A number of recent UDRP decisions remind trademark owners (and counsel) that cybersquatting cases have to be built from the ground up. Each stage has its evidentiary demands. The first two demand either/or proof; the third, the most demanding, requires proof of unified or conjunctive bad faith registration and bad faith use of the accused domain name. Priority, which intuitively would be thought a factor under the first stage (as it is under the ACPA) is actually a factor under the third stage. Entertainment Technology Investments, Inc. d/b/a Gloo, LLC v. Contact Privacy Inc. Customer 011945202, D2017-0606 (WIPO May31, 2017) (<gloo.com>); Technologies Sensopia Inc. v. BLUE NOVA INC, FA1704001725217 (Forum June 9, 2017) (<magicplan.com>). For the third stage, any proof less than both is insufficient to establish infringement. Charles A. Saunders / Saunders Archery Company v. Lisa Katz / Domain Protection LLC., FA1704001727959 (Forum May 31, 2017) (&llt;saundersarchery.com>); Gabs S.r.l. v. DOMAIN ADMINISTRATOR—NAME ADMINISTRATION INC. (BVI), CAC 101331 (ADReu February 26, 2017) (<gabs.com>).

Among country code-anticybersquatting policies including the U.S.‘s statutory scheme, the ACPA the UDRP’s insistence on conjunctive bad faith is sui generis. Under these other policies, bad faith (or abusive registration) can be established with evidence of either registration or use (add “trafficking in” under the ACPA). But for the first two stages of the UDRP complainants succeed by offering either/or proof of standing—the domain name is either identical to a mark in which complainant has a right or it is confusingly similar (paragraph 4(a)(i) of the Policy)—or respondents have neither rights nor legitimate interests in the accused domain names (paragraph 4(a)(ii)) (respondents have the shifted burden of rebutting complainant’s prima facie case by proving the either/or of rights or legitimate interests). I’m going to address only the first stage. It’s a low bar to prove standing; not a no-bar.

There’s no magic in understanding the term “identical”: if the characters of the SLD match the characters of the mark character for character, it is identical (all of the above domain names fit this description). If there’s no perfect match, but there’s some identity of characters, perhaps differently arranged or combining a dictionary word with the dominant element of the mark, then it’s most likely to be confusingly similar. Open Society Institute v. Gil Citro, FA1007001333304 (Forum April 24, 2010) (OPEN SOCIETY INSTITUTE and <opensociety.com>). The burden for proving standing is so low that Panels ordinarily find complainants satisfy the requirement, but this is not always the case. SportSoft Golf, Inc. v. Sites to Behold Ltd., FA 94976, (Nat. Arb. Forum July 27, 2000); Fabricators & Manufacturers Association, International v. Domain Administrator / Namefind, LLC, FA1704001728625 (Forum June 1, 2017) (disclosure: I was Respondent’s counsel in this matter).

There are SLDs that incorporate non-dominant elements of the mark, similarity alone is not sufficient to support standing. SportsSoft Golf is an early example of defining the boundary between similar and confusingly similar, similar but not confusingly so. In its decision, the Panel held that <golfsociety.com> (consisting as it does of two generic words) was not confusingly similar to GOLF SOCIETY OF THE US. The former clearly references to the generic nature of a society interested in golf or the general body or community of golfers while the mark was specific to a particular geographic entity, the United States.

The distinction between confusingly similar and similar but n0t confusing is further illuminated in B2BWorks, Inc. v. Venture Direct Worldwide, Inc., FA 97119 (Forum June 5, 2001) B2BWORKS and <b2badworks.com>, <b2bmediaworks.com>, <b2badswork.com>, and <b2badwork.com>). In this case, the Panel held that Complainant did not have exclusive rights to use of the terms ‘B2B’ and ‘Works’ in association with other words, even with a registered trademark for B2BWORKS).” It continued, “[f]urther, where marks are similar, very small differences become extremely important.” A later appellate court came to the same conclusion in Entrepreneur Media, Inc. v. Smith, 279 F.3d 1135, 1147 (9th Cir. 2002) (“Similarity of marks or lack thereof are context-specific concepts. In the Internet context, consumers are aware that domain names for different Web sites are quite often similar, because of the need for language economy, and that very small differences matter.”)

While the majority in Open Society found there was confusing similarity the third member of the Panel would have found otherwise (citing SportsSoft). He proposed an “objective bystander test”:

The test to be applied has two parts. The first part is to ask if the domain name is similar to the trademark. In this case, it is similar, for the domain name is made up of two of the three words of the trademark. The second part of the test is whether the similarity is confusingly so. This part of the test is satisfied by asking the further question if the Panel can conclude that an objective bystander, comparing the domain name and the trademark, would reasonably conclude that the “open society” of the domain name was referring to the OPEN SOCIETY INSTITUTE of the trademark and by that means giving rise to confusion between the two.

The real obstacle in the way of finding that an objective bystander would reach that conclusion is that the domain name is referring to a concept or a notion, whereas the trademark is referring to an entity or, more precisely, an institute and one institute in particular. An objective bystander is, therefore, unlikely to think that the domain name was referring to the Complainant’s institute or to any institute at all and unlikely to be confused into thinking that the one necessarily referred to the other. (Emphasis added).

SLDs merely similar are not actionable for cybersquatting any more than (under the third stage) are domain names identical to later acquired marks. Domain names can be similar (that is, share some grammatical element) without being confusingly similar. For example, in The Dow Chemical Company and E. I. du Pont de Nemours and Company v. Jung Chang Seap, D2016-0596 (WIPO July 13, 2016) (<duchemical.com>) the Panel held

DuPont’s rights to DUPONT cannot be reasonably expanded to cover uses of DU alone, a term that can also be viewed as a simple French article. Further, there is no evidence that would suggest that DUPONT is commonly abbreviated or referred to as DU alone.

WIPO Overview 2.0, Paragraph 1.2 states that

Some panels have additionally required that, for a domain name to be regarded as confusingly similar to the complainant’s trademark, there must be a risk that Internet users may actually believe there to be a real connection between the domain name and the complainant and/or its goods and services. Such panels would typically assess this risk having regard to such factors as the overall impression created by the domain name, the distinguishing value (if any) of any terms, letters or numbers in the domain name additional to the relied-upon mark, and whether an Internet user unfamiliar with any meaning of the disputed domain name seeking the complainant’s goods or services on the world wide web would necessarily comprehend such distinguishing value vis-à-vis the relevant mark.

WIPO Overview 3.0, Paragraph 1.7 (omits the statement in 2.0) and rephrases the view (not as illuminating I think as 2.0):

Issues such as the strength of the complainant’s mark or the respondent’s intent to provide its own legitimate offering of goods or services without trading off the complainant’s reputation, are decided under the second and third elements. Panels view the first element as a threshold test concerning a trademark owner’s standing to file a UDRP complaint, i.e., to ascertain whether there is a sufficient nexus to assess the principles captured in the second and third elements.

The “objective bystander” test is better represented in WIPO 2.0. Panels are conscious of the “small differences” discussed above. “Distinguishing values” are a better guide. This point is strongly emphasized in Fabricators & Manufacturers. The distinctive feature of THE FABRICATOR is the article and noun combined, not either of the words separately. The law does not sanction monopolizing dictionary words—an example among many is Halo Innovations, Inc. v. Name Administration Inc. (BVI), FA1009001344653 (Nat. Arb. Forum November 3, 2010) (SLEEPSACK and ).

The Panel in Fabricators & Manufacturers concluded that

it is well and long-established rationale and principle of trademark laws across the countries that the exclusivity of a trademark right does not extend to non-distinctive terms, words, or devices which are subject to the free use of the public.

It continued:

Therefore, the Panel agrees with Respondent’s contentions that Complainant does not have any right to monopolize the dictionary word “fabricator” alone because the mark is indivisible: neither the article “the” nor the noun “fabricator” is distinctive in its own right. The Panel agrees that Complainant’s mark is distinctive only in its indivisible wholeness. (Emphasis added).

Other one-element similarities include Barnesandnoble.com, LLC v. Rosenblum, FA 1089020 (Nat. Arb. Forum November 15, 2007) (<noble.com>) and National Grid Electricity Transmission Plc, NGrid Intellectual Property Limited v. Re Tron Technologies, D2013-0925 (WIPO July 11, 2013) (NATIONAL GRID and <mygridpower.com and <mygridstore.com>. The domain names “incorporate only the ‘grid’ element of Complainant’s trademarks. . . . The term ‘grid’ (on its own) is a generic, non-distinctive term, which is not associated exclusively with the Complainant’s business.”)

It is noteworthy that in most of these cases Respondents presented full arguments in defense. Panel (no less than judges) should be pointed in the right direction. I point this out because building a case from the bottom up for their counter-narratives is also something expected of respondents if they have cases to defend. But that’s another story!