are disgorgement
because to do otherwise would be double dipping . . . .” The district court inquired
further: “So you’re not going to make any claim for lost profits to [Shaklee]. Rather,
9
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 10 of 33
your remedy is the disgorgement remedy?” Superior responded: “That is correct.”
It was not an abuse of discretion for the district court to conclude that Superior
waived its claim for statutory and punitive damages when Superior told the district
court it was only seeking disgorgement.
We have found no abuse of discretion under similar circumstances. In
Olmsted, for example, the plaintiff sued “under both Title VII, for which potential
damages awards ha[d] been limited by statute, and 42 U.S.C. [section] 1981(a),
which ha[d] no statutory damages cap.” 141 F.3d at 1461. The jury found for the
plaintiff and awarded him “$10,000 in back pay, $450,000 in compensatory
damages, and $3 million in punitive damages.” Id. at 1460. On the defendant’s
motion, the district court reduced the jury’s damages award because the plaintiff
“effectively . . . abandoned his [section] 1981(a) claim” in the pretrial statement and,
as a result, his recovery was limited by the $300,000 statutory damages cap under
Title VII. Id. at 1460–61.
On appeal, we affirmed the district court’s interpretation of the pretrial
statement despite the “technical” nature of the plaintiff’s mistake:
Here, although we acknowledge the potential confusion for plaintiffs in
differentiating between the amendment to Title VII that is embodied in
42 U.S.C. § 1981a and the cause of action created by 42 U.S.C.
§ 1981(a), we cannot conclude in this instance that the district court’s
construction of the pretrial order—that is, that Olmsted abandoned his
§ 1981(a) claim by virtue of the pretrial stipulation—was unreasonable.
As observed by the district court, the pretrial stipulation does not
reference § 1981(a) at any point. Olmsted contends that the reference
10
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 11 of 33
in the pretrial statement to the Civil Rights Act of 1991 encompasses
both Title VII and § 1981; however, the reference to the Civil Rights
Act of 1991 mentions only § 1981a, not § 1981(a). While we are
reluctant to engage in an overly technical reading of a pleading when
the dispositive factor is the apparent absence of a set of parentheses,
these parentheses unfortunately control our decision. It is worth noting
that the complaint correctly cites to § 1981 and Title VII; we therefore
can assume that, notwithstanding the understandable confusion with
respect to the closely numbered statutory provisions at issue here, the
drafter of both the complaint and the pretrial stipulation knew the
difference between the two statutory avenues of relief.
Id. at 1462 (citation omitted).
And, in Morro v. City of Birmingham, 117 F.3d 508 (11th Cir. 1997), we
found that a district court didn’t abuse its discretion when it found that the defendant
waived a defense by not identifying it at a pretrial conference even though there was
“little doubt that the [defendant] would [have been] entitled to escape judgment
against it on that basis.” Id. at 513–15. There, a police officer for the City of
Birmingham sued the city under 18 U.S.C. section 1983 after the police chief
suspended him without pay. Id. at 511. After the city’s motion for summary
judgment was denied, the district court held a pretrial conference and entered a
pretrial order defining the scope of the issues for trial. Id. Then, after a jury returned
a verdict for the officer and the district court entered judgment, the city filed a motion
for judgment as a matter of law arguing that the police chief was not a “final
policymaker” with respect to disciplinary matters in the police department, so the
city could not be held liable. Id. The district court denied the motion, finding that
11
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 12 of 33
the city failed to preserve the defense because it was not included in the pretrial order
and the city never requested a modification of the pretrial order to inject the defense.
Id.
On appeal, we explained that “we have not hesitated to back up district courts
when they put steel behind the terms of pretrial orders and hold parties to them.” Id.
at 515. We observed that the city “failed to identify its potentially available . . .
defense as an issue at the pretrial conference or obtain a modification of the pretrial
order to permit it to raise the issue later in the proceedings.” Id at 515. Thus, we
concluded that even though we had “little doubt that the City would be entitled to
escape the judgment against it,” we could not “say that the district court abused its
discretion by failing to modify the pretrial order to accommodate presentation of the
. . . defense at the late state of the case at which the City chose to press it.” Id.
So too here. It’s clear Superior knew the difference between disgorgement,
punitive damages, and statutory damages—the complaint sought “actual,”
“general,” “statutory,” and “punitive damages.” Then, when it came time to agree
on the pretrial statement, Superior said it only sought “disgorgement of [Shaklee’s]
profits, pursuant to 15 U.S.C. [section] 1117(a).” And, when the district court asked
Superior what relief it was seeking, it said it only sought disgorgement. If Superior
also wanted to seek punitive and statutory damages, it could have and should have
said so.
12
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 13 of 33
We give substantial deference to the district court’s interpretation and
enforcement of pretrial statements and orders. See Hodges v. United States, 597
F.2d 1014, 1018 (5th Cir. 1979) (“[F]or pretrial procedures to continue as viable
mechanisms of court efficiency, appellate courts must exercise minimal interference
with trial court discretion in matters such as the modification of its orders. Thus, we
ascribe to the trial court a broad discretion to preserve the integrity and purpose of
the pretrial order . . . .” (citation omitted)); Risher v. United States, 465 F.2d 1, 5
(5th Cir. 1972) (“We are not inclined to disturb the district court’s interpretation of
a stipulation agreed upon by the parties during pretrial proceedings and approved by
the court.”). We consider only whether the district court abused its discretion. See
Morro, 117 F.3d at 513. Here, it did not.
Likelihood of Confusion
Next, we consider whether the district court clearly erred by finding that
Superior failed to demonstrate a likelihood of confusion. To determine whether a
defendant’s mark is likely to cause confusion, we consider seven factors: (1) the
type/strength of the plaintiff’s mark; (2) the similarity of the parties’ marks; (3) the
similarity of the products and services the marks represent; (4) the similarity of the
parties’ retail outlets (trade channels) and customers; (5) the similarity of the parties’
advertising media; (6) the defendant’s intent; and (7) actual confusion. Frehling
Enters. v. Int’l Select Grp., 192 F.3d 1330, 1335 (11th Cir. 1999). After making
13
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 14 of 33
subsidiary findings as to each factor, a court must also make an ultimate finding as
to the likelihood of confusion based on the “overall balance” of the factors. Custom
Mfg. & Eng’g, Inc. v. Midway Servs., Inc., 508 F.3d 641, 649 (11th Cir. 2007).
“[W]e review the district court’s findings as to each factor and its ultimate
conclusion regarding the likelihood of confusion only for clear error.” Fla. Int’l
Univ. Bd. of Trs. v. Fla. Nat’l Univ., Inc., 830 F.3d 1242, 1255 (11th Cir. 2016).
“[A] district court’s error in its analysis of one of the subsidiary factors in the
likelihood of confusion test is not enough to allow us to overturn the district court’s
decision.” Id. (citation and quotation marks omitted). “Rather, we must be
convinced that the district court’s ultimate conclusion—that the plaintiff established
a likelihood of confusion—is clearly erroneous.” Id. (citation omitted).
After considering these factors, the district court found that there was no
likelihood of confusion on the Healthprint mark. Superior contends that the district
court erred in applying some of the likelihood-of-confusion factors—namely, the
strength of the mark, the similarity of the marks, the similarity of the services, the
similarity of the advertising, Shaklee’s intent, and actual confusion—and, as a result,
clearly erred in finding no likelihood of confusion. We disagree. Although the
district court erred in its consideration of one of the seven factors—the similarity of
the advertising media—the district court’s likelihood of confusion finding was not
clearly erroneous.
14
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 15 of 33
1. Type/Strength of Superior’s Marks
We “[b]egin at the beginning.” Lewis Carroll, Alice in Wonderland (1865).
There are four categories of marks: (1) “generic”; (2) “descriptive”;
(3) “suggestive”; and (4) “arbitrary” or “fanciful.” Pinnacle Advert. & Mktg. Grp.,
Inc. v. Pinnacle Advert. & Mktg. Grp., LLC, 7 F.4th 989, 1004 (11th Cir. Aug. 2,
2021) (citation omitted). We consider fanciful, arbitrary, and suggestive marks to
be inherently “distinctive,” but not descriptive and generic marks. Id. Only
distinctive marks are entitled to trademark protection under the Lanham Act. Id.
Superior argues that it is unclear whether the district court applied descriptive or
suggestive strength to its Healthprint marks. If the district court applied only
descriptive strength, Superior argues, it clearly erred.
In affirming the denial of Superior’s motion for a preliminary injunction, we
concluded that the district court erred in finding that Shaklee rebutted the
presumption that Superior’s marks were not at least suggestive:
[W]hen Superior registered its marks, the USPTO did not require any
proof of secondary meaning. As a result, the district court correctly
found that Superior’s mark is entitled to the rebuttable presumption that
it is “inherently distinctive” and thus at least suggestive and entitled to
strong trademark protection.
The district court clearly erred, however, in concluding that Shaklee
rebutted that presumption with its evidence of third-party usage.
Superior, 710 F. App’x at 855.
15
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 16 of 33
After trial, the district court fixed its mistake. It recognized the presumption
for Superior and found that Shaklee “did not rebut [it].” The district court thus found
this factor in Superior’s favor, and therefore, did not clearly err in applying this
factor.
2. Similarity of the Parties’ Marks
Next, the district court found that the parties’ marks are “substantially
similar.” Superior argues that the district court erred in finding the subsidiary facts
as to the similarity of marks because it attempted to differentiate Shaklee’s
Healthprint mark by “focusing only on visual appearance.” We know this, Superior
contends, because the district court found that the competing marks are “nearly
identical . . . except for the use of Shaklee’s name preceding the mark.” Superior
argues that the district court’s finding is wrong because neither of Shaklee’s
trademark applications claim use of the name “Shaklee” with either of its Healthprint
marks.
But the district court didn’t focus only on the visual similarities between
Superior and Shaklee’s marks and turn a blind eye to everything else. Rather, it
explained that it was looking at “the overall impressions that the marks create.” It
then found that “Shaklee’s mark is nearly identical to Superior’s mark, except for
the use of Shaklee’s name preceding the mark.”
16
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 17 of 33
And the district court didn’t clearly err by emphasizing the use of the
Healthprint mark with Shaklee’s name. As we’ve explained in discussing the
similarity factor, “the court compares the marks and considers the overall
impressions that the marks create, including the sound, appearance, and manner in
which they are used.” Frehling, 192 F.3d at 1337 (emphasis added). This is
consistent with the Lanham Act, which prohibits a person from “us[ing] in
commerce any reproduction, counterfeit, copy, or colorable imitation of a registered
mark.” 15 U.S.C. § 1114(1)(a); see also J. Thomas McCarthy, McCarthy on
Trademarks and Unfair Competition, § 25A:1 (5th ed. 2020 update) (“[T]he use of
another’s trademark in the body of a web page can cause confusion and therefore,
infringement of that trademark.”). The evidence showed that Shaklee used the mark
on its website by having “Shaklee” next to “Healthprint.” The district court didn’t
clearly err in comparing the similarity of the marks when it considered how Shaklee
actually used the mark.
3. Similarity of the Products and Services the Marks Represent
Third, the district court found that Superior and Shaklee’s products and
services are “dissimilar,” and therefore the third factor “strongly support[ed]”
Shaklee. Superior argues that “[t]he district court’s similarity of services analysis
misapplied the law because it held Superior to a standard of showing identicality
rather than relatedness.” Superior also argues that the district court misunderstood
17
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 18 of 33
“the scope of Superior’s products and services” because Superior’s Healthprint mark
is not “solely directed to blood testing.” The district court didn’t clearly err in
evaluating the similarity of the parties’ products and services.
The similarity of services factor “requires a determination as to whether the
products [or services] are the kind that the public attributes to a single source, not
whether or not the purchasing public can readily distinguish between the products
of the respective parties.” Frehling, 192 F.3d at 1338. At the preliminary injunction
stage, we concluded that “the district court properly found that consumers are not
likely to believe that Superior is producing both comprehensive blood-testing
services and a free questionnaire.” Superior, 710 F. App’x at 856. Although
Superior presented evidence that blood testing was not required for a customer to
obtain a Healthprint, we noted that “the prosecution history files for Superior’s
trademark applications explicitly stated that the ‘HealthPrint requires a blood
draw.’” Id. at 856 (emphasis omitted).
After the bench trial, the district court found that “Superior presented no
evidence . . . to undermine” our finding at the preliminary injunction stage. We
agree. Superior told the patent office that its Healthprint product “requires a blood
draw.” And witnesses at trial confirmed that Superior didn’t advertise its non-blood
test option. In fact, that option wasn’t even listed on Superior’s order form;
customers could only find out about it if Superior mentioned it to them over the
18
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 19 of 33
phone. The parties’ goods and services were not “so related in the minds of
consumers that they get the sense that a single producer is likely to put out both
[parties’ goods and services].” Frehling, 192 F.3d at 1338. The district court didn’t
clearly err in focusing on Superior’s blood-test service and finding that the public
would not attribute it to the same source as Shaklee’s free online questionnaire.
4. Similarity of the Parties’ Retail Outlets and Customers
Superior doesn’t challenge the district court’s finding that the fourth factor
was neutral.
5. Similarity of Advertising Media
As to the similarity of the parties’ advertising media, the district court found
that both parties “use[d] word of mouth, the internet, and e-mail” and that “the
audience[s] for this media may overlap” but that “the disparity between the services
provided render[ed] this factor only slightly positive toward confusion.” Superior
argues that the district court misapplied the law as to this factor because it improperly
focused on the disparity of the parties’ services. According to Superior, “any
similarity or disparity between services has nothing at all to do with resolving the
similarity of advertising media factor; simply put, this was an irrelevant subsidiary
fact.” We agree.
Disparity of services is analyzed in the third factor, whereas the focus of the
advertising factor is on “each party’s method of advertising.” Frehling, 192 F.3d at
19
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 20 of 33
1339. “The key question in assessing similarity of advertising media is whether the
parties’ ads are likely to reach the same audience.” Fla. Int’l Univ., 830 F.3d at
1263. The district court erred by analyzing the advertising factor based on the
difference in services rather than a comparison of their advertising methods.
6. Shaklee’s Intent
Next, as to the intent factor, the district court found that there was “no credible
evidence” that Shaklee adopted the Healthprint mark to piggyback off Superior’s
goodwill. Rather, the district court found, “Shaklee engaged in a deliberate and
reasonable process to develop a mark which addressed Shaklee’s distinct online
questionnaire.” Superior contends that this conclusion was wrong because the
district court “refused to consider any direct evidence of bad faith presented by
Superior at trial” and “summarily discounted” Superior’s circumstantial inferences
as unreasonable. But the district court did no such thing, and its finding of no bad
intent wasn’t clearly erroneous.
As the district court explained, “[t]he evidence at trial showed that Shaklee
used an elaborate and lengthy process to select Healthprint as the mark to use on its
new online questionnaire.” Shaklee organized a working group that conducted
consumer testing and proposed several possible names before finally settling on
“Healthprint.” The working group compared logos prepared by a visual designer
and chose a green thumbprint. Shaklee then hired a law firm to do a trademark
20
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 21 of 33
search and discovered Superior’s registration of its Healthprint marks. George
Shehata, the head of Shaklee’s working group, went to Superior’s website and
noticed it was oriented toward blood testing and medical diagnostics. Shehata
concluded that Superior’s Healthprint service was sufficiently different from what
Shaklee proposed to do, so he didn’t think it was an issue to use the mark.
Superior’s primary evidence of bad intent was that Jennifer Steeves-Kiss, a
member of Shaklee’s working group, had heard about Superior’s Healthprint service
ten years earlier. But prior knowledge of a mark, without more, is not evidence of
an intent to infringe. See George & Co. LLC v. Imagination Ent. Ltd., 575 F.3d 383,
398 (4th Cir. 2009) (“[K]nowledge of another’s goods is not the same as an intent
‘to mislead and to cause consumer confusion.’” (citation omitted)); see, e.g., Amstar
Corp. v. Domino’s Pizza, Inc., 615 F.2d 252, 263 (5th Cir. 1980) (“There is no
evidence that defendants have ever attempted to ‘pass off’ their goods as those of
plaintiff. Evidence introduced at trial established that the name ‘Domino’s Pizza’
was adopted because Mr. Monaghan had been told he could no longer use the name
‘Dominick’s,’ the new name sounded Italian, and it was quite close to the old name.
Although Monaghan was aware of ‘Domino’ sugar at the time, he was unaware of
any other pizzerias by that name. There is no evidence the name was adopted with
any intent to confuse, mislead, or deceive the public.”). Indeed, “[t]here may be
several good faith reasons why the junior user decided to proceed even when aware
21
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 22 of 33
of the senior user’s mark. The most common reason for doing so is a good faith
belief that there is no conflict between the marks as used on the junior user’s goods
or services.” McCarthy, supra § 23:115. The trial evidence supported the district
court’s finding that Shaklee did not have a ‘“conscious intent to capitalize on the
plaintiff's business reputation,’ was ‘intentionally blind,’ or otherwise manifested
‘improper intent.’” Custom Mfg., 508 F.3d at 648. Thus, the district court did not
err as to this factor.
7. Actual Confusion
Finally, the district court found that the actual confusion factor “weigh[ed]
heavily against Superior” because Superior presented no evidence of actual
confusion. Superior argues that the district court erred because it “limited its
confusion analysis to only potential purchasers and stopped there.” Specifically,
Superior argues it presented evidence of actual confusion, including (1) a statement
from Arlys Sanderson, one of Shaklee’s distributors, that she was “confused” and
thought Shaklee and Superior were working together, and (2) “50 or so” calls from
Superior customers who were confused about Superior and Shaklee’s relationship.
But some evidence of actual confusion isn’t necessarily enough to tip the scales in
Superior’s favor when there was substantial evidence of an absence of confusion.
See, e.g., Fla. Int’l Univ., 830 F.3d at 1265 (concluding that one email from an
“actual customer” did not weigh in favor of a likelihood of confusion). We give the
22
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 23 of 33
district court, acting as a fact finder, “great latitude in determining the appropriate
weight to accord particular evidence.” AmBrit, Inc. v. Kraft, Inc., 812 F.2d 1531,
1544 (11th Cir. 1986). Here, it appropriately exercised that latitude.
We have explained that “[t]he actual confusion inquiry turns on both the
number of instances of confusion and the type of person confused.” PlayNation Play
Sys., Inc. v. Velex Corp., 924 F.3d 1159, 1167 (11th Cir. 2019). “All potential
consumers of the relevant product or service, including middlemen, can inform the
inquiry,” but “the ultimate consumers deserve special attention.” Caliber Auto.
Liquidators, Inc. v. Premier Chrysler, Jeep, Dodge, LLC, 605 F.3d 931, 936–37
(11th Cir. 2010). But the “50 or so” calls do not tip the scales for Superior and
compel the conclusion that there was actual confusion. Superior didn’t document
the calls, they were non-specific and vague, and the callers only asked whether
Superior had “partnered” with Shaklee to offer their product. The calls didn’t show
that the callers thought Shaklee was offering blood testing services or that Superior
was offering a free health questionnaire. The callers might have been “confused”
about Superior and Shaklee’s “relationship,” but they were not confused about the
services and products offered by them. See, e.g., The Sports Authority, Inc. v. Prime
Hosp. Corp., 89 F.3d 955, 963 (2d Cir. 1996) (explaining that “actual confusion”
means “confusion that enables a seller to pass off his goods [or services] as the goods
23
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 24 of 33
[or services] of another”). Thus, the district court, as the finder of fact, was entitled
to give little weight to the vague calls in analyzing the confusion factor.
Nor was Sanderson’s statement, alone, enough to show that the district court
clearly erred as to this factor. See, e.g., Fla. Int’l Univ., 830 F.3d at 1265 (holding
that “with only a single probative instance of consumer confusion in the years since
FNU’s name change, the district court reasonably decided that this factor did not
weigh in favor of a likelihood of confusion”). Although one instance of actual
confusion can sometimes be enough, see, e.g., Safeway Stores, Inc. v. Safeway Disc.
Drugs, Inc., 675 F.2d 1160, 1167 (11th Cir. 1982), abrogated on other grounds by
Tobinick v. Novella, 884 F.3d 1110 (11th Cir. 2018), here we are not left with a
“definite and firm conviction” that the district court erred by discounting
Sanderson’s confusion in light of the other evidence of no confusion. See Frehling,
192 F.3d at 1340.
By contrast, the district court had expert testimony that there was no actual
confusion. See AmBrit, 812 F.2d at 1544 n.68 (“[S]urvey evidence can be probative
on the actual confusion issue.”). The district court found Shaklee’s expert survey to
be “credible and highly persuasive.” The survey resulted in “a net rate of confusion
24
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 25 of 33
of no more than 7% between the two marks,” which the district court said
“demonstrate[d] that there is no likelihood of confusion.” 2
Given the vague and irrelevant calls, one confused distributor, and the
uncontested expert testimony that there was little-to-no actual confusion, the district
court was entitled to weigh the evidence and conclude that the actual confusion
factor weighed heavily for Shaklee. See id. (explaining that a finder of fact is
afforded “great latitude in determining the appropriate weight to accord to particular
evidence”); see also Hard Candy, LLC v. Anastasia Beverly Hills, Inc., 921 F.3d
1343, 1362 (11th Circ. 2019) (“If consumers have been exposed to two allegedly
similar trademarks in the marketplace for an adequate period of time and no actual
confusion is detected either by survey or in actual reported instances of confusion,
that can be powerful indication that the junior trademark does not cause a meaningful
likelihood of confusion.” (quoting Nabisco, Inc. v. PF Brands, Inc., 191 F.3d 208,
228 (2d. Cir. 1999), abrogated on other grounds by Moseley v. V Secret Catalogue,
2
Superior faults Shaklee’s expert for not “show[ing] respondents any stimuli directed to
Superior’s trademark for nutritional supplements and printed materials,” but this argument appears
only in the facts section of its initial brief and in Superior’s reply brief. Thus, Superior has
abandoned it. See Miccosukee Tribe of Indians of Fla. v. Cypress, 814 F.3d 1202, 1211–1212
(11th Cir. 2015) (“Our longstanding case law rule is than an appellant who does not raise an issue
in his opening brief may not do so in his reply brief, in a supplemental brief, in a rehearing petition,
or on remand from the Supreme Court.”); Sapuppo v. Allstate Floridian Ins. Co., 739 F.3d 678,
681 (11th Cir. 2014) (“Abandonment of an issue can [] occur when passing references appear in
the argument section of an opening brief, particularly when the references are mere ‘background’
to the appellant’s main arguments or when they are ‘buried’ within those arguments.”).
25
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 26 of 33
Inc., 537 U.S. 418 (2003)). The district court’s actual confusion finding was not
clearly erroneous.
8. Overall Balance
We arrive at the district court’s ultimate finding that there was no likelihood
of confusion. Superior has two issues with this conclusion.
First, Superior argues the district court “inexplicably devoted most of its
likelihood of confusion discussion to Shaklee’s intent.” Superior claims that “when
there is a finding of lack of intent to infringe,” as the district court found, then the
intent factor becomes “irrelevant” in the overall likelihood of confusion analysis.
But the district court didn’t devote most of its analysis to intent. The district
court concluded only that “[t]he intent factor [did] not support Superior’s claim.”
And its only reference to intent in the balancing section was to note that there was
“no evidence” of “malintent on Shaklee’s part.” In other words, the district court
simply recognized the absence of a factor that would have helped Superior; it didn’t
weigh that absence in Shaklee’s favor. See, e.g., Hard Candy, 921 F.3d at 1362
(concluding that where “[t]he district court never said that it was giving any
affirmative (let alone overwhelming) weight to its finding that [the defendant] lacked
intent to infringe” “there [was] no indication in the record that [the defendant]’s
intent received undue consideration”).
26
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 27 of 33
Second, Superior argues that the district court relied on the actual confusion
factor “almost to the exclusion” of the other factors. Superior notes that “[t]his court
has repeatedly held that evidence of actual confusion is not required for a finding of
likelihood of confusion.” Thus, Superior argues, “because a trademark owner need
not produce evidence of actual confusion to prevail on a trademark infringement
claim, the district court erred by weighing this factor ‘heavily’ against Superior.”
It is true that evidence of actual confusion “is not a prerequisite,” but
“evidence of actual confusion is the best evidence of a likelihood of confusion.”
Frehling, 192 F.3d at 1340. For this reason, we have referred to actual confusion as
one of “the two most important factors” (the other being the strength of the mark).
Fla. Int’l Univ., 830 F.3d at 1265; see also Custom Mfg., 508 F.3d at 649 (“We have
consistently held . . . that ‘[t]he type of mark and evidence of actual confusion are
the most weighty of considerations.’” (citation omitted)). And the absence of actual
confusion can weigh against a plaintiff. See, e.g., Hard Candy, 921 F.3d at 1362
(“In this case, there was enough opportunity for confusion to make the absence of
any evidence significant.”). Even so, “it is up to individual courts to assess this
factor in light of the particular facts of each case.” Frehling, 192 F.3d at 1340. The
district court did not err in weighing the actual confusion factor more heavily than
some of the other factors.
27
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 28 of 33
As to the overall balance, we conclude that the district court did not err in
balancing the factors and finding no likelihood of confusion. Even though the
district court erred in finding a lack of similarity of advertising media, the court
properly considered the overall balance of the factors and found no likelihood of
confusion. Unless we are left with a “definite and firm conviction that a mistake has
been committed,” we must affirm. Hard Candy, 921 F.3d at 1363. Here, even
weighing the lack of similarity of advertising media in Superior’s favor, we are not
left with a conviction that the district court’s balance was a mistake. See Superior,
710 F. App’x at 859 (concluding that the district court did not clearly err in denying
a preliminary injunction even though the district court improperly weighed some
factors). The Healthprint marks were used by Superior and Shaklee in different
contexts, the services provided by Superior and Shaklee were dissimilar, there was
no evidence of bad intent of Shaklee’s part, and there was little evidence of actual
confusion.
Exclusion of Superior’s Survey Evidence
Superior finally contends that the district court erred in excluding its expert
testimony. The district court excluded Superior’s expert survey under Federal Rule
of Evidence 702 because the survey was unreliable. The district court concluded
that the survey was “fatally flawed” because it: (1) did not comply with the Squirt
methodology it purported to apply; (2) did not use a control group; and (3) did not
28
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 29 of 33
include a description of Shaklee’s Healthprint service allowing survey participants
to compare it to Superior’s service. Superior argues the district court abused its
discretion in excluding the survey because, “[e]ven if the lack of a control group was
viewed as a methodological flaw, only the weight and not the admissibility should
have been affected.” In support, Superior cites to Jellibeans, Inc. v. Skating Clubs
of Georgia, Inc., 716 F.2d 833, 844 (11th Cir. 1983), where we said that a survey’s
“alleged technical deficiencies affect[ed] the survey’s weight . . . and not its
admissibility.”
Survey evidence is one way a party can demonstrate a likelihood of trademark
confusion. See PlayNation, 924 F.3d at 1169 (explaining that survey evidence is not
required to show a likelihood of confusion). Consumer surveys, like those used by
Superior and Shaklee here, present images and other information to participants and
solicit responses to determine whether there is a likelihood of trademark confusion.
See, e.g., Exxon Corp. v. Texas Motor Exch. of Houston, Inc., 628 F.2d 500, 506–
507 (5th Cir. 1980) (explaining that survey evidence is often introduced to show a
likelihood of confusion among the general public); Charles E. McKenney & George
F. Long, 1 Federal Unfair Competition: Lanham Act 43(a) § 8:2 (Dec. 2020 update)
(“Surveys are now so commonplace in their use to prove confusion . . . that the
absence of a survey has been noted in a number of cases.”). Here, Superior’s expert
conducted an online survey that asked participants if they thought “the Healthprint
29
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 30 of 33
name” was from a single company or more than one company. The survey then gave
participants additional information about Superior’s Healthprint service and asked
them whether “the Healthprint service” was from more than one company. After
considering the responses from 788 participants, Superior’s expert concluded that
the survey demonstrated a likelihood of confusion.
The district court didn’t abuse its discretion in excluding Superior’s survey
under Rule 702 because it was unreliable. See United States v. Brown, 415 F.3d
1257, 1265–66 (11th Cir. 2005) (“What is true about the review of evidentiary issues
in general applies with equal or even greater force to Daubert issues in particular, an
area where the abuse of discretion standard thrives.”); Kumho Tire Co. v.
Carmichael, 526 U.S. 137, 142 (1999) (“[T]he law grants a district court the same
broad latitude when it decides how to determine reliability as it enjoys in respect to
its ultimate reliability determination.” (emphasis omitted)). First, Superior’s survey
never presented images of the Healthprint marks to survey participants—it only
asked whether they associated the Healthprint name with more than one company.
Superior’s expert admitted that his survey didn’t show the Healthprint marks as they
were depicted in the marketplace, and the survey didn’t include any of Superior’s
advertising materials showing how the Healthprint marks were used. The district
court found, and we agree, that “the usage of the word ‘Healthprint’ alone, in plain
30
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 31 of 33
font, [was] particularly odd” given the importance of the visual similarity of Superior
and Shaklee’s Healthprint logos to the likelihood-of-confusion analysis.
Second, “[c]ourts have held that a survey that fails to use a control may be
given less weight or even excluded from evidence altogether.” McCarthy, supra
§ 32:187 (explaining that “control groups are necessary because they can show the
cause of confusion and because they reduce background ‘noise,’ including those
respondents who agree with most any survey assertion and those who tend to
‘quickly guess . . . because they are bored or hurried”). Superior’s survey did not
have a “control question.” Superior’s expert, when explaining the survey
methodology, admitted that controls were not included in the survey.
And third, as the district court found, Superior’s survey never explained
Shaklee’s Healthprint service to participants. Thus, there was no way for survey
participants to compare the two services, so it’s entirely unsurprising that they
thought the described service—i.e., Superior’s service—came from only one
company.
Shaklee’s expert, Poret, discussed these flaws at length in his declaration. He
explained that Superior’s survey was flawed because it didn’t show respondents any
advertising materials, pages from Superior’s website, or even Shaklee’s Healthprint
survey. “In the real world,” Poret explained, both Superior and Shaklee’s websites
convey additional information “that is important to a consumer’s perception of the
31
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 32 of 33
sources, such as company names and logos, web addresses, information regarding
the services, and various design and style elements.” Likewise, the lack of a control
group was problematic because, Poret explained, “[i]ncluding a control is the only
way one knows whether there [was] a ‘placebo effect’ or ‘false positive’ in a
consumer survey.” Finally, Poret observed that Superior’s survey “strongly bias[ed]
respondents to think that ‘Healthprint’ [was] the name of a specific company,” so it
was “unsurprising that, when respondents [were] . . . asked if ‘the’ Healthprint
service is from a single company or more than one company,” many said a single
company.
Although some defects in an expert’s methodology may go to its weight rather
than its admissibility, at some point, an expert’s methodology becomes so flawed
that it is unreliable. See McClain v. Metabolife Int’l, Inc., 401 F.3d 1233, 1245 (11th
Cir. 2005) (“The Daubert ‘requirement that the expert testify to scientific
knowledge—conclusions supported by good grounds for each step in the analysis—
means that any step that renders the analysis unreliable under the Daubert factors
renders the expert’s testimony inadmissible.’”). And where the methodology is
unreliable, the district court has a duty under rule 702 to exclude the expert’s
testimony. See id. Here, because Superior’s expert didn’t show the mark to survey
participants, didn’t use a control group, and didn’t explain Shaklee’s different
32
USCA11 Case: 19-10771 Date Filed: 09/28/2021 Page: 33 of 33
services, we can’t say the district court abused its discretion in excluding Superior’s
survey.
CONCLUSION
We conclude that the district court didn’t abuse its discretion in concluding
that Superior waived its claims for statutory and punitive damages and in excluding
Superior’s expert survey as unreliable. We also conclude that the district court did
not clearly err in finding no likelihood of confusion.
AFFIRMED.
33