SWATCH AG (Swatch SA) (Swatch Ltd.), Plaintiff–Appellant, v. BEEHIVE WHOLESALE, LLC, a limited liability company, Defendant–Appellee.
Appellant Swatch AG brings this action seeking reversal of the district court's order denying its opposition to appellee Beehive Wholesale, LLC's trademark application and dismissing its related claims for federal, state, and common law trademark infringement, trademark dilution, and unfair competition. Swatch appeals on the ground that the district court's underlying factual findings—that there is no likelihood of confusion between Swatch's and Beehive's marks and that Beehive's mark is not merely descriptive—are clearly erroneous. For the reasons that follow we affirm.
Swatch is a well-known Swiss corporation that produces watches, clocks, jewelry, and various materials for watch collectors. It is the owner of three U.S. registrations for the mark SWATCH1 and for materials bearing that mark. Beehive is a Louisiana company engaged in wholesale and retail sales of a variety of products including watches and watch parts.
Beehive produces and sells watch bands and faces under the mark SWAP. The defining feature of these watch parts is that they are interchangeable. A purchaser of Beehive's watch components is able to affix any SWAP watch face to most or all SWAP watchbands. Swatch brand watches, which are typically sold at a higher price point, do not include interchangeable components.
On July 30, 2004, Beehive applied to the U.S. Patent and Trademark Office (“PTO”) to register its mark, SWAP, for use on its “[w]atch faces, ribbon watch bands, slide pendants, and beaded watch bands.” J.A. 315. Beehive's application was preliminarily granted and published for opposition2 on December 26, 2005. On April 14, 2008,3 Swatch filed a notice of opposition to Beehive's application on three grounds: 1) priority of Swatch's mark and likelihood of confusion; 2) mere descriptiveness of Beehive's mark; and 3) dilution of SWATCH by Beehive's use of SWAP. Swatch primarily argued that the similarity of Beehive's SWAP mark to its SWATCH mark in combination with the similar character of their products was likely to result in confusion among consumers as to the origin of the goods. It also argued that SWAP is too generic to be registered. The parties submitted evidence concerning their products, sales, revenue, and advertising, as well as deposition testimony regarding Beehive's selection of the SWAP mark. The opposition was heard and dismissed on all counts by the Trademark Trial and Appeal Board (“TTAB”).
Swatch then filed a civil action in the Eastern District of Virginia seeking the cancelation of Beehive's registration under 15 U.S.C. § 1071(b). Swatch added new claims for trademark infringement and federal unfair competition under the Lanham Act, 15 U.S.C. §§ 1114, 1125(a); trademark dilution under the Trademark Dilution Act, 15 U.S.C. § 1125(c); state trademark infringement under Va.Code § 59.1–92.12; and common law unfair competition. It also submitted documentary evidence not presented to the TTAB including facts relevant to the SWAP clock-face variant, the parties' channels of distribution, and Swatch's dilution-by-blurring claim. The parties agreed to forgo live testimony and have the matter decided solely on the written record. The district court, upon consideration of the materials before it, affirmed the TTAB, holding that its determinations were supported by substantial evidence. It also found facts based on evidence not presented to the TTAB pursuant to its authority under 15 U.S.C. § 1071(b)(3). The district court concluded, on the basis of these combined findings, that there was no likelihood of confusion between the two marks and no likelihood that SWAP would dilute SWATCH. It dismissed Swatch's infringement and unfair competition claims as a matter of law. It also concluded that Beehive's mark is registrable because it is not merely descriptive. This appeal followed.
On appeal, we review the district court's factual findings for clear error and its legal conclusions de novo. Likelihood of confusion is an “inherently factual issue,” and we “review[ ] district court determinations regarding [it] under a clearly erroneous standard.” Petro Shopping Ctrs., L.P. v. James River Petroleum Inc., 130 F.3d 88, 91–92 (4th Cir.1997). The strength of a mark and whether it is capable of being registered are also questions of fact that we review for clear error. Pizzeria Uno Corp. v. Temple, 747 F.2d 1522, 1533 (4th Cir.1984). We have yet to provide definitive guidance as to how dilution claims should be reviewed because the Trademark Dilution Revision Act, 15 U.S.C. § 1125(c) (“TDRA”),4 is a recent enactment. However, likelihood of dilution, like likelihood of confusion, is a fact-intensive inquiry, so the appropriate standard of review is clear error. Cf. Louis Vuitton Malletier S.A. v. Haute Diggity Dog, LLC, 507 F.3d 252, 264–65 (4th Cir.2007) (reciting multifactor factual tests for determining the likelihood of dilution, including proving the fame of the plaintiff mark and the likelihood that an association between two marks will impair the distinctiveness of the plaintiff mark).5
Section § 1071(b) of Title 15 of the United States Code permits a party in a trademark suit to initiate a civil action in the place of an appeal of the TTAB's determination to the Federal Circuit. 15 U.S.C. § 1071(b)(1). A brief description of the procedural features of § 1071(b) is helpful to our analysis.
In a § 1071(b) action, the district court reviews the record de novo and acts as the finder of fact. Durox Co. v. Duron Paint Mfg. Co., 320 F.2d 882, 883–84 (4th Cir.1963). The district court has authority independent of the PTO to grant or cancel registrations and to decide any related matters such as infringement and unfair competition claims. 15 U.S.C. § 1071(b)(1). The district court must admit the PTO record if a party so moves, and if admitted, the record “shall have the same effect as if originally taken and produced in the suit.” Id. at § 1071(b)(3). Whether or not the record is admitted, the parties have an unrestricted right to submit further evidence as long as it is admissible under the Federal Rules of Evidence and Civil Procedure. Id.; see also Kappos v. Hyatt, 132 S.Ct. 1690, 1700 (2012) (interpreting § 1071(b)'s patent parallel, 35 U.S.C. § 145).
Kappos is the primary case interpreting the patent and trademark civil action statutes. In Kappos, the PTO argued that in a § 145 proceeding where new evidence is admitted, the district court should defer to its findings, and “should overturn the PTO's factual findings only if the new evidence clearly establishes that the agency erred.” 132 S.Ct. 1690, 1695–96 (2012). The Supreme Court rejected the PTO's premise that a § 145 suit “creates a special proceeding that is distinct from a typical civil suit filed in federal district court,” id. at 1696, and adopted the Federal Circuit's position that “where new evidence is presented to the district court on a disputed fact question, a de novo finding will be necessary to take such evidence into account together with the evidence before the board.” Id. at 1700 (quoting Fregeau v. Mossinghoff, 776 F.2d 1034, 1038 (Fed.Cir.1985)). It held that the district court “does not act as the ‘reviewing court’ envisioned by the APA,” because the court must determine, among other things, “how the new evidence comports with the existing administrative record,” and “as a logical matter [it] can only make [this] determination[ ] de novo because it is the first tribunal to hear the evidence.” Id. at 1696, 1700.
Kappos also explicitly defines the only situation where consideration of the TTAB decision is permitted. The Court adopted the Federal Circuit's rule that “the district court may, in its discretion, ‘consider the proceedings before and findings of the Patent Office in deciding what weight to afford an applicant's newly-admitted evidence.’ “ Id. at 1700 (quoting Hyatt v. Kappos, 625 F.3d 1320, 1335 (Fed.Cir.2010)).6 In sum, where new evidence is submitted, de novo review of the entire record is required because the district court “cannot meaningfully defer to the PTO's factual findings if the PTO considered a different set of facts.” Id.
We are constrained to conclude that the standard of review articulated by the district court is erroneous. Its statement that in a § 1071(b) proceeding it sits in a “dual capacity” and applies a “unique standard of review,” acting in part as an appellate body, is in tension with the statute and directly conflicts with the requirements of Kappos. Swatch, S.A. v. Beehive Wholesale, L.L.C., 888 F.Supp.2d 738, 745 (E.D.Va.2012).
However, it is not clear from the record that the district court did in fact improperly defer to the factual findings of the TTAB, and remand is therefore unnecessary. As an initial matter, the district court properly reviewed Swatch's dilution-by-blurring claim entirely de novo because the TTAB did not address it on the merits. Swatch, 888 F.Supp.2d at 756 n. 15. The district court also decided Swatch's trademark infringement and unfair competition claims, which were not before the TTAB, de novo. Although the district court stated that it would apply an impermissible hybrid review to its likelihood of confusion and strength-of-the-mark analyses, there are more than sufficient facts recited in its opinion to support its findings.
The district court affirmed the TTAB's finding that SWAP is not merely descriptive and found on a de novo review of newly submitted evidence that SWAP was suggestive and not merely descriptive. Although the district court stated that it was deferring to the TTAB's findings on the administrative record, it also stated that it undertook “de novo review of the Swap's registerability to account for new evidence,” as required by the statute. Id. at 760. While it is not clear exactly what standard of review the district court applied, it appears that it made sufficient factual findings of its own to justify a holding that SWAP is in fact suggestive and not descriptive.
Because of the nature of the strength-of-the-mark inquiry, the district court's possible failure to reexamine all of the evidence de novo does not have a substantial effect on its analysis. Marks are divided into four categories of distinctiveness. George & Co., LLC v. Imagination Entertainment Ltd., 575 F.3d 383, 393–94 (4th Cir.2009). The second and third categories, suggestive and descriptive marks, are at issue here. Suggestive marks are “inherently distinctive.” Id. They “do not describe a product's features but merely suggest[ ] them,” and are therefore entitled to trademark protection. Id. A descriptive mark “define[s] a particular characteristic of the product in a way that does not require any exercise of the imagination.” Id. It is thus not entitled to protection unless it has acquired a secondary meaning. Id.
The question before the district court was whether SWAP is merely descriptive of a feature of Beehive's products, specifically the interchangeability of its watch parts. While we have recognized that it is often difficult to categorize suggestive versus descriptive marks, it is nevertheless an objective factual matter to be decided by the trier of fact based on the meaning of a mark and the attributes of the accompanying product. Because the central distinction is whether or not the mark imparts information about the goods directly, id., it is difficult to see what evidence outside of the mark and the attributes of the product could be of assistance to the court.7
For this reason, we find that even if the district court considered only the newly adduced evidence de novo, it found facts sufficient to support its finding that SWAP is not merely descriptive. The district court found, based on the new advertisements Swatch submitted, that SWAP was suggestive because merely showing the mark and the product together would be insufficient to convey its attributes. While a merely descriptive mark could be used in this manner, “[e]xplaining the function of [Beehive's] product requires a further leap.” Swatch, 888 F.Supp.2d at 760. The court found that individuals, even retail professionals familiar with the products,8 had to “exercise some imagination to connect ‘SWAP’ with the function of Beehive's watches,” as evidenced by diagrams on Beehive's advertisements that “not only add[ ] arrows but adorn[ ] ‘SWAP’ with the word ‘it!’ “ Id.
We find no error in these determinations, and Swatch's arguments on this point are unavailing. Aside from its conclusory assertions that SWAP describes the interchangeability feature of Beehive's watches, Swatch primarily argues that the district court erred by failing to credit admissions by Beehive's executives that SWAP is descriptive. For purposes of this analysis the district court's treatment of these statements does not matter. As noted above, descriptiveness is an objective determination that the district court makes on the basis of the meaning of a mark and the features of any associated products. For this reason, the opinion of a witness, particularly a lay witness, that a mark is descriptive rather than suggestive can be of no assistance.
The district court found that there was no likelihood of confusion between SWATCH and SWAP. Swatch argues on appeal that the district court erred because SWAP is so similar to SWATCH that customers are likely to be confused about the origin of Beehive's products. Although the district court did improperly review TTAB determinations for substantial evidence in some instances, because it was presented with different evidence and applied a test with different factors from the TTAB,9 the court made sufficient de novo determinations to support its finding.
To demonstrate a likelihood of confusion, Swatch must prove both that it has a valid and protectable mark, an element that Beehive concedes in this case, and that Beehive's use of SWAP creates a likelihood of confusion. Petro Shopping Ctrs., 130 F.3d at 91. Likelihood of confusion is an “inherently factual issue.” Id. at 92.
A likelihood of confusion exists between two marks if “the defendant's actual practice is likely to produce confusion in the minds of consumers about the origin of the goods or services in question.” CareFirst of Maryland, Inc. v. First Care, P.C., 434 F.3d 263, 267 (4th Cir.2006) (quoting KP Permanent Make–Up, Inc. v. Lasting Impression I, Inc., 543 U.S. 111, 117 (2004)). To determine if there is a likelihood of confusion between two marks, we consider nine non-exclusive and non-mandatory factors:
(1) the strength or distinctiveness of the plaintiff's mark as actually used in the marketplace; (2) the similarity of the two marks to consumers; (3) the similarity of the goods or services that the marks identify; (4) the similarity of the facilities used by the markholders; (5) the similarity of advertising used by the markholders; (6) the defendant's intent; (7) actual confusion; (8) the quality of the defendant's product; and (9) the sophistication of the consuming public.
George & Co., 575 F.3d at 393. These nine factors serve as a guide rather than “a rigid formula for infringement”; they are not all of equal importance and not all factors are relevant in every case. Id.10
The parties agree on appeal that the first and third factors weigh in favor of Swatch. The district court found, however, that the other factors weigh heavily against a likelihood of confusion.
First, the district court determined under the second factor that SWATCH and SWAP are not confusingly similar. The court stated that it “agree[d]” with the TTAB that the marks themselves were dissimilar in sight, sound, and meaning and that the TTAB's finding was therefore supported by substantial evidence, implying impermissible deference. Swatch, 888 F.Supp.2d at 750. However, in a practical sense, the district court could not have concluded that the TTAB's finding was supported by substantial evidence without performing a de novo review. It is clear from both the TTAB's and the district court's opinions that the only relevant evidence for this finding was the marks themselves and dictionary definitions of the words they contain.11 The district court clearly considered the marks themselves and the dictionary definitions, which appear in its opinion. Although the district court said it agreed with the TTAB, given the nature of the question before it there was little need for deference. It would be apparent to an average consumer that SWATCH and SWAP: 1) look different when written; 2) sound different when spoken; and 3) have completely different meanings in common usage.
Swatch argues that the district court erred because it should have focused on “SWA-“ as the dominant portion of both marks. We compare whole words, not parts, see Sara Lee Corp. v. Kayser–Roth Corp., 81 F.3d 455, 465 (4th Cir.1996), and generally use the phrase “dominant portion” to refer to the non-generic words in multiword marks, see Lone Star Steakhouse & Saloon, Inc. v. Alpha of Virginia, Inc., 43 F.3d 922, 936 (4th Cir.1995) (identifying “Lone Star” as the dominant portion of the “Lone Star Steakhouse & Saloon” and “Lone Star Grill” marks). Therefore the district court correctly considered the similarities between SWAP and SWATCH as whole words and did not err in finding that the stylized marks are dissimilar.12
Moreover, the district court explicitly determined de novo that the marks were dissimilar as they generally appear in commerce. Swatch, 888 F. Supp 2d at 751. The appearance of the mark in commerce is the relevant inquiry under our precedent. Anheuser–Busch, Inc. v. L & L Wings, Inc., 962 F.2d 316, 319 (holding that to determine if the use of a mark creates a likelihood of confusion with a protected trademark “we must examine the allegedly infringing use in the context in which it is seen by the ordinary consumer.”)
The district court correctly noted that accompaniments to marks and the manner in which they are presented in connection with goods can significantly reduce the likelihood of confusion between similar marks. See CareFirst, 434 F.3d at 271–72. It found that SWATCH almost exclusively appears accompanied by a Swiss flag both on products and in advertisements, while SWAP is generally accompanied by the phrase “by Beehive.” It also found that there was a substantial difference in the fonts used by the two marks. Swatch's argument that courts may only consider elements accompanying a mark when that mark is weak, while SWATCH was found to be strong, misstates our precedent. We have held only that disparate design elements are “most significant” when the mark claiming infringement is weak. Id. at 271. Its other argument, that the district court failed to consider the similar placement of SWATCH and SWAP on watch faces, is frivolous. There are a limited number of places that a mark can reasonably appear on a watch face, and, more importantly, most of Swatch's watch faces include the word “Swiss” while SWAP appears alone. The district court's finding of dissimilarity in commerce was therefore not erroneous.
The court also found on a de novo review of the evidence that the SWAP clock-face variant, which frequently appears on products and advertisements, did not create a likelihood of confusion. On appeal, Swatch repeats its argument that the clock-face variant, in which SWAP appears over the word “watch,” and the lower curve of the “S” contains two clock hands, confuses consumers by indicating that they should read “S” and “watch” together. The district court did not err by finding this argument unpersuasive; the picture in the “S” curve is clearly intended to portray clock hands and not arrows connecting the two words. Therefore the district court did not err by finding that the clock-face variant of SWAP is not confusingly similar to SWATCH.
Finally, we have never held, as Swatch argues, that courts in this Circuit must apply a lower threshold for similarity of marks when the parties' products are more similar. Swatch has not advanced any argument that justifies the application of this rule when the similarity of the parties' products is already accounted for under the third factor of the analysis.
The district court properly made all of its findings under the fourth and fifth factors de novo. The TTAB did not find facts related to the similarity of Swatch's and Beehive's facilities because the relevant facts were not before it. Swatch, 888 F.Supp.2d at 752 n. 12. The TTAB did not find facts related to the similarity of Swatch's and Beehive's advertising because advertising is not a factor in the DuPont test. In re E.I. DuPont de Nemours & Co., 476 F.2d 1357, 1361 (C.C.P.A.1973).
The district court found that the similarity of the parties' facilities gave “insignificant support” to Swatch because there are “basic differences between plaintiff's and defendant's modes of distributing their products.” CareFirst, 434 F.3d at 273. Although the district court acknowledged some similarity, it found the overlap to be de minimis between Swatch, which sells primarily to end consumers in Swatch-brand stores and department stores, and Beehive, which is primarily a wholesaler selling to independent retail and gift stores and through trade shows. It found insignificant overlap in Swatch's minimal sales to gift and jewelry stores and Beehive's minimal department store sales. It found no evidence in the record that the parties' products had ever been sold in the same store, and found that both parties' internet sales were limited to their own brand websites. Swatch's attempts on appeal to distinguish the facts of this case from those of CareFirst and Louis Vuitton Malletier with conclusory assertions that Beehive's goods and channels of trade are identical to its own are unavailing. There is no error in the district court's finding that this factor is of no assistance to Swatch.
The district court found that Beehive's advertising is limited in scope and directed almost entirely toward wholesale customers through catalogues and appearances at trade shows. It found that Swatch advertises through television, magazines, social media, and a billboard in Times Square. The district court also found that neither party had purchased advertisements on the internet, and that the mere maintenance of a brand store on the internet does not constitute advertising. Swatch appears to concede on appeal that there are substantial differences in its and Beehive's advertising but argues speculatively that Beehive might have reduced or changed its advertising as a result of Swatch's opposition to its trademark application. This contention finds no support in the record. The district court did not err by finding that the dissimilarity of the parties' advertising favored Beehive.
Although the district court improperly characterized its findings regarding the sixth factor, “ ‘intent to confuse the buying public,’ “ as an affirmance of the TTAB, that characterization did not render its findings erroneous. George & Co., 575 F.3d at 397 (quoting Pizzeria Uno Corp., 747 F.2d at 1535). In support of its allegation that Beehive intended to mislead consumers, Swatch argued in its trial brief, as before the TTAB, only that 1) Beehive had prior knowledge of Swatch's mark, 2) Beehive failed to conduct a trademark dispute despite this knowledge, and 3) the marks are confusingly similar as they are used in commerce.13 We have already determined that the district court's finding of the dissimilarity of the marks in commerce was not erroneous. Although the district court states that it is affirming the TTAB on the other two arguments, the facts that Beehive knew of SWATCH before selecting its mark and that it failed to conduct a trademark analysis are not in dispute. Therefore, the district court's holdings that these facts do not constitute bad faith under Action Temp. Servs. Inc. v. Labor Force Inc., 870 F.2d 1563, 1566 (Fed.Cir.1989),14 and George & Co., 575 F.3d at 398,15 are conclusions of law. Because we review conclusions of law de novo, there can be no prejudice to Swatch from the district court's improper characterization. We find no error in the district court's conclusion that Beehive's actions do not show an intent to mislead consumers.
Because the TTAB applies a different standard for actual confusion, the district court properly found on a de novo review of the record that there was no evidence of actual confusion, and that this factor weighed heavily in favor of Beehive. The district court found that although SWAP had been in use since 2003, Swatch had not adduced evidence of a single instance of actual customer confusion about the origin of the parties' confusion is generally considered to be the “most important factor” in a likelihood of confusion analysis, George & Co., 575 F.3d at 398, and that the “absence of any evidence of actual confusion over a substantial period of time ․ creates a strong inference that there is no likelihood of confusion.” CareFirst, 434 F.3d at 269. On appeal, Swatch does not contend that the record includes any evidence of actual confusion. It argues that the district court accorded too much weight to the lack of actual confusion in its overall analysis. However, as the court noted, the factors are weighed differently depending on the circumstances of the cases, and actual confusion is “often paramount.” CareFirst, 434 F.3d at 268 (quotation marks and citation omitted). Swatch also attempts to argue, as it did under the fifth factor, that the lack of actual confusion could have resulted from Beehive modifying its advertising and use of the mark as a result of Swatch's opposition. Again this argument is speculative and unsupported by the record. The district court's finding that there is no evidence of actual confusion between SWATCH and SWAP was not erroneous.
The district court properly found, on a sufficient de novo review of the entire record, that despite the fame of SWATCH and the similarity of the goods, the “lack of similarity between the marks, lack of predatory intent, lack of similar advertising and only minimal similarity in facilities,” in combination with “the most significant factor, actual confusion,” resulted in no likelihood of confusion between SWATCH and SWAP. Swatch, 888 F.Supp.2d at 756 (internal quotation marks and citations omitted).
The district court dismissed Swatch's trademark infringement and unfair competition claims as a matter of law. We review this holding de novo. Having found no error in the district court's finding that there is no likelihood of confusion between SWATCH and SWAP, we conclude that Swatch's federal, state, and common law trademark infringement and unfair competition claims must be dismissed. The Lanham Act requires a finding of a likelihood of confusion to prove federal trademark infringement, 15 U.S.C. § 1114(a), and federal unfair competition, 15 U.S.C. § 1125(a)(1)(A). Virginia law requires a finding of likelihood of confusion to prove trademark infringement. Va.Code Ann. § 59.1–92.12(i). Virginia common law trademark infringement and unfair competition claims require the same proof, including a likelihood of confusion, as the Lanham Act offenses. See Lone Star Steakhouse & Saloon, 43 F.3d 922, 930 n. 10 (4th Cir.1995). Because there is no likelihood of confusion between the marks, these causes of action fail as a matter of law.
Because the TTAB did not decide Swatch's dilution claim on the merits, the district court properly reviewed the entire record de novo to determine that there is no likelihood that SWAP will dilute SWATCH by blurring. Swatch, 888 F.Supp.2d at 756 n. 15. Therefore, we review the district court's finding for clear error.
Dilution by blurring occurs when the “ ‘association arising from a similarity between’ “ two marks “ ‘impairs the distinctiveness of the famous mark.’ “ Louis Vuitton Malletier, 507 F.3d at 264 (quoting 15 U.S.C. § 1125(c)(2)). To succeed on a dilution claim, the plaintiff must show that 1) it owns a famous, distinctive mark, 2) the defendant uses an allegedly diluting mark in commerce, 3) an association arose from the similarity of the marks, and 4) the association is likely to impair the distinctiveness of the famous mark. Id. at 264–65. The district court assumed without deciding that Swatch had satisfied the first three factors, but found that it had not proven the fourth.
On appeal, Swatch argues only that “the district court erroneously found a lack of similarity between the SWATCH and SWAP marks [and] overlooked evidence that Beehive's goods copied both the SWATCH mark and designs” in its dilution analysis. Appellant's Br. 43. As explained in detail above, the two marks are not confusingly similar and there is no evidence that Beehive intended to confuse consumers by copying Swatch's designs or otherwise. Finding no clear error, we affirm on this ground.
For the foregoing reasons, the district court's order denying Swatch's opposition to Beehive's trademark application and dismissing Swatch's related claims for federal, state, and common law trademark infringement and unfair competition is
1. For clarity, the marks will be rendered throughout in capital letters while “Swatch” will refer to the appellant.
2. Interested parties may challenge the registration of a mark by filing a notice of opposition with the PTO during the mark's publication period.
3. Although Swatch initially filed an opposition on February 3, 2006, its 2008 amended opposition is the relevant filing for our purposes because it was the basis of the Trademark Trial and Appeal Board's proceeding.
4. Pub.L. No. 109–312, 120 Stat. 1730 (2006). The original act required proof of actual dilution and actual economic harm. Louis Vuitton Malletier, 507 F.3d 252, 264 n. 2 (4th Cir.2007).
5. Our only other published case deciding the question of dilution under the TDRA was in the posture of a summary judgment and was therefore reviewed de novo. Rosetta Stone Ltd. v. Google, Inc., 676 F.3d 144 (4th Cir.2012). However in our unpublished decision in Super Duper, Inc. v. Mattel, Inc., we explicitly stated that the elements of the TDRA analysis are factual determinations. 382 F. App'x 308, 314 (4th Cir.2010).
6. Although it is not obvious from Kappos exactly what this means, it is explained quite clearly in Hyatt, the underlying Federal Circuit decision. The Federal Circuit permits district courts to give “less weight to evidence whose reliability is impacted by an applicant's failure, without explanation, to provide it to the Patent Office.”Hyatt, 625 F.3d at 1335. A district court may, but is not required to, review the PTO proceedings and give new evidence less weight if “the facts of a particular case cast suspicion on the new evidence that an applicant failed to introduce before the [PTO].” Id. The district court did not find that Swatch improperly withheld evidence, and Kappos seems to prohibit any other reliance on the TTAB's findings and conclusions.
7. Our precedent supports this position. Although we have considered outside evidence of distinctiveness in the past, it is generally evidence of the meaning of the mark, or evidence that a descriptive mark has acquired secondary meaning, neither of which are at issue here. E.g., U.S. Search, LLC v. U.S. Search.com, Inc ., 300 F.3d 517, 524–25 (4th Cir.2002) (determining the meaning of “search” using web pages, media references, and recruiting materials, and determining if “U.S. Search” had acquired secondary meaning based on a fact-intensive six-factor inquiry).
8. Swatch contends on appeal that the district court erred by failing to consider the descriptiveness of the mark as it would appear to a consumer aware of the specific nature of the goods. This argument is meritless. First, the court's opinion clearly indicates that it evaluated the mark as it would appear to “Beehive's wholesaler customers,” who are undoubtedly familiar with its goods. Swatch, 888 F.Supp.2d at 760. Second, while we have previously held that “the distinctiveness of a mark is measured in connection with the particular goods or services that are being offered,” we have not required that the goods be defined with the degree of specificity Swatch desires. U.S. Search, LLC v. U.S. Search.com, Inc., 300 F.3d 517, 524 (4th Cir.2002); id. (citing as hypothetical examples of “particular goods” generic sellers such as “a shop that sells diamonds, ․ a furniture polish company, and ․ an applesauce cannery.”)
9. The TTAB follows the Court of Customs and Patent Appeals' (predecessor to the Federal Circuit) thirteen factor DuPont test for likelihood of confusion. In re E.I. DuPont de Nemours & Co., 476 F.2d 1357 (C.C.P.A.1973).
10. For example, in this case the district court properly declined to consider the eighth and ninth factors in the George & Co. analysis. Factor eight, the quality of the defendant's product, “is most appropriate in situations involving the production of cheap copies or knockoffs of a competitor's trademark-protected goods.” Sara Lee Corp. v. Kayser–Roth Corp., 81 F.3d 455, 467 (4th Cir.1996). Despite Swatch's assertions that Beehive's products are both similar to its products and sold at a lower price point, there is no evidence that Beehive's products are “markedly inferior” and rely heavily on similarity to Swatch's products “to generate undeserved sales.” Id. As for factor nine, in the typical case, buyer sophistication is only considered “when the relevant market is not the public at-large.” Id. The record clearly reflects that both parties hope and intend to sell their watches to as many consumers as possible without restriction.
11. Swatch submitted the American Heritage College Dictionary definitions for “swap” and “swatch” to the TTAB.
12. Swatch also argues, somewhat confusingly given its focus on “swa-“, that we should consider as evidence of their similarity the fact that both full marks are sometimes accompanied by the word “watch.” However, generic terms do not aid our analysis. If one mark was accompanied by “watch” and the other by “timepiece,” we would not consider that to be evidence of dissimilarity.
13. Swatch argues for the first time on appeal that Beehive mimicked Swatch's product designs and that this serves as evidence of intent. This argument has been waived and is not properly before us. Moreover it is unsupported by the record.
14. “[M]ere knowledge of the existence of the prior user should not, by itself, constitute bad faith.”
15. “[T]he failure to conduct a trademark search or contact counsel shows carelessness at most, but is in any event irrelevant because knowledge of another's goods is not the same as an intent to mislead and to cause consumer confusion.” (internal quotation marks and citations omitted).
DUNCAN, Circuit Judge:
Affirmed by published opinion. Judge DUNCAN wrote the opinion, in which Judge NIEMEYER and Judge KING joined.