Businesses use several advertising tactics. Most are clearly presented to potential consumers in traditional ways such as text on a web page, posters, mailings, brochures, etc. Alternatively, some tactics to increase exposure to consumers using the internet to search for goods and services are clandestine.
For example, using words embedded in the website’s HTML code called “metatags” recognized by search engines to identify the website as relevant to a consumer’s search, the consumer does not see the HTML metatag. Another clandestine way is to bid on (pay for) advertising keywords from a search engine. In the latter circumstance, the keywords are not in the website’s HTML code or the website text. Rather, the search engine will list the purchaser’s website as an ad or popup near the top of searches triggered by the search containing those keywords entered by a consumer.
Purchased advertising keywords are not limited to generic descriptions of goods or services like “shoe” or “accounting.” They can include anything. Use of a competitor’s trademark within metatags or advertising keywords has, not surprisingly, generated claims that the practice constitutes unfair competition even though, arguably, the potential consumer does not see those keywords or metatags as used by the alleged infringer. See e.g., Brookfield Commc’ns., Inc. v. W. Coast Ent. Corp., 174 F.3d 1036 (9th Cir. 1999) (use of metatags comprising of a competitor’s trademark are actionable infringement under the Lanham Act although not encountered by the consumer in the marketplace); Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1145 (9th Cir. 2011) (en banc) (use of purchased advertising keywords is actionable infringement under the Lanham Act, “the use of a trademark as a search engine keyword that triggers the display of a competitor’s advertisement is a ‘use in commerce’ under the Lanham Act.” 638 F.3d at 1145–46).
Such an advertising strategy is called “conquesting.” Lerner & Rowe, at 4-5.
CONSUMER CONFUSION IN TRADEMARK INFRINGEMENT CASE LAW
The “core element of trademark infringement is the likelihood of confusion, i.e., whether the similarity of the marks is likely to confuse customers about the source of the products.” E. & J. Gallo Winery v. Gallo Cattle Co., 967 F.2d 1280, 1290 (9th Cir. 1982). Generally, the focus is upon “whether a reasonably prudent consumer in the marketplace is likely to be confused as to the origin of the goods or service bearing one of the marks.” Rearden LLC v. Rearden Commerce, Inc., 683 F.3d 1190, 1214 (9th Cir. 2012) (quotations and citations omitted). The Ninth Circuit established non-exclusive factors to analyze the likelihood of confusion in AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 348–49 (9th Cir. 1979).
There are circumstances where the Sleekcraft test is inconclusive, particularly where the defendant is using the same mark (at least the words) as the putative plaintiff in comparative advertising circumstances: nominative fair use. See e.g.New Kids on the Block v. News Am. Pub., Inc., 971 F.2d 302, 308-309 (9th Cir. 1982) (“nominative use of a mark—where the only word reasonably available to describe a particular thing is pressed into service—lies outside the strictures of trademark law … such use is fair because it does not imply sponsorship or endorsement by the trademark holder;” adopting a three-factor test to determine if the use of the mark is fair under the circumstances). The Ninth Circuit has explained that the Sleekcraft test is insufficient in the nominative fair use context:
The three-factor test better evaluates the likelihood of confusion in nominative use cases. When a defendant uses a trademark nominally, the trademark will be identical to the plaintiff’s mark, at least in terms of the words in question. Thus, application of the Sleekcraft test, which focuses on the similarity of the mark used by the plaintiff and the defendant, would lead to the incorrect conclusion that virtually all nominative uses are confusing. The three-factor test—with its requirements that the defendant use marks only when no descriptive substitute exists, use no more of the mark than necessary, and do nothing to suggest sponsorship or endorsement by the mark holder—better addresses concerns regarding the likelihood of confusion in nominative use cases.
Playboy Enterprises, Inc. v. Welles, 279 F.3d 796, 801 (9th Cir. 2002).
In the nominative fair use context, the consumer is still encountering the use of the alleged infringed trademark in the marketplace. It is not a clandestine use.
In the clandestine keyword advertising context, the consumer is not encountering the putative defendant’s use of the plaintiff’s trademark. Rather, the consumer sees each competitor’s separate marks. See Lerner & Rowe PC v. Brown Engstrand & Shely, LLC, 119 F.4th 711, 726 (9th Cir. 2024) (when examining the similarity of the marks displayed to the consumer, because the consumer does not see ALG’s use of “Lerner & Rowe”, the court concluded the only marks displayed were clearly dissimilar “‘Lerner & Rowe’ and ‘Accident Law Group’”, supporting a finding of lack of confusion). Therefore there is an adjustment to the Sleekcraft factors, focusing primarily upon four elements in the keyword advertising context: “(1) the strength of the mark; (2) the evidence of actual confusion; (3) the type of goods and degree of care likely to be exercised by the purchaser; and (4) the labeling and appearance of the advertisements and the surrounding context on the screen displaying the results page.” Network Automation, Inc., 638 F.3d at 1154.
RECENT CASE IMPLICATES ‘INITIAL INTEREST CONFUSION’
In a recent keyword advertising case, the Ninth Circuit affirmed the grant of summary judgment against a plaintiff, affirming a finding no likelihood of confusion. Lerner & Rowe, 119 F.4th 711. Two law firms advertised on the internet for consumers interested in legal services related to personal injury claims. Plaintiff Lerner & Rowe PC contended that Brown, Engstrand & Shely, LLC (dba The Accident Law Group (“ALG”)) infringed upon the federally registered trademark “Lerner & Rowe” by ALG’s purchasing from a search engine the term “Lerner & Rowe” as advertising keywords. As a result, ALG’s advertisements were displayed, preceded by the indication “Ad,” when a consumer used the search term “Lerner & Rowe.”
Not able to claim that consumers actually saw ALG’s use of the trademark in the marketplace, the plaintiff resorted to a theory of “initial interest confusion.” Initial interest confusion occurs when one uses a competitor’s trademark as a vehicle to direct a consumer’s attention to its product instead of its competitor’s. Playboy Enters., Inc. v. Netscape Commc’ns Corp., 354 F.3d 1020, 1025 (9th Cir. 2004) (“Although dispelled before an actual sale occurs, initial interest confusion impermissibly capitalizes on the goodwill associated with a mark and is therefore actionable trademark infringement.”). The Ninth Circuit admonished that the initial interest confusion form of infringement only applies to “misleading and deceptive” uses of a trademark and not to “legitimate comparative and contextual advertising.” Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1148 (9th Cir. 2011) (en banc). The owner of a trademark must always “demonstrate likely confusion, not mere diversion” of the consumer to prevail. Id. at 1149. As Judge Marsha Berzon stated in her concurring opinion in Playboy Enters., 354 F.3d at 1035, “[t]here is a big difference between hijacking a customer to another website by making the customer think he or she is visiting the trademark holder’s website (even if only briefly), … and just distracting a potential customer with another choice, when it is clear that it is a choice.”
The Ninth Circuit panel affirmed judgment against Lerner & Rowe, noting that although there was evidence of 236 instances of actual confusion caused by the listing of ALG’s advertisements when searching “Lerner & Rowe,” the confusion was de minimis given the total volume of exposure to 109,322 potential customers. The panel stated that “if a plaintiff can demonstrate that an appreciable number of people are confused, that fact, alone, might entitle the plaintiff to a trial on the likelihood of confusion” and preclude a defendant’s motion for summary judgment. Lerner & Rowe, 119 F.4th at 719 (quotations and citations omitted). However, the evidence proffered indicated merely 0.216% of the total number of those who searched for “Lerner & Rowe” on the search engine and were exposed to the challenged advertisements were confused. Additionally, the evidence showed that only a mere 6.82% of the users who encountered ALG’s ads under the circumstances clicked through to ALG’s website. Lerner & Rowe, 119 F.4th at 719-20; cf. 1-800 Contacts, Inc. v. Lens.com, Inc., 722 F.3d 1229, 1244 (10th Cir. 2013) (a click-through rate might represent an upper limit of initial interest confusion).
CALCULATING ACTUAL CONFUSION
The Ninth Circuit has previously admonished district courts that even one or two instances of actual confusion should weigh in favor of denying a defendants’ motion for summary judgment. Ironhawk Technologies, Inc. v. Dropbox, Inc., 2 F.4th 1150, 1166 (9th Cir. 2021). Although the number of instances of actual confusion was 236, the Lerner & Rowe Court distinguished the two instances that prevented summary judgment in Ironhawk from the evidence presented in Lerner & Rowe. The court was persuaded by not only the number of consumers actually confused, but also the number of consumers actually exposed to the accused advertisements published in the list of search results: “236 calls [to ALG] representing actual confusion … and … 109,322 consumers who saw the advertisements.” Lerner & Rowe 119 F.4th at 720. The court was persuaded that the percentage of those exposed to the ads as a result of the purchase of the keywords and who could be categorized as confused resulted in a mere 0.216%, concluding that “[n]o reasonable jury would conclude that this percentage is anything but de minimis and fails to support a finding of likelihood of confusion.” Lerner & Rowe, 119 F.4th at 720.
The Ninth Circuit was quick to state that courts should not “automatically discount de minimis instances of actual confusion when the record contains additional evidence of consumer confusion.” Lerner & Rowe, 119 F.4th at 721, contrasting the case of Rosetta Stone Ltd. v. Google, Inc., 676 F.3d 144 (4th Cir. 2012) (only five confused consumers from over 100,000 opportunities for confusion, coupled with 262 consumer complaints, in-house Google studies about the defendant’s advertising study was likely confusing, in-house Google trademark attorneys being confused; and an expert survey concluding there was a net confusion rate of 17%.).
Even in light of the evidence of 236 instances of actual confusion, the court was persuaded that the relevant consumers were not easily confused under the circumstances. “[T]he relevant consumers specifically typed in ‘Lerner & Rowe’ as a search term, suggesting that they would be even more discerning of the results they received.” Lerner & Rowe, 119 F.4th at 722. The court concluded that the labeling and appearance of ALG’s advertisements “would not confuse a reasonably prudent consumer searching online for personal injury legal services,” stating: “we find it difficult to believe that consumers searching for the phrase ‘Lerner & Rowe’ would not choose to click on the link that matches their search query word for word.” Lerner & Rowe, 119 F.4th at 724. The court also found that the display of the advertisements and surrounding context on the screen displaying the results weighed against Lerner & Rowe.
The court affirmed the grant of summary judgment against Lerner & Rowe, finding examination of all of the Network Automation, Inc. four principal factors, as well as numerous others, tipped the scales against Lerner & Rowe despite evidence of 236 instances of actual confusion under the circumstances.
REVISITING ‘USE IN COMMERCE’
Judge Roopali H. Desai submitted a concurring opinion, suggesting that the Ninth Circuit en banc decision Network Automation, Inc., should be reexamined in that clandestine use of purchased keywords in internet marketing might not properly be construed as “use in commerce,” as required for application of the Lanham Act. Judge Desai noted that use in commerce is defined in the Lanham Act as, “used or displayed in the sale or advertising of services,” 15 U.S.C. Section 1127, but in the purchased keywords context the consumer does not see the actual use by the putative defendant. In footnote seven of the majority panel opinion, the court indicated discomfort with Network Automation, but that it is required to follow the en banc precedent, stating: “ALG alternatively asks us to affirm the district court’s grant of summary judgment on the ground that ALG never used Lerner & Rowe’s trademark in commerce. Network Automation, however, explicitly held that ‘the use of a trademark as a search engine keyword that triggers the display of a competitor’s advertisement is a ‘use in commerce’ under the Lanham Act.’ 638 F.3d at 1145–46. Because no intervening Supreme Court decision is ‘clearly irreconcilable’ with this holding, we have no power to overrule it. Miller v. Gammie, 335 F.3d 889, 900 (9th Cir. 2003) (en banc).” Lerner & Rowe, 119 F.4th n. 7 at 726 (emphasis added).
NEXT STEPS FOR BUSINESSES
Businesses should include in their program of trademark enforcement an examination as to whether competitors may be using conquesting strategies, and take appropriate action, if any. Conversely, businesses advertising on the internet should be diligent in avoiding potential exposure to infringement suits by the clandestine use of a competitor’s trademarks—although not per se infringement, the practice presents risks.