Plaintiff sought a declaratory judgment that its purchase of defendant’s ACTIVEBATCH trademark as a search-engine keyword did not constitute trademark infringement. The district court issued a preliminary injunction against plaintiff’s use of defendant’s mark as a keyword, emphasizing the “Internet Trinity” likelihood-of-confusion factors (i.e., similarity of marks, relatedness of goods/services, and the simultaneous use of the Internet as a marketing channel). The Ninth Circuit reversed, holding that the district court failed to weigh the likelihood-of-confusion factors flexibly to match the specific facts of this keyword case. The Ninth Circuit set out a new four-factor test for evaluating a likelihood of confusion in keyword-advertising cases, which included a new factor: the labeling and appearance of the sponsored ads and their surrounding context on the search-results page. The court also noted that when examining claims of initial-interest confusion, “the owner of the mark must demonstrate likely confusion, not mere diversion.”
Network Automation, Inc. (“Network”) sold job-scheduling software under its AUTOMATE mark. Advanced Systems Concepts, Inc. (“Systems”) sold competitive job-scheduling software under the federally registered mark ACTIVEBATCH. Network purchased ACTIVEBATCH as a search-engine keyword, which triggered Network’s sponsored ads in the search results. Network’s sponsored ads did not contain the ACTIVEBATCH mark in their title or text. Network identified itself only with its URL in the fourth line of the ads; neither the ad title nor text identified Network. The titles of the ads included “Batch Job Scheduling” and “Windows Job Scheduling.” Systems objected to Network’s actions, and Network filed for a declaratory judgment that its actions did not constitute trademark infringement. Systems counterclaimed for trademark infringement and moved for a preliminary injunction.
The district court sided with Systems and issued a preliminary injunction against Network’s use of ACTIVEBATCH as a search-engine keyword. It initially found that Systems was likely to succeed in proving that Network’s actions constituted “use” of the ACTIVEBATCH mark in commerce. Turning to the merits and the likelihood-of-confusion factors, the district court emphasized the importance of the “Internet Trinity” factors: (1) the similarity of the marks, (2) the relatedness of the goods, and (3) use of the Internet as a marketing channel. It found that all three of these factors favored Systems. Regarding the remaining likelihood-of-confusion factors, the district court found that Systems’ mark was strong, Internet consumers likely used a low degree of care in purchasing both parties’ products, and Network intentionally used Systems’ mark to advertise its own product. The district court also noted that Network’s sponsored ads described above were not “clearly labeled” and thus “may initially confuse consumers” into clicking on Network’s ads.
On appeal, the Ninth Circuit vacated the injunction, holding that Systems did not show a sufficient likelihood of confusion to merit preliminary injunctive relief. As an initial matter, the Ninth Circuit expressly agreed with the Second Circuit’s view that the use of a trademark as a search-engine keyword to trigger the display of a competitor’s advertisement was a “use in commerce” under the Lanham Act. The case thus turned on whether this activity was likely to cause consumer confusion. Network argued that its use of ACTIVEBATCH was legitimate “comparative, contextual advertising” that presented sophisticated consumers with clear choices. Systems argued that Network’s use of ACTIVEBATCH misled consumers by “hijacking their attention with intentionally unclear advertisements.”
The Ninth Circuit began its substantive analysis by clarifying that the eight traditional Sleekcraft likelihood-of-confusion factors are “not a rote checklist” to be applied rigidly. Rather, they were intended to be an “adaptable proxy for consumer confusion” such that courts have misinterpreted its emphasis of the “Internet Trinity” likelihood-of-confusion factors in its Brookfield decision as being the most relevant factors for all Internet-related infringement cases. The court noted that “[w]e did not intend Brookfield to be read so expansively as to forever enshrine these three factors . . . as the test for trademark infringement on the Internet.” According to the court, depending on the facts of each Internet-related case, other factors may be more “illuminating.” For example, the Ninth Circuit noted that while it used the “Internet Trinity” factors in Brookfield to analyze the likelihood of confusion generated by similar domain names, it did not emphasize these factors in its metatag analysis. The Ninth Circuit agreed with Professor McCarthy’s view that the “Internet Trinity” analysis was “appropriate” for domain-name disputes, but concluded that these factors were a “particularly poor fit” for keyword cases.
The Ninth Circuit reviewed all of the Sleekcraft likelihood-of-confusion factors and held that the most relevant factors for keyword-advertising cases were (1) strength of the mark, (2) evidence of actual confusion, (3) type of goods and degree of care likely to be exercised by the purchaser, and (4) labeling and appearance of the sponsored ads and the surrounding context on the search-results page. It also stated that when examining initial-interest confusion, the trademark owner “must demonstrate likely confusion, not mere diversion.”
Regarding the strength of the mark, the court stated that Internet users searching for a distinctive term are more likely to be looking for a particular product, and may be more susceptible to confusion when exposed to sponsored ads touting a similar product from a different source. It agreed with the district court that this factor favored Systems because ACTIVEBATCH was a suggestive, federally registered mark. The Ninth Circuit then held that the district court emphasized the proximity-of-the-goods factor too heavily simply because the parties were direct competitors. The proximity of the goods becomes less important when advertisements are clearly labeled and/or when consumers exercise a high degree of care, because in those cases, consumers would “merely be confronted with choices among similar products.” Accordingly, this factor must be considered in conjunction with an analysis of both the labeling and appearance of the sponsored ads and the degree of care exercised by Systems’ customers.
The Ninth Circuit came to the same conclusion for the similarity-of-the-marks factor, because consumers searching for ACTIVEBATCH did not see two distinct trademarks; rather, they saw a competitor’s sponsored ad that did not display either company’s mark. The labeling and appearance of the ads, including whether they identified Network’s own mark, and the sophistication of the consumers must also be examined to assess this factor’s weight.
As to marketing channels, the appeals court held that the parties’ shared use of a “ubiquitous marketing channel” (i.e., the Internet) did not shed much light on the likelihood of confusion, such that the district court’s weighing of this factor in Systems’ favor was incorrect. Regarding the type of goods and degree of care exercised by consumers, the Ninth Circuit noted that this factor is “highly relevant” to determining likelihood of confusion in keyword cases. According to the court, a “sophisticated consumer of business software exercising a high degree of care is more likely to understand the mechanics of Internet search engines and the nature of sponsored links, whereas an un-savvy consumer exercising less care is more likely to be confused.” The district court found that this factor favored Systems because Internet consumers “generally exercise a low degree of care,” but the Ninth Circuit disagreed and held that this proposition was out-of-date in most contexts. According to the Ninth Circuit, most Internet shoppers are now fairly sophisticated about the process, as online commerce is commonplace, and consumers searching for expensive software products online would likely be even more sophisticated. In particular, Internet consumers are accustomed to the “trial and error” nature of online shopping and “skip from site to site, ready to hit the back button whenever they’re not satisfied with a site’s contents,” and “don’t form any firm expectations about the sponsorship of a website until they’ve seen the landing page─if then.” It thus held that the district court improperly weighed this factor in favor of Systems.
Finally, the Ninth Circuit examined an additional factor not present in the Sleekcraft factors: the labeling and appearance of the sponsored ads and their surrounding context on the search-results page. The district court correctly examined the text of Network’s sponsored ads and concluded that the ads did not clearly identify their source, but it did not consider the surrounding context. In this regard, the Ninth Circuit referenced its decision in Playboy v. Netscape involving keyword-triggered banner ads, which found it important that (1) “clear labeling” of the banner ads might have eliminated the likelihood of initialinterest confusion, and (2) the search-results page did not “clearly segregate” the banner ads from the organic search results.
The Ninth Circuit contrasted this second point with the present case in which “Google and Bing have partitioned their search results pages so that the advertisements appear in separately labeled sections for ‘sponsored’ links.” The Ninth Circuit concluded that the district court did not weigh the likelihood-ofconfusion factors flexibly to match the specific facts of this keyword case, reversed the granting of the preliminary injunction, and remanded the case for further proceedings consistent with its opinion.
This decision is important because of the Ninth Circuit’s departure from its “Internet Trinity” likelihood-ofconfusion test and its creation of a new four-factor test for keyword cases. This decision is also notable for the Ninth Circuit’s clarification that trademark owners must demonstrate likely confusion, “not mere diversion,” when asserting infringement claims based on initial-interest confusion.