IN RE: I.AM.SYMBOLIC, LLC, Appellant
i.am.symbolic, llc (“Symbolic”) appeals from decisions of the U.S. Patent and Trademark Office (“the PTO”) Trademark Trial and Appeal Board (“the Board”) affirm-ing the examining attorney's refusal to register the mark I AM (“the mark”) in standard characters for goods in International Classes (“classes”) 3, 9, and 14 on the ground of a likelihood of confusion with registered marks. See In re i.am.symbolic, llc, 116 U.S.P.Q.2d 1406 (T.T.A.B. Oct. 7, 2015) (Symbolic I); In re i.am.symbolic, llc, No. 85044495, 2015 WL 6746544 (Oct. 7, 2015) (Symbolic II); In re i.am.symbolic, llc, No. 85044496, 2015 WL 6746545 (Oct. 7, 2015) (Symbolic III). Because the Board did not err in its likelihood of confusion conclusion, we affirm.
Symbolic owns the mark I AM (typed drawing) for “clothing, namely, hats, caps, socks, shirts, t-shirts, sweatshirts, tank tops, shorts, pants, sweatpants, jeans, swimwear, swimsuits, beachwear and footwear, namely, shoes, athletic footwear, boots, clogs, sneakers and san-dals” in class 25 (“class 25 registration”). Symbolic I, 116 U.S.P.Q.2d at 1408. It also owns the mark WILL.I.AM (standard characters) for certain goods in class 9 and services in class 41. Id.
I. The Applications and Rejections
Symbolic's predecessor-in-interest, William Adams (“Adams”), filed trademark applications for registration of the mark for goods in classes 3, 9, and 14 on an intent-to-use basis under 15 U.S.C. § 1051(b).1 The examining attorney refused registration on the ground of likelihood of confusion with previously registered I AM marks pursuant to 15 U.S.C. § 1052(d) for the same or similar goods. The applications were amended during prosecu- tion to include with the identification of goods the state-ment “associated with William Adams, professionally known as ‘will.i.am’ ” (the “will.i.am restriction”).
The identification of goods for class 3, as amended, recites:
Cosmetics; artificial eyelashes; body powder; in-cense; nonmedicated lip balm; non-medicated hair care preparations; non-medicated nail care preparations; nonmedicated skin care preparations; makeup remover; massage oils; essential oils for personal use; shaving creams and gels and depila-tory creams and gels; shower and bath gels, bath crystals, milks, oils, bubble bath, powders and salts; soaps and detergents; sun screen and sun tanning preparations; toothpaste and mouthwash, all associated with William Adams, professionally known as “will.i.am” (in International Class 3).
Symbolic I, 116 U.S.P.Q.2d at 1408 (emphases added). The examining attorney refused registration due to a likelihood of confusion with the previously registered mark I AM (typed form) for “perfume” in class 3 (“Siegel Registration”).2 Id.
The identification of goods for class 9, as amended, recites, in relevant part:3 “sunglasses and sunglass cases associated with William Adams, professionally known as will.i.am.” Symbolic II, 2015 WL 6746544, at *1 (emphasis added). The examining attorney refused registration due to a likelihood of confusion with the previously registered mark I am for “sunglasses” in class 9 (“Beeline Registration”).4 Id. at *2.
The identification of goods for class 14, as amended, recites:
Brass wrist cuff bracelets; clocks; jewelry, namely, dog tags for wear by humans for decorative purposes; jewelry; jewelry boxes; jewelry sets, namely, necklaces, earrings and bracelets; key holders, rings and chains of precious metal; lapel pins; medals; ornamental pins, rubber wristbands in the nature of bracelets; silver wrist cuff bracelets; sun dials; watch bands; watch bracelets; watch cases; watches; wrist bands of imitation leather; wrist bands of leather; and jewelry made of resin, namely, wrist cuffs of resin, all associated with William Adams, professionally known as “will.i.am” (in International Class 14).
Symbolic III, 2015 WL 6746545, at *1 (emphases added).
The examining attorney refused registration on the ground of a likelihood of confusion with the previously registered marks I am for “jewelry and fashion jewelry, bracelets, anklets, necklaces, pendants, earrings, ear clips, broaches, finger rings, arm rings; watches, wrist watches, pocket watches, watch chains and watch fobs” in class 14 (Beeline Registration) and I AM (standard character form) for “silicone stretchable wrist band in the nature of a bracelet” in class 14 (“Finch Registration”5 and, collectively with the Siegel Registration and Beeline Registration, “registrants' marks”). Id.
II. The Board Decisions
On appeal, the Board affirmed the examining attorney's refusals to register the mark based on a likelihood of confusion. The Board first rejected Symbolic's arguments based on the will.i.am restriction. It explained:
[W]e view the language “associated with William Adams, professionally known as ‘will.i.am’ ” in [Symbolic's] identification of goods as merely high-lighting an association with [Symbolic's] presumed principal, Mr. Adams. Contrary to [Symbolic's] assertion, we do not see the language as imposing a meaningful limitation on [Symbolic's] goods in any fashion, most especially with respect to either trade channels or class of purchasers. The language does not, in any meaningful way, alter the nature of the goods identified; nor does it represent that the goods will be marketed in any particular, limited way, through any particular, limited trade channels, or to any particular class of customers. It does not even represent that Mr. Adams will be named, or otherwise identified, in the promotion of the goods. The language “associated with William Adams, professionally known as ‘will.i.am’ ” is precatory language, and not binding on consumers when they encounter Applicant's mark.
Symbolic I, 116 U.S.P.Q.2d at 1409–10 (emphases added); Symbolic II, 2015 WL 6746544, at *4 (emphases added); Symbolic III, 2015 WL 6746545, at *4 (emphases added).
The Board found that Adams is the well-known front man for the music group The Black Eyed Peas and is known as will.i.am. The Board also found, however, that the record did not establish that Adams is “widely known by ‘i.am’ (as opposed to ‘will.i.am’), or that ‘i.am’ and ‘will.i.am’ are used interchangeably by either Mr. Adams or the public,” and rejected Symbolic's arguments based on its ownership of the class 25 registration. Symbolic I, 116 U.S.P.Q.2d at 1410; Symbolic II, 2015 WL 6746544, at *4; Symbolic III, 2015 WL 6746545, at *4.
The Board then addressed the DuPont Factors. See In re E. I. duPont deNemours & Co., 476 F.2d 1357 (CCPA 1973). The Board found that the legal or literal identity of the marks; the similarity or the identity of the goods; and the identity of the trade channels and purchasers, as well as the conditions of sale, favored a likelihood of confusion conclusion. Symbolic I, 116 U.S.P.Q.2d at 1411–12; Symbolic II, 2015 WL 6746544, at *5–8; Symbolic III, 2015 WL 6746545, at *5–7. The Board also found that the “purported lack of fame” of registrants' marks was of “little consequence” and noted that “[t]o the extent that Mr. Adams and [Symbolic's] mark are well-known, such fact supports refusal of [Symbolic's] application, because when confusion is likely, it is the prior Registrant which must prevail.” Symbolic I, 116 U.S.P.Q.2d at 1412 & n.7; Symbolic II, 2015 WL 6746544, at *8 & n.9; Symbolic III, 2015 WL 6746545, at *8 & n.10.
For the class 14 application, the Board rejected Symbolic's argument that “the coexistence of the [Beeline and Finch Registrations] on the register is evidence that there is no likelihood of confusion between each of these marks and [Symbolic's] mark” because “third-party registrations cannot assist an applicant in registering a mark that is likely to cause confusion with a previously registered mark.” Symbolic III, 2015 WL 6746545, at *8.
Symbolic timely appealed to this court. We have ju-risdiction pursuant to 28 U.S.C. § 1295(a)(4)(B).
We review the Board's legal conclusions without def-erence and its factual findings for substantial evidence. In re Pacer Tech., 338 F.3d 1348, 1349 (Fed. Cir. 2003). “Substantial evidence is ‘more than a mere scintilla’ and ‘such relevant evidence as a reasonable mind would accept as adequate’ to support a conclusion.” Id. (quoting Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)).
Likelihood of confusion is a question of law based on underlying findings of fact. In re Chatam Int'l Inc., 380 F.3d 1340, 1342 (Fed. Cir. 2004). We assess a likelihood of confusion based on the factors set forth in DuPont. 476 F.2d at 1361. “The likelihood of confusion analysis con-siders all DuPont factors for which there is record evidence but ‘may focus ․ on dispositive factors, such as similarity of the marks and relatedness of the goods.’ ” Herbko Int'l, Inc. v. Kappa Books, Inc., 308 F.3d 1156, 1164–65 (Fed. Cir. 2002) (quoting Han Beauty, Inc. v. Alberto–Culver Co., 236 F.3d 1333, 1336 (Fed. Cir. 2001) (alteration in original)). While evidence of actual confusion may be considered in the DuPont analysis, “a showing of actual confusion is not necessary to establish a likelihood of confusion.” Id. (citing Giant Food, Inc. v. Nation's Foodservice, Inc., 710 F.2d 1565, 1571 (Fed. Cir. 1983)). In the likelihood of confusion analysis “doubts are to be resolved against the newcomer and in favor of the prior user.” San Fernando Elec. Mfg. Co. v. JFD Elecs. Components Corp., 565 F.2d 683, 684 (CCPA 1977).
Symbolic argues that the Board erred in its likelihood of confusion analysis by: (1) holding that the will.i.am restriction is “precatory” and “meaningless” and therefore not considering it in analyzing certain DuPont factors; (2) ignoring third-party use and the peaceful coexistence on the primary and supplemental registers and in the marketplace of other I AM marks; and (3) finding a likelihood of reverse confusion. We will address each argument in turn.
I. The will.i.am Restriction
Symbolic argues that the Board erred by ignoring the will.i.am restriction because it is a meaningful limitation that negates any likelihood of confusion. Symbolic contends that the restriction is a “subject matter limitation,” Appellant's Br. 24 (citing DuPont, 476 F2d. at 1362; M2 Software, Inc. v. M2 Commc'ns, Inc., 450 F.3d 1378, 1382 (Fed. Cir. 2006)), that distinguishes the mark from registrants' marks by “impact[ing] the connotation and commercial impression of [the] mark, the nature of the goods identified in the [a]pplications, and the channels of trade in which said goods move,” id. at 22.
The PTO responds that the Board correctly held that the will.i.am restriction imposes no meaningful limitation for purposes of likelihood of confusion analysis and substantial evidence supports the Board's factual findings on the relevant DuPont factors.
We agree with the PTO that Symbolic has failed to show reversible error in the Board's determination that the will.i.am restriction does not impose a meaningful limitation in this case for purposes of likelihood of confusion analysis. It is well established that the Board may not read limitations into an unrestricted registration or application. See SquirtCo v. Tomy Corp., 697 F.2d 1038, 1043 (Fed. Cir. 1983) (“There is no specific limitation here, and nothing in the inherent nature of SquirtCo's mark or goods that restricts the usage of SQUIRT for balloons to promotion of soft drinks. The board, thus, improperly read limitations into the registration.”); see also 15 U.S.C. § 1057(b) (“A certificate of registration of a mark upon the principal register ․ [is] prima facie evidence ․ of the owner's exclusive right to use the registered mark in commerce on or in connection with the goods or services specified in the certificate, subject to any conditions or limitations stated in the certificate.” (emphases added)). However, where both the applicant's and the registrant's identifications of goods recite limitations, those limitations may be considered in analyzing the DuPont factors. See M2 Software, 450 F.3d at 1382–83. Here, the registrations do not contain an express limitation, and, for the reason discussed in more detail below with respect to the DuPont factors, the will.i.am restriction does not distinguish the mark sufficiently from the registrants' marks to overcome the evidence of likelihood of confusion.
A. The First DuPont Factor
The first DuPont factor requires consideration of “[t]he similarity or dissimilarity of the marks in their entireties as to appearance, sound, connotation and commercial impression.” DuPont, 476 F.2d at 1361. “The proper test is not a side-by-side comparison of the marks, but instead whether the marks are sufficiently similar in terms of their commercial impression such that persons who encounter the marks would be likely to assume a connection between the parties.” Coach Servs. Inc. v. Triumph Learning LLC, 668 F.3d 1356, 1368 (Fed. Cir. 2012) (internal quotation omitted). We compare the applicant's and registrant's “marks themselves.” Denney v. Elizabeth Arden Sales Corp., 263 F.2d 347, 348 (CCPA 1959).
Symbolic argues that the will.i.am restriction impacts the meaning and the overall commercial impression of the mark. According to Symbolic, “consumers will recognize the ‘i.am’ portion of will.i.am's name” because its “goods must be associated with will.i.am” per the will.i.am restriction. Appellant's Br. 30. Symbolic contends that the Board's finding that Adams is not widely known or identified by the term i.am or I AM is not supported by substantial evidence. Symbolic maintains that the Board erred by not considering extrinsic evidence of the connotations of the registrants' marks.
The PTO responds that the Board did not err in determining that the first DuPont factor weighs heavily in favor of a likelihood of confusion conclusion. The PTO contends that the Board's finding that Adams is not identified by i.am or I AM is supported by substantial evidence. The PTO argues that extrinsic evidence concerning the actual uses of the marks cannot be considered because the restrictions are not reflected in the four corners of the application and registrations.
We agree with the PTO that the Board correctly determined that the first DuPont factor “weighs heavily” in favor of a likelihood of confusion. See In re Majestic Distilling Co., Inc., 315 F.3d 1311, 1315 (Fed. Cir. 2003) (explaining that “when word marks are identical but neither suggestive nor descriptive of the goods associated with them, the first DuPont factor weighs heavily against the applicant”). The Board's findings that the mark and the registrants' marks are at least legally identical in appearance, identical in pronunciation, have the same meaning, and engender the same overall commercial impression are supported by substantial evidence.
Symbolic does not, and cannot, dispute that the mark, I AM in standard character form, and the registrants' marks, I AM in standard character, typed, or stylized form, are pronounced the same way and, at a minimum, legally identical. See In re Viterra Inc., 671 F.3d 1358, 1364 n.2 (Fed. Cir. 2012) (“ ‘[S]tandard character’ marks formerly were known as ‘typed’ marks, but the preferred nomenclature was changed in 2003 to conform to the Madrid Protocol. ․ [W]e do not see anything in the 2003 amendments that substantively alters our interpretation of the scope of such marks.”); SquirtCo, 697 F.2d at 1041 (“[T]he argument concerning a difference in type style is not viable where one party asserts rights in no particular display. By presenting its mark merely in a typed drawing, a difference cannot legally be asserted by that party.”).
The Board did not err in holding that the will.i.am restriction does not change the meaning or the overall commercial impression of the mark. The identification of goods does not specify how Adams will be “associated with” the goods. And we discern no error with the Board's determination that the will.i.am restriction “does not even represent that Mr. Adams will be named, or otherwise identified, in the promotion of the goods.” Symbolic I, 116 U.S.P.Q.2d at 1410. In short, there is nothing in the record that persuades us that the will.i.am restriction changes the meaning or overall commercial impression of the mark.
To the extent that Symbolic is advocating that we consider another mark, will.i.am, that is not part of the applied-for mark in analyzing the similarity of the marks, we decline to do so. The correct inquiry requires comparison of the applied-for mark, which only includes the words “I AM,” to the registrants' marks. See, e.g., Denney, 263 F.2d at 348 (“In determining the applicant's right to registration, only the mark as set forth in the application may be considered; whether or not the mark is used with an associated house mark is not controlling. Therefore the likelihood of confusion must be determined by a comparison of the [applicant's and registrant's] marks themselves.” (internal citations omitted)); In re Microsoft Corp., 68 U.S.P.Q.2d 1195, 1198 (T.T.A.B. Sept. 11, 2003) (“To the extent applicant's argument reflects anticipated use of the applied-for mark with applicant's house mark, it is well-settled that use of a house mark in conjunction with a product mark will not serve to prevent a finding of likelihood of confusion when the house mark is not included in the mark for which registration is sought.”).
Similarly, we discern no error in the Board's treat-ment of Symbolic's proffered examples of use of the registrants' marks. None of the examples of use of the registrants' marks pointed to by Symbolic reflect an actual limitation in the registrations. The Board properly analyzed likelihood of confusion based on the mark as applied to the goods recited in Symbolic's application compared to registrants' marks and the goods recited in their registrations. See In re Dixie Restaurants, Inc., 105 F.3d 1405, 1407–08 (Fed. Cir. 1997); see also In re St. Helena Hosp., 774 F.3d 747, 751 (Fed. Cir. 2014) (affirm-ing Board's determination that applicant's and registrant's marks have similar connotations where although “the [use] specimens refer to days and minutes, respectively, neither identification specifies a certain period of time or suggests any specific meaning of the word ‘TEN’ or the numeral ‘10’ ” present in the applicant's and registrant's marks).
Moreover, the Board's finding that Adams is not known by I AM or i.am is supported by substantial evidence. The websites and media coverage pointed to by Symbolic consistently show that Adams is known as will.i.am, not I AM or i.am. See, e.g., J.A. 116, 123, 2011–12, 2107, 2203, 2214. Additionally, Symbolic's ownership of the class 25 registration and use of that mark in pro-moting and selling Adams's clothing line does not compel a finding that Adams is known by I AM or i.am. Indeed, Adams is identified as will.i.am, not I AM or i.am, in that context too. See, e.g., J.A. 2024 (“will.i.am of the Black Eyed Peas launches a creative and unique men's wear collection under the i.am clothing label ․”).
Furthermore, Symbolic's reliance on Coach is misplaced. Although it is true that word marks identical in sound and appearance may have different meanings and commercial impressions, 668 F.3d at 1369, the Board's finding that this is not such a case is supported by substantial evidence. Unlike in Coach, Symbolic did not establish that the words “I AM,” which comprise the totality of the mark, by themselves “ha[ve] many different definitions in different contexts,” or that I AM when applied to Symbolic's goods “brings to mind” something different from I AM when applied to registrants' marks. Id.
B. The Second, Third, and Fourth DuPont Factors
Under the second, third, and fourth DuPont factors, one must consider “[t]he similarity or dissimilarity and nature of the goods or services as described in an application or registration or in connection with which a prior mark is in use”; the similarity or dissimilarity of the trade channels; and “[t]he conditions under which and buyers to whom sales are made, i.e., ‘impulse’ vs. careful, sophisti-cated purchasing.” DuPont, 476 F.2d at 1361. In review-ing the second factor, “we consider the applicant's goods as set forth in its application, and the [registrant's] goods as set forth in its registration.” M2 Software, 450 F.3d at 1382. Likelihood of confusion “must be resolved on the basis of the goods named in the registration and, in the absence of specific limitations in the registration, on the basis of all normal and usual channels of trade and meth-ods of distribution.” SquirtCo, 697 F.2d at 1042–43.
Symbolic argues that the Board erred by finding that its goods and registrants' goods are the same or related and move in the same or similar trade channels because the will.i.am restriction distinguishes its goods and channels of trade by fundamentally changing them. According to Symbolic, consumers purchase goods associated with celebrities, like will.i.am, “because the goods are associated with the celebrity and to associate themselves with the celebrity” and “goods associated with celebrities constitute extensions of the celebrities' personas and brands.” Appellant's Br. 39. Symbolic further argues that the registrants' marks are prohibited from drawing a connection or association with will.i.am, as the mark does, because it would violate Adams's right to publicity under various state laws and infringe his rights in his identity and persona under 15 U.S.C. § 1125(a). Symbolic also contends that the PTO's acceptance of the will.i.am. restriction in the applications and the PTO's registration of marks containing similar restrictions that “describe goods that are distinguishable for being related to or associated with a musician or entertainer” support the registrability of the mark. Id. at 26.
The PTO responds that substantial evidence supports the Board's findings that the goods are identical or closely related, the trade channels are identical, and the classes of purchasers overlap. According to the PTO, the will.i.am restriction does not alter the intrinsic nature of the goods or trade channels. Even if the will.i.am restriction were a meaningful limitation to the goods, the PTO contends, Symbolic, a later applicant, cannot avoid a likelihood of confusion with unrestricted registrations, like the registrants' marks, merely by restricting the goods in its application. The PTO further responds that its acceptance of the will.i.am restriction in the identification of goods and prior acceptance of similar wording in other registrations does not make the will.i.am restriction significant for likelihood of confusion purposes.
We agree with the PTO that substantial evidence supports the Board's findings that the goods are identical or related and the channels of trade are identical. Symbolic does not dispute that the goods would be related and the channels of trade would be identical without the will.i.am restriction. Symbolic, however, points to no evidence in the record to support its contention that the will.i.am restriction changes the nature of the goods or trade channels. The Board thus did not err in holding that the will.i.am restriction does not: (1) limit the goods “with respect to either trade channels or class of purchasers”; (2) “alter the nature of the goods identified”; or (3) “represent that the goods will be marketed in any particular, limited way, through any particular, limited trade channels, or to any particular class of customers.” Symbolic I, 116 U.S.P.Q.2d at 1410; Symbolic II, 2015 WL 6746544, at *4; Symbolic III, 2015 WL 6746545, at *4.
Symbolic's reliance on M2 Software is thus misplaced. In M2 Software, we affirmed the Board's determination that the goods were not related and the channels of trade and purchasers were different where the identification of goods for both the registrant's and applicant's marks were limited to “CD-ROMs produced for a particular field[,]” “music or entertainment” and “pharmacy and medicine,” respectively. M2 Software, 450 F.3d at 1382.
Here, substantial evidence supports the Board's findings that the will.i.am restriction does not impose a meaningful limitation and the registrations at issue do not contain any express limitations.6 Thus, unlike in M2 Software, the application and registrations here do not contain meaningful limitations in the identification of goods. As a result, the Board properly declined to import restrictions into the identification of goods based on alleged real-world conditions. See Stone Lion Capital Partners, L.P. v. Lion Capital LLP, 746 F.3d 1317, 1323 (Fed. Cir. 2014) (“It was proper, however, for the Board to focus on the application and registrations rather than on real-world conditions, because ‘the question of registrability of an applicant's mark must be decided on the basis of the identification of goods set forth in the application.’ ” (quoting Octocom Sys., Inc. v. Hous. Comp. Servs. Inc., 918 F.2d 937, 942 (Fed. Cir. 1990))); Octocom, 918 F.2d at 942 (“The authority is legion that the question of registrability of an applicant's mark must be decided on the basis of the identification of goods set forth in the application regardless of what the record may reveal as to the particular nature of an applicant's goods, the particular channels of trade or the class of purchasers to which sales of the goods are directed.” (collecting cases)).
Additionally, the PTO's mere acceptance of the will.i.am restriction in the identification of goods does not render it meaningful to likelihood of confusion. Contrary to Symbolic's suggestion, not all language in an identification of goods that is “specific, definite, clear, accurate, and concise,” Appellant's Br. 25 (quoting TMEP § 1402.01), imposes a meaningful limitation for purposes of likelihood of confusion analysis. Similarly, the PTO's alleged allow-ance of similar limitations in other registrations does not dictate the outcome in this case. See, e.g., Juice Generation, Inc. v. GS Enters. LLC, 794 F.3d 1334, 1338 (Fed. Cir. 2015) (explaining that “[l]ikelihood of confusion must be analyzed on a case-by-case basis”).
In the absence of meaningful limitations in either the application or the cited registrations, the Board properly presumed that the goods travel through all usual channels of trade and are offered to all normal potential purchasers. See Viterra, 671 F.3d at 1362 (“[I]t is well established that, ‘absent restrictions in the application and registration, goods and services are presumed to travel in the same channels of trade to the same class of purchasers.’ ” (quoting Hewlett–Packard Co. v. Packard Press, Inc., 281 F.3d 1261, 1268 (Fed. Cir. 2002))). Indeed, Symbolic does not contest the Board's findings with respect to potential purchasers. In sum, the Board's findings that the second, third, and fourth DuPont factors weigh in favor of a likelihood of confusion conclusion are supported by substantial evidence.
II. Other Registered I AM Marks
Under the sixth, eighth, and twelfth DuPont factors, we must consider “[t]he number and nature of similar marks in use on similar goods”; “[t]he length of time during and conditions under which there has been con-current use without evidence of actual confusion”; and “[t]he extent of potential confusion, i.e., whether de mini-mis or substantial.” DuPont, 476 F.2d at 1361. We have explained in an opposition context that “evidence of third-party use bears on the strength or weakness of an opposer's mark.” Juice Generation, 794 F.3d at 1338. “The weaker an opposer's mark, the closer an applicant's mark can come without causing a likelihood of confusion and thereby invading what amounts to its comparatively narrower range of protection.” Id.
Symbolic argues that the Board erred by failing to consider the sixth, eighth, and twelfth DuPont factors in light of our recent precedent. Appellant's Br. 41–42 (citing Jack Wolfskin Ausrustung Fur Draussen GmbH & Co. KGAA v. New Millennium Sports, S.L.U., 797 F.3d 1363, 1373–74 (Fed. Cir. 2015), cert. denied, 136 S. Ct. 982 (2016); Juice Generation, 794 F.3d at 1338). According to Symbolic, the Board was obligated to consider third-party use and the peaceful coexistence on the Regis-ters and in the marketplace of Symbolic's class 25 registration with the registrants' marks for at least ten years. Symbolic also contends that the Board should have considered the fact that the Beeline, St. Germain,7 and Finch Registrations have coexisted on the registers and in the marketplace for years, despite all offering bracelets or jewelry.
The PTO responds that Jack Wolfskin and Juice Generation are “inapposite based on the record of this case.” Appellee's Br. 43. Here, unlike in Jack Wolfskin and Juice Generation, the PTO contends “there is no evidence of extensive or voluminous third-party registrations or use of I AM marks co-existing for the same or similar goods.” Id. at 44. The PTO further argues that those cases did not change “well-settled precedent” “that the coexistence of similar marks in third-party registrations does not justify registering a mark that is likely to cause confusion with previously registered marks.” Id. (citing AMF Inc. v. Am. Leisure Prods., Inc., 474 F.2d 1403, 1406 (CCPA 1973)).
We agree with the PTO that Jack Wolfskin and Juice Generation are distinguishable. Both cases involved oppositions to the registration of applied-for composite marks that were not identical to the opposers' marks. The applied-for marks included a component, a paw print, Jack Wolfskin, and “peace” and “love” with a product-identifying term, Juice Generation, used “ubiquitous[ly],” Jack Wolfskin, 797 F.3d at 1374, or in a “considerable number of,” Juice Generation, 794 F.3d at 1339, third-party marks for the same or similar goods. Based on the evidence showing the frequency of use of paw prints on clothing and peace and love in connection with restaurant services or food products, we concluded that the Board gave inadequate consideration to the actual or potential weakness of registrants' marks in the likelihood of confusion analysis. Jack Wolfskin, 797 F.3d at 1374 (explaining that “the considerable evidence of third-party registration and usage of marks in commerce that depict paw prints on clothing ․ indicates that the paw print portion of [opposer's] mark is relatively weak”); Juice Generation, 794 F.3d at 1340 (“[W]e conclude that the Board gave inadequate consideration to the strength or weakness of [opposer's] marks.”).
The factual scenario here, involving identical word marks, stands in sharp contrast. As an initial matter, Symbolic has neither introduced evidence, nor provided adequate explanation to support a determination that the existence of I AM marks for goods in other classes, e.g., its class 25 registration for clothing, support a finding that registrants' marks are weak with respect to the goods identified in their registrations. Symbolic's ownership of the class 25 registration does not give it the “right to register [the same] mark on an expanded line of goods [in the applications at issue] where the use of the mark covered by such registration would lead to a likelihood of confusion, mistake or deception.” Jackes-Evans Mfg. Co. v. Jaybee Mfg. Corp., 481 F.2d 1342, 1345 (CCPA 1973).
Furthermore, Symbolic's evidence of third-party use of I AM for the same or similar goods falls short of the “ubiquitous” or “considerable” use of the mark compo-nents present in its cited cases. See Jack Wolfskin, 797 F.3d at 1374; Juice Generation, 794 F.3d at 1339; see also id. at 1337 n.1 (listing 26 third-party marks). For classes 3 and 9, Symbolic has only pointed to the mark over which the Board concluded that there was a likelihood of confusion for the same or similar goods. In other words, Symbolic has not pointed to any record evidence to support a finding that multiple third parties use the mark I AM for the listed goods in its class 3 and 9 applications.
For class 14, Symbolic points to three I AM marks al-legedly coexisting without confusion for the same or similar goods. Thus, vis-à -vis each of the two registrations relied on by the Board in its refusal, Symbolic points to two other third-party registrations. Assuming arguen-do that the existence of two third-party registrations could demonstrate the weakness of a mark, “the Board may focus its analysis on dispositive factors, such as similarity of the marks and relatedness of the goods.” Han Beauty, 236 F.3d at 1336. “Indeed, any one of the factors may control a particular case.” Dixie Restaurants, 105 F.3d at 1407. As we discussed supra, substantial evidence supports the Board's findings that the mark and registrants' marks are identical word marks for the same or related goods with identical trade channels and purchasers. Given the strength of these DuPont factors and the limited evidence of third-party use, any error in failing to specifically analyze the potential weakness of registrants' marks based on such limited third-party use was harmless error. See Opryland USA Inc. v. Great Am. Music Show, Inc., 970 F.2d 847, 850 (Fed. Cir. 1992) (“Not all of the duPont factors are relevant or of similar weight in every case.”).
Accordingly, we find no reversible error with respect to the sixth, eighth, and twelfth DuPont factors.
III. Likelihood of Reverse Confusion
“The term ‘reverse confusion’ has been used to describe the situation where a significantly larger or promi-nent newcomer ‘saturates the market’ with a trademark that is confusingly similar to that of a smaller, senior registrant for related goods or services.” In re Shell Oil Co., 992 F.2d 1204, 1208 (Fed. Cir. 1993) (quoting Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947, 957 & n.12 (7th Cir. 1992)). In that situation, “[t]he junior user does not seek to benefit from the goodwill of the senior user; however, the senior user may experience diminution or even loss of its mark's identity and goodwill due to extensive use of a confusingly similar mark by the junior user.” Id. Trademark law “protects the registrant and senior user from adverse commercial impact due to use of a similar mark by a newcomer.” Id.; see also Wall-paper Mfrs., Ltd. v. Crown Wallcovering Corp., 680 F.2d 755, 762 (CCPA 1982) (explaining “even where there is reverse confusion ․ another source with superior de jure rights may prevail regardless of what source or sources the public identifies with the mark”).
Symbolic argues that the Board found a likelihood of reverse confusion and that that finding is not supported by substantial evidence. The PTO responds that the Board did not find a likelihood of reverse confusion and that its finding that the fame of the marks, the fifth DuPont factor, is neutral is supported by substantial evidence.
We agree with the PTO that the Board did not find a likelihood of reverse confusion. The Board treated the fifth DuPont factor as “neutral” and explained that the “purported lack of fame” of registrants' marks was of “little consequence.” Symbolic I, 116 U.S.P.Q.2d at 1412; Symbolic II, 2015 WL 6746544, at *8; Symbolic III, 2015 WL 6746545, at *8. That determination is consistent with our precedent holding that the purported lack of fame of a registrant's mark has “little probative value” in the likelihood of confusion analysis. Majestic Distilling, 315 F.3d at 1317.
Symbolic points to a footnote in the Board's opinions in support of its contention that the Board found a likelihood of reverse confusion. Therein the Board noted that “[t]o the extent that Mr. Adams and [Symbolic's] mark are well-known, such fact supports refusal of [Symbolic's] application, because when confusion is likely, it is the prior Registrant which must prevail.” Symbolic I, 116 U.S.P.Q.2d at 1412 n.7 (emphasis added); Symbolic II, 2015 WL 6746544, at *8 n.9 (emphasis added); Symbolic III, 2015 WL 6746545, at *8 n.10 (emphasis added). The Board made no factual findings with respect to the fame of the mark.
When viewed in the context of the Board's opinion as a whole, the cited footnote appears to be the Board's response to Symbolic's arguments regarding the alleged fame of Adams and the mark. The footnote merely ex-plains that to the extent the Board agreed with Symbolic that Adams or the mark are famous, such a finding would not support registration of the mark. Therefore, the Board's statements do not amount to a finding of reverse confusion.
We have considered Symbolic's remaining arguments, but conclude that they are without merit. The Board's factual findings were supported by substantial evidence and its legal conclusions were not erroneous as a matter of law. For the reasons set forth above, we affirm the Board's decision.
1. Adams assigned the applications to Symbolic while they were pending at the PTO. For simplicity, we refer to the owner of the applications throughout as Symbolic.
2. Registration No. 2,045,626 owned by Danica Siegel.
3. The examining attorney's final refusal for class 9 was limited to the goods “sunglasses and sunglass cases associated with William Adams, professionally known as will.i.am.” The application with the remaining goods was allowed to proceed to publication. Symbolic II, 2015 WL 6746544, at *8 n.4. Because the goods in the application allowed to proceed to publication are not a part of this appeal, we do not address them.
4. Registration No. 3,188,447 owned by Beeline GmbH.
5. Registration No. 3,935,952 on the Supplemental Register owned by Justin Finch.
6. Because we affirm the Board's determination that the will.i.am restriction imposes no meaningful limitation in this case, we need not, and do not, decide whether other legal constraints, e.g., 15 U.S.C. § 1125(a), may ever inherently restrict an otherwise unrestricted identification of goods in a registration in a manner meaningful to the likelihood of confusion analysis.
7. Registration No. 1,357,947 for a composite mark including I AM and a design for “jewelry” owned by Saint Germain Foundation.
LOURIE, Circuit Judge.