In a recent decision addressing the propriety of Google’s “sale” of the trademarks of others for the purpose of triggering Sponsored Links in search engine results, the district court for the Eastern District of Virginia concluded that Google’s actions were not likely to give rise to consumer confusion in the marketplace, that Google was using plaintiff’s trademarks in a functional way, that Google was not liable under plaintiff’s theories of contributory and vicarious infringement, and that Google’s “use” of plaintiff’s famous ROSETTA STONE trademarks did not result in trademark dilution. As a result, the court granted summary judgment in favor of Google, dismissing Rosetta Stone’s case.
Plaintiff Rosetta Stone, a company specializing in software for learning foreign languages, alleged in its complaint that Google, by way of its AdWords program, was actively assisting third party advertisers to mislead consumers and misappropriate Rosetta Stone’s trademarks by using the trademarks (i) as keyword triggers for paid advertisements and (ii) within the title and text of paid advertisements on Google’s website. Briefly, the AdWords program permits advertisers to “bid” on keywords that, when combined with their ad’s quality and maximum bid price, will trigger a “Sponsored Link” to the advertiser’s chosen web site. Those familiar with Google’s search engine will recognize the Sponsored Link section of a search result page, with links to specific websites of third party advertisers who typically sell the types of products or services that were the subject of the searcher’s inquiry.
Rosetta Stone’s complaint grew out of Google’s policy of permitting advertisers unrelated to Rosetta Stone to use Rosetta Stone words, marks or phrases as keyword triggers that cause sponsored links to appear when a searcher enters a search term. As alleged by Rosetta Stone, Google, by way of its Sponsored Links, was “helping . . . advertisers misdirect web users to websites of companies that (i) compete with Rosetta Stone, (ii) sell language education programs from Rosetta Stone’s competitors, (iii) sell counterfeit Rosetta Stone products, or (iv) are entirely unrelated to language education.” Google’s AdWords program, therefore, in Rosetta Stone’s eyes, amounted to “trademark infringement . . . driven by an economic incentive to increase the number of Sponsored Links that appear for every term entered into its search engine because Google is paid by its AdWords advertisers on a ‘cost-perclick’ basis.”
The court was not persuaded by Rosetta Stone’s theories and, in fact, in granting summary judgment on every count, made it clear that Rosetta Stone had failed to show the existence of genuine issues of material fact sufficient to support any of its claims. The court first addressed the trademark infringement allegation by Rosetta Stone and concluded that “no reasonable trier of fact could find that Google’s practice of auctioning Rosetta Stone’s trademarks as a keyword triggers to third party advertisers creates a likelihood of confusion as to the source or origin of Rosetta Stone’s products.” In reaching this conclusion the court focused on three of the traditional “likelihood of confusion” factors, namely intent, actual confusion, and the sophistication of the consuming public (the court concluded that the other factors were not at issue between the parties).
As to intent, the court found that no reasonable fact finder could conclude that Google intended to confuse potential purchasers by auctioning Rosetta Stone’s marks. The court explained that this was not a situation where a defendant (in this case Google) was creating knock-off goods in an attempt to pass them off as another’s. Rather, the court analogized Google to a newspaper or magazine selling advertising space -- Google provides the space for others to advertise; it does not sell goods of its own. Similarly, the court was not persuaded by Rosetta Stone’s argument that Google profited from its sale of keywords. The fact that Google made money by way of its AdWords program did not mean that it intended to cause confusion amongst the buying public. The court noted that there must be something more than financial gain to establish an intent to confuse. Indeed, the court reasoned that it would not be in Google’s best interest to facilitate confusion because if consumers were confused they would not be as likely to continue to utilize the Google search engine, which in turn would mean less revenue for Google.
On the issue of actual confusion, neither party disagreed that evidence of actual confusion would be the best evidence of likelihood of confusion. However, the court found de minimis Rosetta Stone’s purported “confusion” evidence -- testimony from only five individuals after a six-year period wherein Rosetta Stone’s marks had been used to trigger more than 100,000,000 advertisement impressions on Google search result pages. Moreover, the court found that the five individuals put forward by Rosetta Stone in support of its “actual confusion” argument (allegedly as a result of reaching the vendors’ website through Sponsored Links on a Google search page) were not actually confused. The witnesses knew they had not purchased products directly from Rosetta Stone and thus there was no confusion as to source.
Regarding consumer sophistication, the court also was not persuaded by Rosetta Stone’s argument that the appropriate market was the public at large. Rather, the court recognized that purchasers of Rosetta Stone’s products (which often sell for several hundreds of dollars), were “well-educated consumers willing to invest money and energy in the time-intensive task of learning a language.” Such consumers are, the court felt, likely to be discerning shoppers who will be able to distinguish between Sponsored Links and the organic search results displayed on Google’s search results page. On balance, the court concluded that Google’s use of Rosetta Stone’s marks did not amount to trademark infringement.
Turning to the unique argument of functionality, the court concluded that the functionality doctrine protected Google’s use of Rosetta Stone’s trademarks as keyword triggers. Specifically, the court found that Google’s use of the trademarks “ha[d] an essential indexing function because they enable Google to readily identify in its databases relevant information in response to a web user’s inquiry.” Essentially, the court agreed with Google that the use of keywords was the most efficient way of ensuring that advertisers were able to put their products and services in front of the appropriate group of consumers and thus was an appropriate “functional” use of Rosetta Stone’s marks.
With respect to Rosetta Stone’s allegations of contributory trademark infringement, the court explained that in order to succeed on such a claim Rosetta Stone would have to prove that Google intentionally induced another to infringe a trademark, or that Google continued to supply its product (service) to an entity Google had reason to know was engaging in trademark infringement. Rosetta Stone’s argument in support of its theory relied upon the fact that Google’s keyword suggestion programs for helping advertisers generate appropriate keywords for their business often suggested trademarks or brand names as keywords and thus directly induced advertisers to infringe. Rosetta Stone also argued that by allowing alleged counterfeiters to open AdWords accounts -- despite Google having been informed by Rosetta Stone that such advertisers were allegedly infringers -- Google was “supplying a service to those it knows or has reason to know is engaging in trademark infringement.”
The court found two main problems with Rosetta Stone’s argument. First, the court ruled that the mere fact that Google offered a tool that could be used to assist advertisers (albeit in some cases counterfeiters) to optimize their advertisements was not something that, in and of itself, indicated an intention to induce infringement. Indeed, the court stated that it only made good business sense for Google to have offered such a tool because the more “hits” an advertiser gets, the more money Google makes. Thus, putting a tool in place to optimize hits was logical. Second, the court found no evidence that Google supplied its services to entities it knew to be infringers or counterfeiters. While there was no question that counterfeiters and infringers operated though Google by way of the AdWords program, the court was not convinced that Google actually knew which advertisers were guilty and/or what specific products were infringing or counterfeit. Drawing on the recent decision in Tiffany, Inc. v. eBay Inc., the court found that Google’s “generalized knowledge” that some participants in the AdWords advertising program may be involved in infringing or counterfeit activity did not shift the burden to Google to seek out the alleged wrong-doers, verify the alleged infringements, and thereafter ban them from using the AdWords program. Certainly, in the court’s opinion, Google’s failure to do all of this did not raise issues of contributory trademark infringement.
Rosetta Stone’s vicarious liability theory also failed. Recognizing that in order to be found vicariously liable the alleged infringer must exercise some control over the infringing use, the court found that because Google did not have joint ownership over the allegedly infringing advertisements and did not control those advertisements, Google could not be vicariously liable for trademark infringement.
Finally, the court addressed Rosetta Stone’s dilution claim. While the court had no problem finding the ROSETTA STONE trademark sufficiently famous for purposes of Rosetta Stone’s claim, it could not side with Rosetta Stone on the ultimate question. The Dilution Act, the court explained, “provides that a claim for dilution is not actionable if it involves ‘any fair use . . . of a famous mark by another person other than as a designation of source for the other person’s own goods or services, including use in connection with . . . advertising or promotion that permits consumers to compare goods or services.’” Google’s “use” of the plaintiff’s marks, the court concluded, was not actionable because Google did not use the marks on its own goods or services. Moreover, the court was skeptical of Rosetta Stone’s dilution claim because Rosetta Stone put forth no evidence that Google had caused a blurring or tarnishing of Rosetta Stone’s trademarks. Rather, the evidence of record showed that Rosetta Stone’s brand awareness had actually increased in the time since Google began its AdWords program. Without any evidence that the ROSETTA STONE trademark had lost distinctiveness and with no evidence that Google’s actions had resulted in tarnishment or blurring, the court concluded that no reasonable trier of fact could find in Rosetta Stone’s favor on the issue of dilution, and thus dismissed the dilution claim as well.
The Rosetta Stone decision is perhaps the most comprehensive United States decision to date on the question of the propriety of Google’s AdWords program. Rosetta Stone’s appeal is currently pending.