Case Information
*2 Before M OORE , Chief Judge , UNNINGHAM , Circuit Judge ,
and S CARSI , District Judge [1] S CARSI , District Judge
Appellant Chateau Lynch-Bages (“Opposer”) appeals from a Trademark Trial and Appeal Board (“Board”) deci- sion partially dismissing its opposition to a trademark ap- plication filed by Appellee Chateau Angelus S.A. (“Applicant”). The Board determined that there was no likelihood of confusion between Applicant’s mark, “ECHO D’ANGÉLUS,” and Opposer’s mark, “ECHO DE LYNCH BAGES.” For the reasons below, we vacate the Board’s rul- ing and remand for further proceedings consistent with this opinion.
B ACKGROUND
Opposer asked the Board to deny an application seek- ing to register the mark “ECHO D’ANGÉLUS,” claiming that the proposed mark was likely to cause confusion with Opposer’s mark, “ECHO DE LYNCH BAGES.” Appx. 34, 36–37. The Board dismissed the opposition in part, finding that there was no likelihood of confusion as to the use of Applicant’s mark with certain classes of goods. Appx. 12– 27.
In its analysis, the Board considered the relevant fac- tors outlined in In re E.I. DuPont de Nemours & Co. , 476 F.2d 1357 (C.C.P.A. 1973) (“ DuPont factors”). Appx. 13– 26. First, the Board found that the similarity of the goods described in both Opposer’s registration and Applicant’s application weighed in favor of a finding of likelihood of confusion. at 14–15. Second, the Board concluded that the identical goods at issue, wine, “are presumed to move in the same channels of trade to the same classes of con- sumers,” such that “the parties’ trade channels and classes of customers are the same.” Id. at 15–16. Third, the Board determined that the degree of purchaser care was neutral. Id. at 16–17. And fourth, based on Applicant’s submission of third-party registrations that included the term “ECHO,” the Board ruled that “ECHO” is “a fairly com- monly-chosen term in the field,” which “weighs somewhat against finding that confusion is likely.” Id. at 17–19.
With two factors weighing in favor of confusion, one factor neutral, and one factor weighing “somewhat” against confusion, the Board then considered the similarities of the marks, which the Board correctly noted was “one of the most important considerations.” Id. at 20. The Board an- alyzed each mark as a “unitary expression” and found that the term “ECHO” did not dominate either mark. Id. at 22. Instead, the Board placed substantial weight on the terms “ANGÉLUS” and “LYNCH BAGES.” Id. While neither party appeared to argue that the subject marks included house marks, the Board nevertheless opined that “[i]t ap- pears that ANGÉLUS and LYNCH BAGES are the parties’ ‘house marks.’” Id.
In support of this finding, the Board noted that Appli- cant owned four registrations that included the term “ANGELUS,” and that Opposer’s name and its letterhead contained the phrase “LYNCH BAGES.” Id. Based on this, the Board “assess[ed] the effect of house marks” in evalu- ating the similarity of the marks. at 22–25. Indeed, the bulk of the Board’s similarity analysis consisted of evalu- ating case law to determine how much weight to place on the presence of the supposed house marks. Ultimately, the Board found that “ECHO” had “some conceptual weakness in connection with the goods at issue,” and that “the use of ECHO with D’ and DE followed by the parties’ house marks contributes significantly to the overall commercial impres- sions of the marks as invoking the respective house marks.” at 25–26. Because the common term between the marks was “somewhat weak,” the Board concluded that “the dissimilarities outweigh[ed] the similarities in the re- spective marks” given that the marks “incorporate[d] dif- ferent-appearing house marks as part of unitary expressions.” Id. at 26. Weighing all the factors “in light of all the evidence,” the Board found the dissimilarities in the marks to be a “predominant” factor, and “the overall balance of factors weigh[ed] against likely confusion.” at 26–27.
Opposer then filed the present appeal. We have juris- diction under 28 U.S.C. § 1295(a)(4)(B).
S TANDARD OF R EVIEW
A mark may be denied registration under Section 2(d) of the Lanham Act if it is “likely, when used on or in con- nection with the goods of the applicant, to cause confusion” with an already registered mark. 15 U.S.C. § 1052(d). Likelihood of confusion is a legal determination based on factual findings relating to the DuPont factors. On-Line Careline, Inc. v. Am. Online, Inc. , 229 F.3d 1080, 1084 (Fed. Cir. 2000). We review the Board’s factual findings for sub- stantial evidence, and we review the ultimate weighing of the DuPont factors de novo. QuikTrip W., Inc. v. Weigel Stores, Inc. , 984 F.3d 1031, 1034 (Fed. Cir. 2021). “Sub- stantial evidence is such relevant evidence as a reasonable mind would accept as adequate to support a conclusion.” Cai v. Diamond Hong, Inc. , 901 F.3d 1367, 1371 (Fed. Cir. 2018) (internal quotation marks omitted).
D ISCUSSION
On appeal, Opposer argues that the “Board erred in de- termining that each of the subject marks contained a house mark” because such a finding lacked substantial evidence. Appellant’s Br. 4, 15–17. In Opposer’s view, this “errone- ous determination” warrants reversal and remand. at 4, 25. We agree remand is appropriate, and conclude that the Board’s similarity analysis was flawed because its finding that the marks contained each party’s house mark was not supported by substantial evidence. See Stratus Networks, Inc. v. UBTA-UBET Commc’ns Inc. , 955 F.3d 994, 998 (Fed. Cir. 2020) (“[W]e evaluate whether the Board’s factual findings for each considered DuPont factor are supported by substantial evidence.”).
In considering the similarity of the marks, the Board declined to place an emphasis on the common term “ECHO” because it believed that the marks were unitary expressions not dominated by this term. [2] Appx. 21–22. Therefore, the Board’s analysis primarily revolved around comparing the terms “ANGÉLUS” and “LYNCH BAGES.” Here, rather than simply concluding that the substantial differences in the sound, appearance, and commercial im- pression of these terms rendered the marks dissimilar, see In re i.am.symbolic, llc , 866 F.3d 1315, 1323 (Fed. Cir. 2017), the Board opted to find that the terms “appear” to be the parties’ house marks. Appx. 22.
“The term ‘house mark’ refers to a trademark that is used throughout the commercial operations of the trade- mark owner.” 1 Anne Gilson LaLonde, G ILSON ON T RADEMARKS § 1.02[1][e] (Dec. 2024). Typically, parties ap- ply for the registration of a house mark and must provide evidence showing the broad use of the mark, as well as the mark’s use in commerce. See Trademark Manual of Exam- ining Procedure § 1402.03(b) (Nov. 2024). Therefore, the Board must first find substantial evidence of a mark’s use in commerce before determining that it is a house mark in the context of the DuPont factors.
There was limited evidence of such commercial use be- fore the Board. As a preliminary matter, the record does not suggest that either party argued that the subject marks contained house marks. Instead, the Board made this ob- servation based on its own investigation. Regarding “ANGÉLUS,” the Board noted that four of Applicant’s other registrations contain variations of the term. Appx. 22. The Board did not cite any authority supporting the proposition that merely owning multiple registrations with a common term renders that term a house mark. Nor did the Board find that these marks had a “broad use” or were used in commerce. The evidence that “LYNCH BAGES” was Opposer’s house mark was even more sparse. The Board dedicated a single sentence to the analysis, noting that Opposer’s name is “Chateau Lynch Bages”—impre- cisely omitting a hyphen—and that its letterhead used “LYNCH BAGES.” Id. The Board provided no further analysis, law, or citations to the record to support its find- ing. Id.
If that were the extent of the Board’s discussion of house marks, perhaps the error would be harmless. How- ever, the Board appeared to place great weight on its find- ings. Indeed, the Board expressly “assess[ed] the effect of house marks” on the similarity factor. Id. at 23. But, even here the Board’s analysis was unclear, as its survey of the case law resulted in a conclusion that “none of [its] prece- dents” governed the situation that it concocted itself: where the competing marks supposedly contained both a shared word and each party’s house mark. at 24–25.
After this underdeveloped house mark analysis, the Board considered the similarities of the two marks. It acknowledged that the identical term “ECHO” contributes to a finding of similarity, but that the “remainder of the marks are visually and aurally different.” at 25. Again, if the Board had arrived at this conclusion without the house mark discussion, its analysis may have been sound. However, it opined, “ECHO with D’ and DE followed by the parties’ house marks contributes significantly to the over- all commercial impressions of the marks as invoking the respective house marks.” at 25–26. And, because the Board considered “ECHO” weak, the Board reasoned that the marks were dissimilar given that they “incorporate dif- ferent appearing house marks.” at 26. The house mark findings were critical in the Board’s analysis and nearly dispositive to its conclusion that the marks were not simi- lar.
Applicant does not attempt to justify the Board’s house mark findings. Instead, Applicant simply states that “the ‘house’ marks are the ‘main’ components” in each mark. Appellee’s Br. 7–8. However, Applicant fails to engage with the relevant issue as to whether there was substantial evidence to determine that the marks contained house marks at all. Indeed, if there had been such evidence or argument, we agree that the Board would have had discre- tion to consider the prominence of the house marks in its similarity analysis. But that was not the case.
For the reasons above, the Board’s determination that the competing marks contained house marks was errone- ous, and because this finding was critical in the Board’s similarity analysis, the conclusion of dissimilarity lacked substantial evidence. Moreover, since the Board gave “pre- dominant” weight to the similarity factor in its DuPont analysis, it is not clear whether the Board would have arrived at the same conclusion of no likelihood of con- fusion absent this error. ONCLUSION
For the foregoing reasons, we vacate the Board’s deci- sion and remand the case for additional proceedings con- sistent with this opinion.
VACATED AND REMANDED OSTS
Costs to Opposer.
[1] The Honorable Mark C. Scarsi, District Judge, United States District Court for the Central District of Cal- ifornia, sitting by designation.
[2] Opposer argues that the Board also erred in con- sidering the marks as unitary expressions, but we do not expressly address this issue because it is unclear to what extent the house mark findings informed the Board’s deci- sion to view the marks as unitary expressions.