July 1, 2020
Brian Winterfeldt and David Rome

U.S. Supreme Court Rules on Generic Terms in Booking.com Decision

External Article

This article is hosted away from the Winterfeldt website at

The Supreme Court of the United States held yesterday that a generic term combined with a generic top level domain name is not necessarily generic; such a designation can be registered as a trademark, provided that consumers perceive it as indicating a source of goods or services.

The decision in United States Patent and Trademark Office v. Booking.com B.V. No. 19-46, rejected the Office’s proposed rule that so-called “generic.com” designations be deemed generic almost per se and without regard to evidence of their significance to consumers. While Justice Ginsburg’s majority opinion is particularly well suited to a future where new generic top level domains are more widely adopted and become more numerous, the decision will also likely create problems in the short term. The decision creates ambiguity regarding the nature and amount of evidence that ostensible trademark owners must provide to show distinctiveness of a “generic.com” mark or any arguably generic mark. This uncertainty will burden brand owners, the courts and the Office.

Refining the Standard for Genericness for Domain Names and in General

The majority opinion in Booking.com establishes that the genericness of a given designation must be determined based on its significance to consumers. The Office had argued that genericness could be determined as a matter of law without regard to consumers, even if consumers understood a mark to be distinctive, if the designation is a “Generic.com.” It based this conclusion on the Supreme Court’s 1888 decision in Goodyear’s India Rubber Globe Mfg. Co. v. Goodyear Rubber Co., 128 U.S. 598, which held generally that the addition of a generic corporate designation, like “Inc.” or “Corporation,” to an otherwise generic term does not supply any distinctive significance and therefore does not make the combination of the two any less generic. The majority disagreed, noting that Congress had since enacted the Lanham Act in 1946, well after the Supreme Court’s decision in Goodyear. Section 14(3) of the Act provides that findings of genericness must be based on the “primary significance of the … mark to the relevant public.” The Office argued that Section 14(3) only relates to cancellation of registrations for generic designations, but the Court maintained that “significance to the relevant public” is a necessary component to findings of genericness generally, stating: “whether a term is generic depends on its meaning to consumers. That bedrock principle of the Lanham Act is incompatible with an unyielding legal rule that entirely disregards consumer perception.” The Court pointed out that the holding in Goodyear was not inconsistent with this understanding of trademark law, even though its reasoning was not framed in terms of the designation’s impression on consumers.

Accepting that Goodyear is still good law, the Court in Booking.com found that generic top-level domains are not analogous to generic corporate designations. “Generic.com” designations are more likely to have a source-identifying characteristic because domain names are exclusively controlled by their owners. If that owner does business through that domain name, the domain name can be associated with that single source of goods or services.

This decision still sets a very low bar that the Office must meet in refusing registration for “Generic.com” designations. The majority opinion emphasized that “[w]hile we reject the rule proffered by the [Office] that ‘generic.com’ terms are generic names, we do not embrace a rule automatically classifying such terms as nongeneric.” In other words, the Office cannot merely conclude that a designation is generic because it is a “Generic.com.” It must go further to show that the designation has a generic meaning to consumers, just as it does ordinarily for any combination of generic terms. Justice Sotomayor’s concurring opinion points out that the Office may very well have met its burden of proof in that regard as to BOOKING.COM and that the courts below may have erred in finding to the contrary, but that particular issue was not raised on appeal and was not before the Supreme Court.

The Booking.Com Decision and New gTLDs

The Supreme Court’s decision in Booking.Com is likely to apply well to the wider adoption of so-called new generic top level domain names. The Office’s proposed rule that the addition of a generic top level domain name could never render an otherwise generic designation distinctive based on Goodyear makes far more sense when considering the group of “legacy” generic top level domain names that are still most familiar to consumers in the United States. These originally denoted particular types of entities: “.com” for commercial, “.org” for non-commercial, “.net” for network-infrastructure-related, “.edu” for educational, “.gov” for governmental and “.mil” for military. These short strings of characters were arguably “generic” in a trademark sense because they indicate the genus or category of services with which they were used. They also functioned very similarly to generic corporate designations such as “Inc.” for corporations, “Partners” for partnerships, “Limited” for limited companies and the like. However, as the internet has grown and especially since ICANN began accepting applications to delegate new gTLDs in 2012, the number of generic top level domain names has expanded to more than twelve hundred. Some of these, like “.xyz,” have no inherent significance, whereas others have a plain meaning in English, but may be used in connection with an enterprise for which that meaning would be essentially arbitrary. Even while arguing in favor of the Office’s proposed rule, Justice Breyer’s dissent admitted that it “may apply differently to the newly expanded universe of top-level domains ….” Instead of adopting a rule that would already require considerable modifications in light of already extant business realities, the Supreme Court has instead adopted a more flexible and fact sensitive view that can embrace evolving ways in which businesses, consumers and other internet users are utilizing the domain name system.

The impact of brand owners wielding exclusive trademark rights in “Generic.com” domain names is also greatly diminished by the increasing availability and acceptance of new generic top level domain names. There are now more ways for a generic term to be reflected in the domain name system than ever before. This decision may even help invigorate the still gradual transition from the ubiquity of “.com” top level domain and interest in a second round of new generic top level domain name applications. The ability to potentially own rights in both a generic term in the second level of those domains and a corresponding trademark for the entire string might also increase the value placed on new “generic.gtld” strings.

Reconsidering the Applicability of Surveys to Questions of Genericness

It is painfully unclear how the Booking.com decision will affect evidence showing distinctiveness in more substantial disputes concerning genericness, beyond even ex parte proceedings before the Office. The decision below in the Eastern District of Virginia had found that BOOKING.COM was distinctive based on a “Teflon survey.” That type of survey is appropriately used to determine whether a coined or fanciful trademark, like TEFLON, has suffered “genericide,” or has become synonymous with the class or category of goods or services with which it is used. As such, the survey was designed in a way that already assumed that the designation was inherently distinctive. However, the issue in Booking.com was whether the designation was inherently distinctive. Neither the Office nor the Court of Appeals for the Federal Circuit, which ordinarily reviews the Office’s decisions, will accept this form of evidence as probative. There are a myriad problems with using surveys to differentiate between generic and descriptive terms, as pointed out by Justice Breyer in his dissent:

Consumer surveys often test whether consumers associate a term with a single source. But it is possible for a generic term to achieve such an association—either because that producer has enjoyed a period of exclusivity in the marketplace … or because it has invested money and effort in securing the public’s identification …. Evidence of such an association, no matter how strong, does not negate the generic nature of the term. For that reason, some courts and the TTAB have concluded that survey evidence is generally of little value in separating generic from descriptive terms.

(citations omitted). The majority insisted that it had “no authority to reweigh” the survey evidence presented at trial when the issue had not been raised on appeal. Without considering the particular survey used, however, the majority was willing to offer, generally, that “[s]urveys can be helpful evidence of consumer perception but require care in their design and interpretation.” Justice Sotomayor’s concurring opinion offered the further caveat that “[f]laws in a specific survey design, or weaknesses inherent in consumer surveys generally, may limit the probative value of surveys in determining whether a particular mark is descriptive or generic in this context.” The upshot seems to be that it is possible that some kind of survey could be probative in cases like these, but there is no guidance on how it should be designed or interpreted. There seems to be no clear call for the Office or the lower courts to begin accepting forms of surveys that they already deem irrelevant. Even if they do, the Supreme Court’s decision in Booking.com does not seem to require that the surveys be accorded so much weight that they would substantially impact the eventual outcome.

Rebalancing the Anticompetitive Impact of Trademark Registration

The Supreme Court offered an important rebuke to the Office’s arguments that registration should be denied because it would unduly impair competition. The Office routinely fashions itself and its power to deny registration as the bulwark against brand owners that would otherwise deny competitors the ability to describe their goods or services. This ignores the legal reality that a trademark registration is neither necessary nor sufficient for a party to wrongfully assert ostensible trademark rights and the courts, not just the Office, bear some of the responsibility for preserving competition. The Supreme Court observed that “trademark law hems in the scope of [descriptive] marks short of denying trademark protection altogether.” Courts already recognize that less distinctive marks have a diminished scope of protection, that marks  existing in a crowded field are less likely to be confused by consumers, and that there is a defense for fair and good faith use of terms that accurately describe goods and services. Registration of a “Generic.com” would not necessarily result in a monopoly on its generic component, nor would it necessarily create enforceable rights against slight variations on the “Generic.com” domain name. In response to the dissent’s rhetorical question of “why would a firm want to register its domain name as a trademark unless it wished to extend its area of exclusivity beyond the domain name itself,” the majority easily supplied the obvious answer: to more effectively enforce the exclusivity of the domain name itself with the enhanced remedies afforded by the Lanham Act. Referencing the limitations on the agency’s power conferred by the statute , the Supreme Court concluded that “[w]e have no cause to deny Booking.com the same benefits Congress accorded to other marks qualifying as nongeneric.”

Functionality as a Bar to Registration

The Office may yet find alternative grounds to refuse registration for otherwise “Generic.com” designations based on a theory that they are “as a whole … functional” within the meaning of Section 2(e)(5) of the Lanham Act. This may capture some of the intuitions reflected in Justice Breyer’s dissent contending that a top-level domain name is “merely a necessary component of a web address.” Although the Office presented arguments based on this theory in its brief, the Supreme Court declined to consider it because the application at issue had not been refused as functional. It will be interesting to see whether the Office develops this theory in the future, as an alternative to refusals of domain names for genericness.

If you have any questions regarding this alert or wish to discuss these matters in more detail, please contact any of the following Winterfeldt IP Group team members:

Brian Winterfeldt, brian@winterfeldt.law, +1 202 903 4422

David Rome, david@winterfeldt.law, +1 847 757 3790

download this document as a PDF