(January 19, 2023) - Domain name investors register appealing internet domain names for resale. An investor's portfolio may hold tens of thousands or hundreds of thousands of domain names, or more.

A catchy term that may appeal to a domain name investor may also appeal to a business that registers a trademark on that term. This can lead to disputes under the Internet Corporation for Assigned Names and Number's Uniform Domain Name Dispute Resolution Policy. ICANN adopted the UDRP in 1999 as a remedy for the problem of cybersquatting. A complainant bringing a successful complaint under the UDRP is awarded the disputed domain name, which is then seized from the respondent without compensation. There are several venues around the globe that are accredited by ICANN to administer disputes under the UDRP, most prominently the World Intellectual Property Organization, part of the United Nations based in Geneva. WIPO was the lead drafter of the policy concept that ICANN adopted as the UDRP.

While many say the UDRP works quite well in the face of frequent cybersquatting, some domain name investors find fault with it for being employed too broadly. Their view is that the UDRP is not only used against bad-faith cybersquatted domain names, but is also misused to seize inherently appealing domain names held by domain name investors that are offered for sale to the general public and are not harming the complainant.

Westlaw Today asked Nat Cohen, a longtime domain name investor, domain industry advocate and board member of the Internet Commerce Association, to offer his expert knowledge of the domain industry to shed some light on the situation.

Nat Cohen: Cybersquatting is a recurring, persistent problem for brand owners. The UDRP over its 20-plus-year history has been effective in remedying tens of thousands of instances of abusive domain name registrations in a low-cost, expeditious manner. While it is still burdensome for brand owners to have to spend thousands of dollars on a UDRP proceeding to shut down a domain name that may have been registered for $20 or less, this is still far preferable to filing in court and incurring tens of thousands of dollars in legal expenses.

The UDRP is a bare-bones, expedited proceeding where the decision is delegated to an administrative panel. In the vast majority of cases, the panel comprises a single panelist paid a rather token amount to render a decision. The heart of the UDRP is whether the respondent acted in bad faith. The panelist must therefore look into the soul of the respondent to determine his or her intentions based on written submissions and exhibits, without the opportunity for depositions or cross-examinations or oral argument.

This streamlined process is suitable for the most clear-cut instances of cybersquatting. Most disputes brought under the UDRP are "no-brainer" cases since merely from looking at the domain name it is clear that it is targeting a famous mark. The UDRP offers an expedited way of dealing with such blatant cybersquatting without the need to go to court. Yet the UDRP was not designed nor is it well suited to adjudicate more nuanced or complicated issues, such as infringement claims, defamation claims, stolen domain names, business disputes, or where an investor claims a right to market a nondistinctive, widely used term that may also happen to be used in a particular location by a company for certain goods or services.

Nevertheless, complaints are brought under the UDRP to address all these issues. Whether these disputes fall within the scope of the UDRP is left to the discretion of panels. While some panels will appropriately deny a complaint on the basis that it is outside of the scope of the UDRP, many panels will plunge ahead, perhaps with the view that they are being paid to render a decision on the pleadings and that they'll give it their best shot.

While most panelists give both sides a fair hearing and require that the complainant provide adequate evidence and offer a sound argument to prevail, the UDRP is lacking the procedures to properly assess certain types of disputes and to develop a consistent jurisprudence that is more than the luck of the draw as to which panelist is assigned to resolve the dispute. There are hundreds of panelists, with no set qualifications to serve, who come from varied legal traditions, who are called upon to make subjective inferences based upon an often-scant evidentiary record, who are appointed at the discretion of a various UDRP administrators, who are themselves chosen for their roles and empowered to operate the UDRP without community input and who wield their power without any accountability to the community. These structural defects are inherent in the policy that was adopted 23 years ago and have not been addressed since.

If one is a brand owner seeking the transfer of a domain name that is blatantly infringing on your well-known mark, the UDRP can deliver quick, efficient justice. If one is a domain name investor making a livelihood dealing with inherently appealing, nondistinctive domain names that have some similarity to terms that are registered trademarks, the UDRP is ill-suited to make a determination as to whether one registered the domain name due to its inherent appeal or instead in bad faith to target the trademarked use. The subjective inferences that UDRP panelists make in assessing bad faith in disputes involving domain name investors vary widely from panelist to panelist, such that the UDRP puts the livelihood of domain name investors in jeopardy by failing to consistently deliver reliable, fair or credible outcomes.

NC: Most UDRP panelists are experienced IP attorneys. Many are active brand enforcement attorneys who will at times file UDRP complaints to protect clients' brand rights. This gives them experience with the UDRP that is helpful when they serve as UDRP panelists and most of them have demonstrated objectivity in their decisions.

Yet in many cases it can't but help color the perspective that they have towards the UDRP, which they view as a brand enforcement tool. The UDRP is loosely written and open to interpretation and relies on subjective inferences by the panelists.

Borrowing the more familiar terminology from criminal cases, since in the UDRP the "judges" are usually active "prosecutors" who are concurrently filing complaints under the UDRP on behalf of clients, these judges look favorably on arguments made by other prosecutors, as they are the same arguments that they themselves make. Likewise, they look less favorably on arguments made by defense attorneys. This dynamic occurs in UDRP disputes decided by panels composed of active brand enforcement attorneys. Notably, the "defense bar" has been intentionally excluded from an equivalent role in administering the UDRP. This exclusion calls into the question the fairness and neutrality with which the UDRP is administered.

While the UDRP was adopted by ICANN as a consensus policy with input from various stakeholder communities, it has since been implemented and interpreted nearly exclusively by those aligned with brand owners. The three criteria a complainant must demonstrate are not well defined in the policy with illustrative examples offered "without limitation." As a result, the consensus views as to how the UDRP should be interpreted offers clearer guidance than the policy itself. Indeed, at WIPO training sessions, panelists are advised to refer to WIPO's distillation of the jurisprudence developed by its roster of hand-selected panelists, notably from which respondent representatives have been excluded.

Twenty years of the compounding effects of the UDRP being interpreted nearly exclusively by brand-side advocates have led to a jurisprudence that bears little resemblance to the actual language of the policy and that advantages complainants while disadvantaging respondents. The first test of "identical or confusingly similar" has strayed far from its legal meaning such that panels will look at the content of the webpage rather than the similarity of the domain name to the mark to find that, for instance, "Guaiky" is confusingly similar to "Quirky," "Boksoe" is confusingly similar to "Brooks," and "Hoicap" is confusingly similar to "Cat."

The second test stating that legitimate interest can be demonstrated by evidence that "before any notice to you of the dispute, your use of, or demonstrable preparations to use, the domain name or a name corresponding to the domain name in connection with a bona fide offering of goods or services," is often ignored by panels who will only consider use at the time the complaint is filed. The third test regarding bad-faith intent with respect to the complainant is often ignored, as some panels will find bad faith even while acknowledging that there may not be sufficient evidence that the respondent was even aware of the complainant. The disproportionate influence that brand interests have over the UDRP has led to a judicial process that is neither sufficiently neutral nor balanced.

NC: No, it does not.

There is a limited supply of inherently meaningful terms and memorable short acronyms. This supply is dwarfed by the demand for intuitive monikers by countless companies around the globe. Numerous articles in the popular press and even law review studies are asking, "Are we running out of trademarks?" See Barton Beebe & Jeanne C. Fromer, "Are We Running Out of Trademarks? An Empirical Study of Trademark Depletion and Congestion" at https://bit.ly/2EuRQge. This article questions the long-held assumption underlying trademark law that "when we grant exclusive rights in a trademark, the cost to competitors, consumers, and more generally to the public domain is inconsequential."

Growing mark congestion means sharing of marks is increasingly common. The most appealing names have long been adopted for commercial use and registered as trademarks by one or more companies in one or more countries around the globe. A new entrant to the marketplace wishing to adopt an appealing identity will usually find that it has already been chosen by another company. But this is not necessarily a problem, as a newcomer is entitled to use a mark that is the same as a previously registered mark held by another company so long as it is for a non-infringing use.

Yet what if the new entrant wishes to make use of the dot-com domain name matching their shared mark? Does being the first to register a trademark on a term also entitle the company to a monopoly over all similar domain names such that the company that is the first to register a mark is entitled to seize any similar domain names held by domain name investors?

Is it in the public interest for the new entrant to be able to purchase the exact match dot-com domain name if it is available for sale and if the new entrant so chooses? Or is it in the public interest for older trademark registrants to be able to hoard all the most desirable domain names and deny them to new entrants?

The first UDRP complaint filed against a dictionary word domain name, in January 2000, was on the domain name craftwork.com. Gen. Mach. Prods. Co. Inc. v. Prime Domains, No. FA0001000092531, (Nat. Arb. Forum Mar. 16, 2000). I had registered the craftwork.com domain name a couple of years prior to the dispute. The complainant had two U.S. trademark registrations for stylized versions of the mark "Craftwork" that predated my registration. The three-person UDRP panel unanimously found that the complainant's trademark registrations did not entitle the complainant to exclusive commercial rights to craftwork.com such that I was free to register craftwork.com and offer it for sale so long as I was not targeting the complainant's goods and services.

Coincidentally, 23 years later, a few days before I am writing this, craftwork.com was purchased by an unrelated buyer. This is consistent with an established principle of trademark law: that nondistinctive terms are not exclusively associated with any one party and that, absent a famous mark, multiple companies can make non-infringing use of the same mark for various products and services. If I had conducted a trademark search back in 1998 and as a result discovered that a trademark registration already existed for a specific industrial use on the term "Craftwork," that information would have been of little utility. The domain craftwork.com would still have been an inherently appealing domain name that it was legitimate to register for resale, as the recent sale to a third party demonstrates.

Domain name investors serve in effect as online naming agencies. We offer a curated assortment of appealing domain names for startups and other companies looking for a memorable online presence. Just as many companies often coexist while using the same trademarked term for different products and services, a domain investor can offer a domain name that may match an existing trademarked term for sale to new entrants to the marketplace that are looking for an appealing term to serve as a brand that does not infringe on any existing use.

Domain names from Telepathy's portfolio have been chosen for branding initiatives by Sony (crackle.com), Facebook (Meta) (scaler.com, meta.in), Alphabet (Google) (crr.com, verily.com), Apple (acquired Chomp using chomp.com), AOL (ygm.com), among others. A global search of trademark databases would have revealed prior registered trademarks for some or all of these domain names. If it is a violation of the UDRP merely to offer for sale a domain name that matches an existing registered trademark, this would stifle the dynamism of the internet by denying new companies the opportunity to acquire desirable, intuitive and memorable domain names that are held by domain name investors. Instead, older companies who have prior trademark registrations for similar terms could use the UDRP to seize and hoard all the most desirable domain names that are similar to their trademarked term and thereby deny innovative new companies the opportunity to acquire an appealing, intuitive domain name matching its brand identity.

Unfortunately, some UDRP panelists have an outdated notion of trademark exclusivity. They would grant the owner of a registered mark the presumptive right to similar domain names, even in the absence of evidence of bad-faith intent or targeting. The reasoning that once a company acquires a trademark registration it is bad faith to offer for sale a similar domain name, even if the domain name is not exclusively associated with the complainant, is articulated in the dissenting panelist's opinion in the reza.com dispute Reza IP Holdings LLC v. Alireza, No. D2022-0945, (WIPO Arb. June 28, 2022). See "Dissenting Panelist Would Have Decided Case Wildly Differently," at https://bit.ly/3vUgrmX.

The panel majority in the reza.com dispute rejected this reasoning, and indeed the complainant in the reza.com dispute was found guilty of "reverse domain name hijacking." Too often though, similar views are offered to justify transfers despite insufficient evidence of bad faith.

Even when a domain name owner truthfully claims not to have previously been aware of the complainant, certain UDRP panelists will nevertheless rely on the concept of "willful blindness" to justify transfers despite an absence of evidence of bad-faith targeting. These panelists assert that domain name investors have a duty to conduct searches and then to avoid any domain names that are similar to trademark registrations that appear during the searches. As applied to the UDRP, willful blindness is based on the fallacious notion that a trademark registration grants the registrant the rights to enforce a global, perpetual prohibition on the buying and selling of similar domain names.

When a domain name investor discovers an inherently appealing nondistinctive domain name, perhaps because it is being offered during the daily auctions of deleting domain names, the domain name investor is under no obligation to conduct trademark searches before registering the domain name and offering it for general sale, as such an offer for sale infringes no company's trademark rights. Nor would discovering, after the investor had already identified the domain name as an inherently appealing term that fit her investment strategy, that a company had prior trademark rights to a similar term mean that the domain investor was acting in bad faith in proceeding to register the domain name and offer it for sale.

NC: Certain complainants attempt to use the UDRP as a low-cost domain name acquisition tool. When a complainant desires a nondistinctive domain name that is not causing it any harm, not because the domain name is being used in bad faith but because the complainant wants to use the domain name itself and attempts to use the UDRP to seize the domain name without compensating the domain name owner, that is often a circumstance in which a finding of RDNH, or reverse domain name hijacking, is appropriate. There have been over 500 such findings made throughout the history of the UDRP. See Zak Muscovitch, "The UDRP 'Celebrates' Its 500th Reverse Domain Name Hijacking Case," at https://bit.ly/3X1lv4Z.

While an RDNH finding is rather toothless, as it comes with no penalty or sanction, it is not without risk to the complainant and to the complainant's counsel. There can be reputational harm, and a frivolous UDRP complaint can itself create a cause of action against the complainant, as discussed in Zak Muscovitch's "The Hidden Perils of Filing a Baseless UDRP Complaint," at https://bit.ly/3CL0eo1.

NC: Yes, legitimate domain name registrants can be unfairly labeled as cybersquatters. A common recurring situation is that some panelists expand a complainant's limited trademark rights into a global entitlement to similar domain names. This can result in domain name registrants being unfairly labeled as cybersquatters and losing their domain names through a UDRP dispute.

The UDRP was developed in the late 1990s, a generation ago in human terms, and eons ago in terms of the development of the internet. It was developed in part to solve a problem that no longer exists. The problem was that at the dawn of the commercial internet, brands like Coca-Cola, McDonald's and 7UP (see "7UP and March Madness," at https://bit.ly/3XtYxTB) did not realize the importance of registering their matching dot-com domain names, thereby allowing others to register those domain names first. The UDRP was therefore enacted to prohibit the registration and resale of domain names "primarily for the purpose" of targeting the complainant's trademark rights. Now a third-party registering the exact match dot-com domain name of a well-known brand is rarely an issue, as one of the first things a company does when naming itself or developing a new brand is to secure the matching domain name.

The domain name resale marketplace was in its infancy when the UDRP was developed. It is now a robust industry in its own right, with an estimated $2 billion in transactions occurring annually (See "What's in a (Domain) Name? The $2 Billion Secondary Market for Dot-Com Domains," at https://on.bcg.com/3QAaSU6). Domain name investors and brand owners both draw from the same pool of appealing word formations to source their portfolios and brand identities, respectively. Domain name investors are looking to offer the brands of the future, while trademark registrants have already secured their brands.

This creates a clash. When a domain name investor is offering for sale a domain name that is similar to an existing brand, did the domain name investor register the domain name "primarily for the purpose" of selling it to the existing brand, or is the investor free to offer the domain name for sale to a third party that might wish to make a non-infringing use of the domain name?

There is little clear guidance under the UDRP. A UDRP panel is empowered to draw whatever inferences seem appropriate to it. As the panel is given, at times, the seemingly impossible task of determining, based on little evidence, the intentions of a domain investor many years earlier when it first registered the disputed domain name, and as the evidentiary standard is merely a "balance of probabilities," the ownership rights of participants in the domain name aftermarket are made precarious by being subject to an inconsistent, speculative and, at times, capricious assessment.

One flaw of the UDRP is that it does not require a nondistinctive domain name to be used to specifically target the complainant for a panel to make a finding of cybersquatting. Nor does the complainant need to demonstrate that it was harmed by the use. Thus, complainants can use the UDRP to seize nondistinctive domain names to which they do not have exclusive rights and that were put to a use that did not cause them any meaningful harm.

UDRP panels have controversially ordered the transfers, for example, of the domain names canary.com to an oil field services company (see https://bit.ly/3IGmhjv) and the valuable two-letter dot-com domain name lh.com to Lufthansa. Deutsche Lufthansa AG v. Future Media Architects Inc., No. FA0802001153492, (Nat. Arb. Forum Apr. 17, 2008). Also consider visitqatar.com, which was transferred to the Qatar Tourism Council despite the respondent having a U.S. registered trademark (see https://bit.ly/3kt9eYF), and the case of buttonmakers.com to a competing company that also made buttons (see Rockstar Industries LLP v. ButtonMakers.com, No. FA1601001658525, (Nat. Arb. Forum Mar. 10, 2016)). See also "Small Businesses and Investors Losing Domains in Flawed UDRP Decisions," at https://bit.ly/3IWWSCf.

These decisions were controversial because these domain names are all based on nondistinctive terms that were not exclusively associated with the complainant, there were a multitude of other companies that could and did make a legitimate non-infringing use of the terms on which the domain names were based, and the evidence did not appear to justify a finding that the respondent had violated the UDRP. When a UDRP panel orders an unjustified transfer, a policy that was created to remedy harm becomes itself a cause of harm.

Especially problematic for domain name investors is that panelists will at times draw inferences of bad-faith cybersquatting from the asking price an investor sets on the domain name. If the panelist considers that the asking price is "too high," then the panel may draw an inference that this is evidence that the domain registrant registered the domain name to target the complainant's trademark rights. See "Smells Like Cybersquatting – How the UDRP Smell Test Can Go Awry," at https://bit.ly/3X8R6S3.

Unfortunately, many panelists are not aware that the domain investor business model with its typically very low sell-through rate requires high markups to make the business viable. Nor are panelists necessarily experts on domain name valuations. Nor is it the proper role of panelists to set price controls on aftermarket domain names by ordering the seizure of domain names that in their subjective opinion are priced "too high." Drawing inferences of bad faith from a supposedly high asking price has led to many instances of domain investors being unfairly labeled cybersquatters and their property seized from them. One well-known example is in the dispute over the ado.com domain name. Autobuses de Oriente ADO SA de CV v. Carrillo, No. D2017-1661, (WIPO Arb. Feb. 1, 2018). See "ICA Statement on ADO.com UDRP Decision: Overreaching Panelists and Interference With the Domain Market," at https://bit.ly/3istB7t.

This is not to say that cybersquatting does not exist. Cybersquatters tend to target well-known distinctive brands. Domain name investors deal in inherently appealing, nondistinctive domain names. It may be a challenge to differentiate a cybersquatter who registered a domain name to target an existing trademark holder from a domain investor who registered a similar domain name but with the intention of selling it to a third party. The key elements used to distinguish the two cases are often the degree of distinctiveness of the mark, the extent of third-party use, whether the investor can show a strategy of registering nondistinctive domain names that fit a similar pattern or theme, and whether the use is specifically targeting a particular trademark owner.

Domain name investors who engage competent counsel and present a thorough defense, with adequate evidence, have a good record of success under the UDRP, though there are unfortunate exceptions. Sometimes this requires that the investor incur the burdensome expense of filing a lawsuit in court to overturn an erroneous UDRP panel decision to transfer one's domain name.

An overly expansive view of trademark rights does not only affect domain name investors, but it can also interfere with free speech rights, as companies have been successful in using the UDRP to shut down criticism sites. I discuss the misuse of the UDRP to suppress critical speech in "Does the UDRP Interfere With Free Speech Rights – The StopSpectrum.com Decision" (https://bit.ly/3wcp4t9).

NC: Legitimate domain name investing is an unusual business. The critical thing to know about domain name investing is that in a good year, a typical domain name investor will sell only about 2% of the domain names she holds. See Elliot Silver, "My Sell Through Rate is Too High," at https://bit.ly/3vXqEin.

For most investors to earn a reasonable profit on the time and capital committed, the investor must set an asking price that is far higher than her acquisition cost for any individual domain name. Only in this way can the very small percentage of domain names that sell each year generate sufficient revenue to make it a viable business and compensate for the vast majority of domain names that the investor acquires that will never find a buyer. As a rule of thumb, an investor will usually be willing to commit $1,000 to acquiring a domain name only if she believes that it has the potential to attract a buyer at a sales price of between $5,000 to $20,000 or more.

NC: One should have realistic expectations. Far from being glamorous, domain name investing is a very time-intensive and tedious business and often requires a significant upfront cash investment.

As is said about professional poker players, "it is a hard way to make an easy living." Competition for appealing domain names is global and intense. As only a small percentage of domain names sell each year, most of the names you acquire will never find a buyer. This can also make it hard to learn from your mistakes, as meaningful customer feedback is so infrequent. One must either be willing to hustle as a salesperson for often small margins or be willing to wait decades to find a buyer. For instance, many of the domain names that I sold in the past couple of years were names I registered over 20 years ago.

It helps if you focus on a certain niche that you believe gives you a competitive advantage — perhaps a certain domain name extension appeals to you, or perhaps you have expertise in a certain subject area or industry. Otherwise, when you are winning domain name auctions against other, more experienced domain name investors, it likely means that you are overpaying.

Fortunately, there is a wealth of free or inexpensive ways to learn about domain name investing, with numerous informative blogs, podcasts, videos and research tools that offer useful guidance before you invest funds on perhaps ill-considered speculative domain name acquisitions.

The key question I ask myself when considering whether to register a domain name is whether I can imagine a company choosing the domain name for its online identity. If the answer is "no," then it probably is not worth registering.

NC: The viability and credibility of the UDRP depends upon you. Recognize that the UDRP is a simplified procedure intended for clear-cut instances of cybersquatting. The goal is not to come up with the most clever and novel way to justify a transfer despite circumstances that would argue against ordering the transfer. Nor, if you are a practicing IP attorney, is serving as a UDRP panelist an opportunity to advocate on behalf of your clients' interests.

Rather, it is perfectly acceptable to find that the circumstances in a dispute are more complex or nuanced than can be adequately resolved by the UDRP such that the dispute should be considered outside the scope of the UDRP. Similarly, if you are unsure of the proper outcome, do not feel that you are required to flip a coin and to make a transfer decision based on a "balance of probabilities" when that means that there is a high probability that you are making the wrong decision and causing avoidable harm.

If the complainant has a meritorious claim to make that is beyond the scope of what the UDRP is suited to adjudicate, that case should be decided on a full factual record in a court of law. For the UDRP to be a tool for good rather than for harm requires that UDRP panelists exercise restraint and resist the temptation to dispose of valuable domain names based on little more than guesswork, even if the UDRP empowers you to do so.