An 8-1 Decision That Opened a Door Almost No One Can Walk Through
In June 2020, the Supreme Court decided the Booking.com trademark case, holding 8-1 that adding ".com" to a generic term does not automatically disqualify the resulting combination from trademark protection. USPTO v. Booking.com B.V., 591 U.S. 378 (2020). Justice Ginsburg, writing for the majority, rejected the USPTO's position that "generic.com" combinations are generic as a matter of law. Justice Breyer, alone in dissent, warned that the decision would allow companies to monopolize commercially useful language.
The ruling was legally significant. It eliminated a categorical rule that the USPTO had applied for years and replaced it with a fact-intensive inquiry into consumer perception. But its practical significance has been far narrower than the initial reaction suggested. Six years later, the door the Court opened remains one that almost no applicant can afford to walk through.
The Booking.com trademark case is frequently cited and frequently mischaracterized. What follows is an analysis of what the Court actually decided, what it left unresolved, and what the ruling means now, for IP professionals tracking the evolution of genericness doctrine (the legal framework for determining whether a term is too common for trademark protection) and for developers working with trademark data who encounter "generic.com" registrations and need to understand the thin protection they carry.
What Booking.com Actually Asked For
The facts are straightforward. Booking.com B.V., the Netherlands-based online travel company, filed four trademark applications with the USPTO for the mark BOOKING.COM. The applications covered online travel reservation services in International Classes 039 (transportation services) and 043 (hotel and temporary accommodation services). The company wanted federal registration for a term that combines a word everyone uses to describe reservation services with a top-level domain suffix.
The USPTO refused all four applications under Section 2(e)(1) of the Lanham Act, which bars registration of marks that are "merely descriptive" or "generic" for the applicant's goods or services. The examining attorney's reasoning was blunt: "booking" is generic for reservation services, and appending ".com" adds no distinguishing character. A generic term does not become registrable just because someone attaches a domain extension to it.
Booking.com appealed to the Trademark Trial and Appeal Board (the TTAB, the USPTO's internal adjudicative body that hears appeals from examining attorney decisions). The TTAB affirmed, applying what amounted to a per se rule: when a generic term is combined with a top-level domain like ".com," the resulting composite is necessarily generic. No amount of consumer perception evidence could overcome this categorical bar. The TTAB treated the question as one of linguistic composition rather than marketplace reality.
The Fourth Circuit reversed. The appellate court considered consumer survey evidence, specifically a Teflon survey (a survey format developed in connection with DuPont's TEFLON mark, designed to test whether consumers perceive a term as a brand name or a common word). The survey showed that 74.8% of consumers identified BOOKING.COM as a brand name rather than a generic term for hotel reservation services. The Fourth Circuit held that this evidence was relevant and sufficient: consumer perception, not a linguistic formula, determines whether a term functions as a trademark.
The Supreme Court granted certiorari to resolve whether the per se rule the USPTO advocated was correct. The question as framed was narrow: does combining a generic word with ".com" yield a generic composite as a matter of law? Understanding the trademark distinctiveness spectrum is essential background for following the doctrinal arguments that came next.
Per Se Rule or Consumer Perception: The Doctrinal Fault Line
The core legal question in USPTO v. Booking.com exposed a fault line in trademark law that had been papered over for more than a century. On one side: the principle that generic terms belong to the public and cannot be monopolized by any single producer. On the other: the principle that consumer perception is the ultimate test of whether a term functions as a trademark. These principles coexist easily in most cases. They collided in this one.
The USPTO's argument rested on Goodyear's India Rubber Glove Mfg. Co. v. Goodyear Rubber Co., an 1888 Supreme Court decision holding that adding a corporate designation like "Company" to a generic term does not create a protectable trademark. Goodyear Rubber Company could not monopolize "Goodyear Rubber Company" because the addition of "Company" was a generic corporate suffix that did nothing to distinguish the source. The USPTO argued that ".com" operates the same way: a generic internet suffix, functionally equivalent to "Company," that adds no trademark significance to the underlying generic term.
Booking.com's counterargument targeted the analogy directly. A corporate designation like "Company" can be shared by thousands of entities. Goodyear Rubber Company, Goodyear Tire Company, and countless other Goodyear-prefixed companies can all coexist. But a domain name is different. Only one entity can operate booking.com. That exclusivity, Booking.com argued, gives the domain name a source-identifying capacity that "Company" inherently lacks. When consumers see "booking.com," they understand it refers to a specific company, not to the general category of booking services, because only one company can be at that address.
The Goodyear analogy was always strained, and the Court was right to reject it. The 1888 decision addressed a naming convention (corporate suffixes) that is genuinely generic in character. Multiple entities can and do share the same "Generic + Company" format. Domain names operate under fundamentally different constraints. The one-to-one mapping between a domain and an entity is not a minor technical detail; it is a structural feature that shapes how consumers process the term. The USPTO's insistence on treating ".com" as linguistically identical to "Company" ignored this structural difference and elevated formal categorization over marketplace reality. That is the kind of rigid formalism that trademark law, with its grounding in consumer perception, is designed to avoid.
Ginsburg's Reasoning: Why Language Rules Don't Settle Trademark Questions
Justice Ginsburg's majority opinion articulated a principle that sounds obvious but has significant doctrinal implications: whether a term is generic depends on whether consumers perceive it as a brand name or a common descriptor, and no rule of linguistic combination can substitute for that inquiry.
The opinion dismantled the USPTO's argument in two steps. First, the Court rejected the premise that combining two generic components necessarily yields a generic composite. The whole can differ from the sum of its parts. "Tennis" is generic for a sport. "Channel" is generic for a broadcasting medium. But "Tennis Channel" may function as a brand name if consumers perceive it that way. The same logic applies to "Booking" and ".com." The fact that each component, considered in isolation, lacks trademark significance does not foreclose the possibility that the combination has acquired source-identifying meaning.
Second, the Court rejected the Goodyear analogy on its merits. The 1888 case addressed "generic + Company" constructions, and the Court held that ".com" is not equivalent to "Company" for the reasons Booking.com identified: a domain name points to a single entity. "Booking Company" is something many businesses could call themselves. Booking.com is a specific address on the internet that only one entity controls. That factual distinction matters because trademark law asks what consumers understand, and consumers understand that booking.com refers to one particular company.
But Ginsburg threaded a narrow needle, and the opinion's most important practical language is easily overlooked. The Court emphasized that even if a generic.com term qualifies for registration, it would receive only "thin" protection, meaning the scope of the trademark right would be narrow. The owner of a generic.com mark could prevent exact copying but would have limited ability to stop competitors from using the generic term in descriptive ways. A registration for BOOKING.COM would not give Booking.com the right to prevent other travel companies from advertising "booking" services.
This is the tension the decision never fully resolved. The Court opened the door to registration but simultaneously warned that the mark, once registered, would provide limited enforcement value. The practical result: a company can spend hundreds of thousands of dollars on survey evidence to prove acquired distinctiveness (the legal doctrine recognizing that a term originally lacking distinctiveness can gain trademark significance through use in commerce), only to receive a registration that gives narrow rights against close copycats. For a company the size of Booking.com, the registration has strategic value as a defensive asset. For most companies, the cost-benefit calculation does not work.
Breyer's Warning: Monopolizing Useful Language
Justice Breyer's solo dissent raised a concern that the majority acknowledged but did not, in his view, adequately address: the anti-competitive cost of allowing trademark rights in generic.com terms.
Breyer's argument centered on language monopoly. If BOOKING.COM is a registered trademark, then Booking.com B.V. can send cease-and-desist letters to any competitor who uses "booking.com" in advertising or domain variations. The legal reality, Breyer argued, is that trademark owners wield enforcement tools that are blunter than the nuanced doctrinal protections the majority envisioned. A cease-and-desist letter does not come with a careful explanation of thin-mark limitations and the boundaries of the Abercrombie spectrum. It comes with an implicit threat of litigation. Small competitors facing such a letter from a company with Booking.com's resources will fold, even if the legal merits favor them.
Breyer identified a genuine risk. The gap between the theoretical scope of a thin mark and the practical coercive power of a trademark registration is real. Cease-and-desist letters do not engage in careful doctrinal analysis; they assert rights and threaten consequences. The majority's assurance that generic.com marks receive only narrow protection assumes that the protection will be exercised through formal legal channels where courts can police the boundaries. In practice, most trademark enforcement happens through demand letters that never reach a courtroom.
But Breyer likely overestimated the scale of the problem. The evidentiary burden for registering a generic.com mark remains high enough that few companies can clear it. The feared wave of language monopolization has not materialized, in part because the cost barrier functions as a natural check. The companies capable of mounting a successful generic.com registration are, by definition, large enough that their mark has already acquired genuine source-identifying significance in the marketplace. The concern about small competitors being bullied by trademark holders with generic.com registrations is legitimate in theory, but the number of such registrations has remained small enough that the systemic harm Breyer predicted has not occurred.
Where Generic.com Marks Sit on the Distinctiveness Spectrum
Trademark law organizes marks along a spectrum of distinctiveness, known as the Abercrombie spectrum after Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976). From weakest to strongest: generic (the common name for a product, never protectable), descriptive (describes a feature or quality, protectable only with acquired distinctiveness), suggestive (hints at a quality without directly describing it), arbitrary (a real word used in an unrelated context, like APPLE for computers), and fanciful (an invented word, like XEROX).
Generic terms sit at the bottom. They can never function as trademarks because they are the vocabulary everyone needs to describe the product category, making a "generic trademark" a legal contradiction. "Aspirin" was once a brand name; it became generic because consumers started using it as the common word for the product. You cannot register a generic term, period.
BOOKING.COM sits precisely at the boundary between generic and descriptive. The word "booking" is generic for reservation services. The Court held that the combination with ".com" can push the term across the line into descriptive territory, but only if consumer perception evidence supports that conclusion. This places generic.com marks in a peculiar position on the Abercrombie spectrum: they are not inherently descriptive (they start from a generic base), but they can achieve descriptive status with acquired distinctiveness, and they receive narrower protection than marks that are descriptive from the outset.
The practical effect is something close to a new category that the Abercrombie framework does not explicitly account for. Generic.com marks with proven consumer recognition sit above the generic floor but below ordinary descriptive marks in terms of enforcement power. Courts evaluating likelihood of confusion for these marks (using the DuPont factors, the thirteen criteria US courts apply to determine if two marks are confusingly similar) will weigh the mark's inherent weakness heavily. The owner of a generic.com mark may prevail against an identical or near-identical copy, but will struggle to establish confusion against marks that merely use the same generic term in a different format. In opposition and cancellation proceedings, a generic.com registration is both harder to obtain and easier to challenge than a registration for a mark with inherent distinctiveness.
This doctrinal positioning matters for anyone working with trademark data. When you encounter a generic.com registration in a trademark database, the registration alone does not tell you much about enforcement risk. The mark's actual scope depends on the strength of the consumer perception evidence that supported its registration and on how courts in the relevant jurisdiction have interpreted thin-mark protection.
Six Years After the Booking.com Trademark Case
The immediate aftermath of the Booking.com decision was predictable: a noticeable increase in generic.com trademark applications at the USPTO. Attorneys who had previously counseled clients away from generic.com filings suddenly had Supreme Court authority to cite. Applications for marks like HOTELS.COM, CARS.COM, and similar generic-domain combinations appeared or were renewed with fresh arguments based on the decision.
Most of these applications have failed. The Booking.com ruling removed the per se bar, but it did not lower the evidentiary standard. Applicants still need to demonstrate that consumers perceive their specific generic.com term as a brand name rather than a category descriptor. That means survey evidence, and credible surveys are expensive. A properly conducted Teflon survey, the format most directly endorsed by the Fourth Circuit's reasoning in the Booking.com litigation, costs between $50,000 and $150,000, depending on the scope of the survey and the market at issue.
The cost barrier creates a filtering effect that has, in practice, limited the ruling's impact to exactly the kind of applicant that least needed the help: large, well-established companies whose generic.com domains already function as de facto brand names in the marketplace. These are companies that have spent years and millions of dollars building consumer recognition. The Supreme Court told them they could register what they had already built. Smaller companies with generic.com domains, the ones who might theoretically benefit most from registration, typically cannot afford the survey evidence required to prove consumer perception.
The ruling ratified market reality rather than creating a new one. Companies like Booking.com had already invested enough in marketing and brand building that their domain names functioned as source identifiers regardless of formal registration status. The decision gave these companies a federal registration certificate along with the procedural advantages that come with registration (constructive notice, the ability to record with Customs, access to federal courts). But it did not fundamentally change the competitive dynamics of the online travel, hotel, or car rental markets. The companies that won the right to register were already the dominant players. The registration process for generic.com marks simply added an expensive evidentiary step to an already lengthy procedure.
The more interesting question is whether the decision has affected how consumers interact with generic domain names. There is a plausible argument that as more generic.com terms gain trademark recognition, the ".com" suffix itself takes on greater source-identifying significance in the minds of consumers. If that perception shift is real, it could make future generic.com applications progressively easier to support. But that is speculation, and the data over six years does not clearly support it.
What This Means for Brand Strategy, Legal Practice, and Trademark Data
For Brand Owners and Founders
If you own a generic.com domain and are building a business around it, can you trademark a generic domain name? After the Booking.com decision, registration is theoretically available. But "theoretically available" is not "practically achievable." Unless your company has invested significantly in building consumer recognition of the domain as a brand name (not just as a web address), you are unlikely to clear the consumer perception hurdle.
The more actionable lesson is about brand strategy. Building a brand around a generic.com domain is not inherently wrong, but understand what you are signing up for. You will spend more on brand building, more on legal protection, and receive narrower trademark rights than a competitor who chose a distinctive name from the outset. BOOKING.COM has been a commercial success in spite of its generic name, not because of it. The trademark system will accommodate you, but it will not make the path easy or cheap.
For Trademark Attorneys and IP Professionals
The Booking.com decision shifted the genericness analysis from a categorical rule to a factual inquiry, which means more work at both the prosecution and enforcement stages. At prosecution, practitioners advising clients with generic.com marks need to build an evidentiary record early: consumer surveys, marketing expenditure documentation, media coverage showing brand recognition. The per se rule, whatever its doctrinal flaws, had the virtue of providing clear guidance. Its absence means more uncertainty and more litigation.
At the enforcement stage, the thin-mark limitation creates a practical challenge that the decision did not resolve with much precision. How thin is thin? Can the owner of a generic.com mark bring an infringement action against a competitor using the same generic term in a slightly different domain format (say, "bookinghotels.com")? The answer is "probably not, but it depends on the facts," which is the kind of answer that generates billable hours and unpredictable outcomes.
For Developers Building Trademark Tools
Generic.com registrations present a data challenge. In a trademark database, a generic.com trademark registration like BOOKING.COM looks identical in format to a registration for NIKE or GOOGLE. The registration status field does not distinguish between a fanciful mark with broad protection and a generic.com mark with thin protection. Any clearance or screening tool that treats all live registrations as carrying equivalent risk will produce misleading results.
The relevant signal is not the registration itself but the underlying distinctiveness classification and the prosecution history. A mark that was registered over a genericness refusal, with survey evidence in the file, carries a different risk profile than a mark that sailed through examination on inherent distinctiveness. Trademark tools that surface prosecution history and distinctiveness data alongside registration status give users the context they need to make informed decisions. Without that context, a generic.com registration can trigger false positives that either overstate the infringement risk or waste attorney time investigating a mark with narrow rights.
Consult a trademark attorney for legal guidance specific to your situation, particularly when evaluating the enforceability of generic.com marks or assessing clearance risk.
The Booking.com trademark case resolved a doctrinal question that needed resolving. Consumer perception, not linguistic formulas, determines whether a term functions as a trademark. That principle is sound. But the practical impact has been modest, contained by the same evidentiary standards the Court left in place. The door is open. The price of admission ensures that few will enter.
For teams building trademark screening into applications or domain evaluation tools, Signa's API provides programmatic access to trademark registration data across 200+ offices, including distinctiveness classifications and prosecution history.