Since our previous article on the emerging issue of trademark rights in hashtags, the use of hashtags in social media marketing has continued to grow. Described as the “ignition keys to a social media keyword search,” hashtags can be powerful tools for creating communities around a brand. Indeed, recent scholarship suggests that modern brand narratives are written in collaboration with consumer communities rather than by brand owners acting alone.

A catchy hashtag creates its own social media channel and brand owners naturally want to prevent competitors from hijacking the content stream tied to their cleverly crafted messages. To safeguard the investment in this narrative, companies are increasingly seeking trademark protection for their hashtags. Applications for hashtag trademarks continue to soar, with over 1,042 hashtag trademark applications in 2015 in the United States alone. However, despite the United States Patent and Trademark Office’s (or USPTO’s) guidance regarding hashtag trademarks in its Trademark Manual of Examining Procedure (or TMEP), hashtags continue to pose challenges for both USPTO examiners and the courts.

At the USPTO, examination and registration of hashtag marks remain somewhat inconsistent. As we noted in our previous article, the USPTO has addressed the issue of hashtags’ ability to function as trademarks in the “Hashtag Marks” section of the Trademark Manual of Examination Procedure. In essence, the TMEP states that the hashtag symbol should be ignored by the examiner and the hashtag mark should be examined in the same manner that any other tag line or phrase would be. In other words, according to the USPTO, a hashtag is no more—but also no less—capable of functioning as a trademark than the non-hashtag form of the relevant tag line or phrase would be. But does this approach ignore some unique features of hashtag marks?

First, while descriptiveness is an issue for both hashtag and non-hashtag marks, the fact that hashtag marks also function as online search terms would seem to increase the need for a hashtag to have a close and obvious connection to a particular brand if it is to be recognized as a trademark—i.e., an identifier of the source of goods and services—and not merely a search term. For example, the word “Tasty” may be merely descriptive when used on a package of bread, but #Tasty is arguably even less distinctive when used only in a social media campaign, considering that the hashtag does not actually appear on the product and, considered as a search term, could be relevant to any number of topics.

The TMEP notes this problem and instructs that #Skater for skateboarding equipment would not be registrable as merely descriptive. However, in practice, the USPTO has not always been entirely consistent in assessing the descriptiveness of hashtag trademarks. For example, the USPTO has allowed #LetsBowl for bowling balls and #Smart for clothing without raising a descriptiveness objection. The USPTO also allowed Abercrombie & Fitch to register #SoCalStylist for retail store services featuring clothing and accessories, and a non-profit to register #KickHunger for promoting public awareness of hunger and hunger relief. But the USPTO found #WeatherWednesday for an online newsletter about the weather and #MusicVideoMonday for advertising services and mobile marketing to be merely descriptive, allowing only registration on the supplemental register.

Second, the fact that a hashtag often appears only on social media rather than on the goods themselves or in advertising raises questions regarding what constitutes an acceptable specimen for a hashtag mark. The USPTO has not formulated a clear policy on this issue. The TMEP notes that if #SewFun was the subject of a trademark application for “instruction in the field of sewing” with a specimen consisting of a screenshot of a social networking site used to organize user comments about sewing classes that the applicant offers, the mark would be refused registration for failure to function as a service mark. Accordingly, the USPTO rejected the initial specimen for #LeadershipFlow in connection with business education services that consisted of the applicant’s website with posts about business topics. However, the USPTO allowed Procter & Gamble to register #LikeAGirl for “providing information in the field of female empowerment, anti-gender discrimination via social media” with a screenshot of its Twitter page as a specimen.

Only a handful of court decisions have dealt with the subject of trademark rights in hashtags to date, with similarly inconsistent outcomes. As we previously wrote, a district court in Mississippi held that use of the tags #FratCollection and #FraternityCollection by a competitor of the clothing maker Fraternity Collection was sufficient to state a claim for false advertising under the Lanham Act and for trademark infringement under state law. However, in a recent California case, Eksouzian v. Albanese, the court concluded that a competitor’s use of a hashtag did not violate a settlement agreement on trademark usage between the parties because the hashtag was “merely a functional tool.”

The parties in Eksouzian had jointly developed a compact vaporizer pen, but later separated and entered into a settlement agreement pursuant to which the plaintiffs were permitted to use the terms “Cloud” or “Cloud Vapes” as trademarks, but not in such close association with the words “pen” or “penz”—common descriptors for compact vaporizers—as to form a unitary trademark. Plaintiffs then used the hashtags #cloudpen and #cloudpenz in connection with promotional contests on social media. The court found that plaintiffs did not breach the settlement agreement because “hashtags are merely descriptive devices, not trademarks, unitary or otherwise, in and of themselves” and use of the hashtag “is merely a functional tool to direct the location of Plaintiffs’ promotion so that it is viewed by a group of consumers, not an actual trademark.” This conclusion seems to be at odds with the USPTO’s willingness to register hashtags as trademarks.

In another recent case, Public Impact, LLC v. Boston Consulting Group, Inc., a Massachusetts court came to the opposite conclusion. In that case, Public Impact, LLC, an education policy and management consulting firm that owns a federal registration for the mark PUBLIC IMPACT, sought a preliminary injunction to prevent the defendant, Boston Consulting Group (or BCG), from using the hashtag #PublicImpact and the username @4PublicImpact on social media. After determining that BCG had not submitted sufficient evidence to show that “public impact” is generic for consulting services in light of the fact that Public Impact’s federal registration had attained incontestable status, the court concluded that BCG’s use of the username and hashtag was likely to constitute trademark infringement, particularly given the similarity of the services provided by the two organizations. Accordingly, the court enjoined BCG from using the phrase “public impact” with two or fewer letters, numbers, or characters appended in any form on social media or in other commercial activities.

In sum, the application of trademark law to hashtags and the rapidly evolving social media landscape is still in its very early stages. With regard to federal registration, one scholar has argued that the USPTO should treat hashtag marks as “primarily merely a hashtag” until the applicant can establish that the mark actually functions as a source indicator, an approach that could avoid some of the inconsistencies seen in the registration process today. The bigger questions regarding the scope of protection afforded to hashtag marks and the analysis of trademark infringement involving use of hashtag marks on social media, however, have yet to be resolved.