CONFUSION MÉLANGE: A CALL TO REVISE THE LAW ON TRADEMARK INFRINGEMENT, INITIAL INTEREST CONFUSION, AND DIGITAL MARKETING JOHNNY WALSH* TABLE OF CONTENTS Introduction I. Background A. Internet Marketing, AdWords, and Cookies B. Trademarks, the Lanham Act, and the Internet 1. Congressional Intent Behind the Protection of Trademarks 2. The Lanham Act 3. Secondary Splits on Initial Interest Confusion II. The Issues A. The Inefficiencies of the Multifactor Tests for Likelihood of Confusion B. Circuit Court Splits on Initial Interest Confusion C. Increasing Breadth, Intricacy, and Cost of Trademark Litigation III. Proposed Solutions A. Standardize Initial Interest Confusion 1. Bar Initial Interest Confusion if the Parties’ Products are Unrelated 2. Allow Initial Interest Confusion Regardless of Brevity of Confusion or Disclaimer Usage 3. Bar Initial Interest Confusion if Consumers are Sophisticated or Careful 4. Do Not Assume Consumers on the Internet to be Sophisticated or Careful 5. Do Not Assume Consumers of Expensive Products to be Sophisticated or Careful B. Abridge the Multifactor Likelihood of Confusion Tests in Digital Marketing Cases C. Replace the Multifactor Likelihood of Confusion Tests for Digital Marketing Cases Conclusion INTRODUCTION On February 7, 2022, Yasha received word that the pending summary judgment motion against him had been decided.1 His years-long litigation with a former roommate had been a conundrum: a case that involved personal history but which had played out through the decidedly impersonal medium of internet commerce.2 As if sprung from the most cliché business drama, it all started with * J.D. Candidate, 2024, Indiana University Robert H. McKinney School of Law; B.S. 2016, Purdue University. 1. Boost Beauty, LLC v. Woo Signatures, LLC, No. 2:18-CV-02960-CAS-EX, 2022 WL 409957 at *11 (C.D. Cal. Feb. 7, 2022). 2. See id. at *2. 282 INDIANA LAW REVIEW [Vol. 57:281 a simple idea: to purchase a generic “eyelash enhancement” from a third-party vendor, relabel it, and then sell it directly to consumers over the internet.3 Simple though it was, this idea proved successful enough that Yasha shortly made it his full-time job; it was inevitable that his roommate, Arash, would take notice.4 The two still shared an apartment when Arash began his competing business, and he was not shy about the inspiration: the identical products had similar names—Arash having chosen “WooLash” to compete with Yasha’s original “BoostLash”—and sold for the same price.5 There was even evidence that Arash had paid to have “WooLash” advertisements shown in the Google search results for “BoostLash.”6 When Arash moved for summary judgment on Yasha’s trademark infringement claim, the decision came down to whether Yasha’s potential consumers were “likely to be confused” by the competing products’ similar branding and characteristics.7 The Ninth Circuit Court of Appeals has outlined eight factors for considering whether such confusion is likely; the District Court provided serious analysis of just four.8 First, the Court concluded that Yasha’s “BoostLash” was a weak trademark whose “obviousness” meant it was entitled to less stringent protection.9 Although the Court’s conclusion was correct—the trademark itself simply described its stated purpose of enhancing its users’ eyelashes—it was, in large part, arrived at on the back of specious evidence regarding the number of other cosmetic trademarks bearing the components “Boost” or “Lash”;10 many trademark scholars do not rank such evidence among those required to determine trademark strength.1 1 Second, the Court found the lack of “evidence of actual confusion” to weigh against plaintiff Yasha, despite precedent that such evidence merely “provides strong support for the likelihood of confusion” when a plaintiff 3. Id. 4. Id. (defendants claim plaintiff’s business was conducted “in the open” of a shared apartment). 5. Id. at *3. 6. Id. at *4, *12. 7. Id. at *9. 8. Id. at *9-11. 9. Id. at *10. 10. Id. 11. See 5 J. THOMAS MCCARTHY, MCCARTHY ON TRADEMARKS & UNFAIR COMPETITION § 11:81 (5th ed. 2023) (listing “amount of third party use, amount and type of advertising and promotion which draws attention to the mark, amount of sales and number and type of customers, recognition by the trade, by the media and by potential customers, survey of likely customers showing their recognition of the mark” as appropriate evidence for this factor); see also 3 MARY M. SQUYRES & NANETTE NORTON, TRADEMARK PRACTICE THROUGHOUT THE WORLD § 26:19 (3rd ed. 2022) (listing “History and meaning of the mark; Advertising and sales figures; Unsolicited publicity about the goods/services; Lack of third party use; Consumer correspondence referring to the goods/services; Affidavits from people in the industry regarding recognition of the mark” as appropriate evidence). 2023] CONFUSION MÉLANGE 283 does demonstrate it.12 Third, the “degree of consumer care” factor was credited to defendant Arash because purchasers of obscure brands of “non-medicated eyelash growth serum” were considered sophisticated, scrutinizing consumers in the eyes of the Court,1 3 despite the fact that even rudimentary research on such products should cause severe doubt as to their effectiveness.14 Fourth, the Court explained that the “similarity of the marks” factor weighed against plaintiff Yasha’s infringement claim because, if one cleaves the products’ names in half and ignores “the portion . . . that refer[s] to ‘lash’,” it is quite easy to conclude that “BoostLash” and “WooLash” are “not similar.”1 5 Although Yasha had also argued that other factors “create[d] genuine issues of material fact,” the Court curtly dismissed these as “not meaningfully factor[ing] into the analysis.”16 With that, in just eighteen pages, Yasha’s four-year pursuit of a remedy had evaporated.17 In a vacuum, the Boost Beauty decision might merely be considered an oddity. But imagine if the case had arisen in the Eighth Circuit, where precedent dictates that consumer sophistication and degree of care should scarcely be settled at the summary judgment stage.18 Maybe the plaintiff’s claim would also have survived summary judgment in the Tenth Circuit, where courts consider internet shoppers to have a relatively low degree of care and be easily swayed.19 Or consider the Second Circuit, where the plaintiff’s case would almost certainly have been dismissed early on for failing to meet that jurisdiction’s unique requirement for “intentional deception” in such cases.20 Despite the existence of federal-level trademark protection,21 federal courts have created a body of law that may afford a trademark owner protection against certain tactics in one Circuit’s jurisdiction, yet render them completely powerless against the same 12. Id. at *11 (internal quotation marks omitted). 13. Id. at *3, *11. 14. See Roberta Schroeder, The Eyelash Growth Serums That Really Work, HARPER’S BAZAAR (Jan. 25, 2023), https://www.harpersbazaar.com/uk/beauty/skincare/g40386701/lash-growth-serum/ [https://perma.cc/TZC2-BKEM] (“Independent studies into the effectiveness of eyelash growth serums are sparse” and most supporting evidence is purely anecdotal). 15. Boost Beauty, LLC v. Woo Signatures, LLC, No. 2:18-CV-02960-CAS-EX, 2022 WL 409957 at *11 (C.D. Cal. Feb. 7, 2022). 16. Id. at *10-11. 17. See id. at *13 (defendant’s motions for summary judgment granted on plaintiff’s fourth and sixth through eleventh claims). 18. Select Comfort Corp. v. Baxter, 996 F.3d 925, 937 (8th Cir. 2021) (“[A] jury rather than a judge should assess the level of consumer sophistication.”). 19. SanMedica Int’l, LLC v. Amazon.com, Inc., No. 2:13-cv-00169-DN, 2016 WL 527055, at *8 (D. Utah Jan. 20, 2016) (finding that an internet shopper “misled into an initial interest, . . . may satisfy him or herself that the other related products . . . are at least as good”). 20. Alzheimer’s Disease & Related Disorders Ass’n v. Alzheimer’s Found. of Am., Inc., 307 F. Supp. 3d 260, 286 (S.D.N.Y. 2018) (stating that “the Second Circuit has also expected a plaintiff to prove ‘intentional deception’” in internet trademark infringement cases). 21. 15 U.S.C. §§ 1114, 1125. 284 INDIANA LAW REVIEW [Vol. 57:281 tactics in another jurisdiction. The principle of stare decisis is meant to promote “evenhandedness, predictability, and consistency” in the law,22 but the Federal Circuits’ various precedents on trademark infringement—particularly on the internet—have increasingly fostered unpredictability and inconsistency. Given that the use of trademarks on the internet can be expected to increase in line with the predicted growth of e-commerce,23 it would be beneficial for federal courts to attempt to cure the many ailments currently afflicting trademark law. This Note argues that the current state of trademark infringement litigation, although ostensibly still grounded in its original Congressional intent, has become contorted into an overwrought, expensive, and inconsistent mess. This Note further argues that these issues may be addressed by simplifying the rules and doctrines that govern this type of litigation, particularly in regard to digital marketing. Part I of this Note addresses the background and current state of digital marketing, the motivation behind trademark protection, the required elements for a claim of infringement, and the Circuit splits that have developed around the controversial doctrine in trademark law, initial interest confusion. Part II explains how a combination of the Circuits’ decades-old tests and the disorder of the relatively new initial interest confusion doctrine have led to irregular and potentially unfair litigation. Part III then states a number of means by which these issues may be remedied, either in concert or in the alternative: standardizing the initial interest confusion doctrine; abridging the tests for likelihood of confusion in digital marketing cases; and replacing the tests for likelihood of confusion in digital marketing cases. I. BACKGROUND A. Internet Marketing, AdWords, and Cookies Digital marketing is the use of the internet to increase awareness of a brand and connect that brand to relevant consumers.24 It is a broad field that ranges from pop-up advertisements to the use of social media influencers.25 This Note will only cover the two most common forms of digital marketing: the first—search engine marketing—is one of the oldest and most established; the second—online behavioral advertising—is one of the newest and most controversial. 22. Johnson v. United States, 576 U.S. 591, 592 (2015). 23. Here’s Why E-Commerce Growth Can Stay Stronger for Longer, MORGAN STANLEY (June 14, 2022), https://www.morganstanley.com/ideas/global-ecommerce-growth-forecast-2022 [https://perma.cc/SUT7-UQXU] (e-commerce currently makes up twenty-three percent of sales in the U.S. and could reach thirty-one percent by 2026). 24. Digital Marketing, AM. MKTG. ASS’N, https://www.ama.org/topics/digital-marketing/ [https://perma.cc/N6AG-5C59] (last visited Nov. 11, 2022). 25. See Indeed Editorial Team, What Is Pop-up Marketing and How Do You Use It?, INDEED (May 3, 2021), https://www.indeed.com/career-advice/career-development/pop-up-marketing [https://perma.cc/NVE8-3G5K]; Adam Hayes, Social Media Marketing (SMM): What It Is, How It Works, Pros and Cons, INVESTOPEDIA (Oct. 22, 2022), https://www.investopedia.com/terms/s/social- media-marketing-smm.asp [https://perma.cc/42RS-UW3E]. 2023] CONFUSION MÉLANGE 285 In simplest terms, search engine marketing (“SEM”) is the monetization of internet search results; a business that is willing to pay a search engine more will have its advertisements appear more prominently in the search engine’s results pages.26 This concept is made possible by “AdWords.”27 At their base, the search queries that users enter into search engines are merely combinations of words and phrases.28 Search engine providers may turn these words and phrases into AdWords by auctioning them off to advertisers, whose bids represent their willingness to pay for a single user interaction on their advertisement, also called a “click.”29 When a user runs a search query, the search engine generates “organic” results determined to be relevant by the engine’s algorithm.30 These organic results will also be intermixed with “sponsored links”— advertisements for the AdWords in the query—ranked in part by the amount each advertiser has bid.31 A click on any one of these sponsored links will take the user to that advertiser’s “landing page.”32 In 2004, Google became one of the first search engine providers to allow trademarks to be sold as AdWords in its SEM system.3 3 This decision proved successful, and has been followed by Microsoft and Amazon Marketplace.34 This open approach to trademark usage has made it common practice for online businesses to purchase their competitors’ trademarks as AdWords in the hopes of diverting the competitors’ potential customers.35 Many businesses go one step 26. Leeron Hoory & Kelly Main, The Ultimate Guide To Search Engine Marketing (SEM) in 2022, FORBES (Aug. 17, 2022), https://www.forbes.com/advisor/business/search-engine-marketing- sem/ [https://perma.cc/S9BA-SZUT]. 27. Kimberly Vogel & Cindy McCaffrey, Google Launches Self-Service Advertising Program, GOOGLE (Oct. 23, 2000), http://googlepress.blogspot.com/2000/10/google-launches-self- service.html [https://perma.cc/ATR9-EFJL]. 28. Hoory & Main, supra note 26. 29. Id. 30. Id. 31. Tony Tran, A Beginner’s Guide to Using Google Ads (Previously Google Adwords), HOOTSUITE (Sep. 17, 2020), https://blog.hootsuite.com/google-ads/ [https://perma.cc/PD4C-3G9J]. 32. Id. 33. Rosetta Stone Ltd. v. Google, Inc., 676 F.3d 144, 151 (4th Cir. 2012). 34. Compare Grow Your Business With Google Ads, GOOGLE, https://support.google.com/ google-ads/answer/6336021 [https://perma.cc/3V2K-NCMJ] (last visited Oct. 7, 2022), with What Is Search Engine Marketing?, MICROSOFT, https://about.ads.microsoft.com/en-us/get-started/search- engine-marketing-basics [https://perma.cc/S6LS-ZSNZ] (last visited Oct. 7, 2022), and How Does Bidding Work For Sponsored Products?, AMAZON ADS, https://advertising.amazon.com/library/ videos/campaign-bidding-sponsored-products [https://perma.cc/Q3RE-27M5] (last visited Oct. 21, 2022). 35. See Tempur-Pedic North America, LLC v. Mattress Firm, Inc., No. CV H-17-1068, 2017 WL 2957912, at *3 (S.D. Tex. July 11, 2017) (case concerning online mattress retailers’ purchase of each other’s trademarks as AdWords); Alzheimer’s Disease & Related Disorders Ass’n v. Alzheimer’s Found. of Am., Inc., 307 F. Supp. 3d 260, 284 (S.D.N.Y. 2018) (similar case between Alzheimer’s charities); 1-800 Contacts, Inc. v. Lens.com, Inc., 722 F.3d 1229, 1234 (10th Cir. 2013) 286 INDIANA LAW REVIEW [Vol. 57:281 further: crafting their sponsored links to conceal their identity or imitate their competitors to further increase the chance of generating clicks.36 While SEM could perhaps best be expressed as the application of a user’s current activity in order to display advertisements that are likely relevant, online behavioral advertising (“OBA”) has the goal of displaying advertisements that are foreseeably relevant.37 This goal is pursued—and, more often than not, accomplished—using “cookies”: small text files embedded in a user’s web browser that contain data on the user’s browsing history, recent purchases, demographic information, and more.38 Traditionally, a website uses “first party” cookies—which track activity on that website—and “third-party” cookies—which track activity on other websites—to create individualized user profiles, which it may then utilize to direct specific advertisements and product recommendations toward that user.39 Most often, these cookie data are put to use in OBA systems through artificial intelligence (“AI”) software, which generates predictions on what a user ‘wants’ to see by combining that user’s first-party cookie profile with the website’s accumulated data on user behavior.40 An OBA system will consistently produce recommendations and advertisements with a high degree of effectiveness; such recommendations account for about thirty percent of product interactions on the Amazon Marketplace.41 However, many trademark law scholars have little doubt that AI-assisted OBA systems are capable of misleading consumers with recommendations to infringing or counterfeit products or being manipulated by infringers into doing so.42 Regardless of whether it is SEM or OBA, digital marketing systems are only becoming more advanced and consumer-specialized. By tailoring the sponsored links and recommendations that are shown to users, digital marketing providers (similar case between online retailers of contact lenses). 36. See Edible Arrangments, LLC v. Provide Com., Inc., No. 3:14-CV-00250 (VLB), 2016 WL 4074121, at *3 (D. Conn. July 29, 2016) (dispute over the use of “edible arrangements” as AdWord and sponsored link text by a competitor of a famous brand); Edina Realty, Inc., v. TheMLSonline.com, No. Civ. 04-4371JRTFLN, 2006 WL 737064 at *3 (D. Minn. Mar. 20, 2006) (identical fact pattern between real estate brokerages); Jim S. Adler, P.C. v. McNeil Consultants, L.L.C., 10 F.4th 422 (5th Cir. 2021) (similar fact pattern between attorneys). 37. Steven C. Bennett, Regulating Online Behavioral Advertising, 44 J. MARSHALL L. REV. 899 (2011). 38. In re Facebook Internet Tracking Litig., 140 F. Supp. 3d 922, 926 (N.D. Cal. 2015); Klein v. Facebook, Inc., 580 F. Supp. 3d 743, 792 (N.D. Cal. 2022). 39. Bennett, supra note 37. 40. Brent Smith & Greg Linden, Two Decades of Recommender Systems at Amazon.com, IEEE INTERNET COMPUTING, May/June 2017, at 12-18, https://assets.amazon.science/76/9e/7eac89c14 a 8 3 8 7 4 6 e 9 1 d d e 0 a 5 e 9 f / t w o - d e c a d e s - o f - r e c o m m e n d e r - s y s t e m s - a t - a m a z o n . p d f [https://perma.cc/KGA2-CNUZ]. 41. Id. at 13-14. 42. Jeff Rojek et al., Comments from the Center for Anti-Counterfeiting and Product Protection, USPTO (Dec. 16, 2019), https://www.uspto.gov/sites/default/files/documents/Jeffrey- Rojek_RFC-84-FR-58141.pdf [https://perma.cc/6XV8-A4XK]. 2023] CONFUSION MÉLANGE 287 have created an incredibly competitive area of marketing in which it is a near- universal practice for competitors to use one another’s trademarks. The issue of digital marketing technology has always been determining the point at which this laissez-faire use of trademarks crosses over into impropriety. However, it would be impossible to discern whether any use of trademarks is improper without first laying the foundation about what protections trademarks are actually entitled to. B. Trademarks, the Lanham Act, and the Internet 1. Congressional Intent Behind the Protection of Trademarks.—The position of many scholars is that the primary motivation behind trademark law is to implement consumer protection safeguards by incentivizing the creation of better, safer products and punishing the use of dishonest sales and advertising practices.43 The idea is that the owner of a trademark, having been granted a monopoly on its use, will make greater efforts to better her brand’s image by investing in quality control and product improvements.44 Furthermore, by punishing businesses who misleadingly use another’s trademarks, the law also seeks to prevent consumers from being deceived or manipulated.45 As an added bonus, trademarks also enable consumers to exert less effort deliberating on purchase decisions, as the respective strengths of certain brands may be easier to weigh than those of individual products.46 In addition to this consumer-focused view, however, conventional wisdom and scholarly consensus alike conclude that trademark law benefits trademark- owning producers at least as much as it does consumers. Labeling products with a trademark allows businesses to accumulate goodwill with customers,47 who will in turn “reward [the trademark owner] with continued patronage.”48 In this way, trademarks act as a form of property into which the owner makes value- improving investments.49 Trademark law can therefore be viewed as protection for producers against harm to their investment and “illegitimate diversions of their trade.”50 Today, scholars generally agree that the consumer-focused view and producer-focused view are of equal importance in trademark law.51 Any 43. See MCCARTHY, supra note 11, at § 27:25; Pratheepan Gulasekaram, Policing the Border Between Trademarks and Free Speech: Protecting Unauthorized Trademark Use in Expressive Works, 80 WASH. L. REV. 887, 895 (2005) (“The underlying tenet of the Lanham Act is consumer protection.”). 44. MCCARTHY, supra note 11, at § 2:3. 45. Id. at § 2:9. 46. Id. at § 2:5. 47. Id. at § 3:2. 48. Smith v. Chanel, Inc., 402 F.2d 562, 566 (9th Cir. 1968). 49. Trademark Law Revision Act of 1988, S.1883, 100th Cong. (1988). 50. Mark P. McKenna, The Normative Foundations of Trademark Law, 82 NOTRE DAME L. REV. 1839, 1848 (2007); see MCCARTHY, supra note 11, at § 5:2 (“When the owner of a trademark has spent considerable time and money bringing a product to the marketplace, trademark law protects the producer from pirates and counterfeiters.”). 51. See Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in 288 INDIANA LAW REVIEW [Vol. 57:281 normative statement on how trademark law ought to be shaped must therefore take into account these dual goals: protecting consumers from being confused or deceived, and producers from having their brands devalued or misappropriated. 2. The Lanham Act.—Even if the motivation behind their protection is known, an answer to the question of what may classify as a trademark remains somewhat nebulous. Subject to a few limitations,52 a trademark can be nearly anything that serves the purpose of identifying or distinguishing its owner’s particular product offering from competing product offerings.5 3 Although trademark law in the United States existed as a collection of state-level schemas for the first century and a half of the country’s history, calls for standardization and national regulation of trademarks began to be made in the 20th century.54 To confront these concerns, Congress passed the Lanham Act in 1946.55 Beyond implementing the means for registering trademarks nationally, the Lanham Act also established two causes of action for infringing use of a trademark: Section 32, Subsection 1 (“Section 32(1)”),56 and Section 43, Subsection a (“Section 43(a)”).57 Section 32(1) concerns the infringement of federally registered trademarks, the required elements for which are: (1) that the mark allegedly infringed has been properly registered with the United States Patent and Trademark Office; (2) that the defendant made a “use in commerce” of either the mark itself or a “reproduc[tion], counterfeit, copy, or colorabl[e] imitat[ion]” of the mark; (3) that the defendant’s use was made “without the consent of the registrant”; (4) that the defendant’s use was “in connection with the sale, offering for sale, distribution, or advertising” of a good or service; and (5) that the defendant’s use was “likely to cause confusion, or to cause mistake, or to deceive . . . .”58 Although these elements are relatively few in number, they are steeped in redundancy: it is a decidedly short list of activities that are “in commerce” but nevertheless fall completely outside the purview of selling, offering, distributing, or advertising, and one struggles to imagine how a trademark is likely to be confused with Trademark Law, 86 B.U. L. REV. 547, n.59 (2006) (“[B]oth perspectives are necessary to fully appreciate the whole [of trademark law].”); MCCARTHY, supra note 11 at § 2:2 (recognizing “the dual goals of trademark law”). 52. 7 RUDOLF CALLMANN ET AL., CALLMANN ON UNFAIR COMPETITION, TRADEMARKS & MONOPOLIES § 26:11 (4th ed. 2022) (marks are unregistrable if they are “merely descriptive” or “primarily merely a surname,” as well as if they “disparage or falsely suggest a connection with persons, . . . institutions, beliefs, national symbols,” as well as in other limited scenarios). 53. See 15 U.S.C. § 1127 (“The term ‘trademark’ includes any word, name, symbol, or device, or any combination thereof . . . used by a person, or . . . which a person has a bona fide intention to use . . . to identify and distinguish his or her goods, including a unique product, from those manufactured or sold by others.”). 54. MCCARTHY, supra note 11 at § 5:3. 55. 15 U.S.C. §§ 1051-1141. 56. Id. § 1114(1). 57. Id. § 1125(a). 58. Id. § 1114(1). 2023] CONFUSION MÉLANGE 289 another without at least being its colorable imitator. If a registrant inadvisably brings an infringement suit against a party to whom they have given consent to use their trademark, then that consent may of course be used to their detriment;59 for obvious reasons, litigation in such cases is rarely brought and usually quickly resolved. A plaintiff suing for trademark infringement under the Lanham Act is not even held to the requirement for federal registration: Section 43(a) offers an alternative cause of action for the infringement of any trademark, registered or unregistered, as long as it can be shown that the trademark is a valid identifier of plaintiff’s products, that the defendant used the trademark in commerce, and that the use was likely to cause confusion or a false association with the plaintiff.60 In this way, a defendant facing claims of trademark infringement under both Section 32(1) and Section 43(a) might not even bother contesting the plaintiff’s claim of federal registration status; even if it can be shown that the registration was gained through bold-faced fraud, the plaintiff’s case may continue practically unabated.61 The only difference being that plaintiff must now hurdle the low barrier of showing that her trademark is valid.62 For this reason, the majority of cases of trademark infringement in federal court will naturally boil down to the two factors common to both causes of action: whether the allegedly infringed trademark was used in commerce and whether consumers were likely to be confused by that use. a. Requirement for use in commerce.—The requirement under Sections 32(a) and 43(1) for infringing activity to be a “use in commerce” of the trademark exists to limit the Lanham Act’s causes of action to “bona fide use[s] of a mark in the ordinary course of trade.”63 Suppliers and service providers—including those whose platforms and software are used by others on the internet—may also be found to have “contributed” to another party’s improper use of a trademark if they “intentionally induce[d] another to infringe a trademark, or . . . continue[d]to supply its product to one whom it knows or has reason to know is engaging in trademark infringement,” and as a result may be liable for “contributory” infringement.64 In the context of the internet, federal courts have found that 59. See MCCARTHY, supra note 11, at § 31:41 (stating that a defense exists “where the trademark owner has by affirmative word or action conveyed to the infringer the message that its acts are not objectionable”). 60. 15 U.S.C. § 1125(a)(1). (“Any person who . . . uses in commerce any word, term, name, symbol, or device . . . likely to cause confusion, or to cause mistake, or to deceive as to the affiliation, connection, or association of the defendant with the plaintiff.”). 61. See MCCARTHY, supra note 11, at § 31:60 (“It has been held several times that even if [a] defendant succeeds in proving that the plaintiff’s registration was fraudulently obtained, plaintiff’s common law rights in the mark continue unabated.”). 62. See supra note 53 and accompanying text. 63. Id. § 1127. 64. Inwood Lab’ys, Inc. v. Ives Lab’ys, Inc., 456 U.S. 844, 854 (1982); see also Gov’t Emps. Ins. v. Google, Inc., 330 F. Supp. 2d 700, 705 (E.D. Va. 2004) (holding that a search engine provider “monitor[ing] and control[ing]” its SEM system showed sufficient “knowledge required to allege contributory infringement”); but see Tiffany (NJ) Inc. v. eBay Inc., 600 F.3d 93, 107 (2d Cir. 2010) 290 INDIANA LAW REVIEW [Vol. 57:281 individual users and sites may meet the Lanham Act’s use in commerce requirement by writing another’s trademark into a website’s metatags,65 placing another’s trademark in a domain name,66 and purchasing or selling another’s trademark as an AdWord.67 b. Requirement for a likelihood of consumer confusion.—Once it has been determined that the alleged infringement is a use in commerce, the infringement inquiry turns to whether that use was likely to cause consumer confusion, a question that federal courts answer by applying balancing tests consisting of between seven and thirteen factors that weigh either in favor of or against the probability that the use of a trademark created confusion.68 The Circuits’ respective multifactor tests vary slightly, but each one includes the “similarity of the marks,” “proximity of the goods,” “evidence of actual confusion,” and “strength of the mark” factors, and almost all also include the “intent of the defendant” and “consumer sophistication” factors.69 The various tests contain some combination of other factors as well; for instance, the Second Circuit’s Polaroid factors take under consideration “the defendant’s good faith,” “the quality of defendant’s product,” and “the likelihood that the prior owner will bridge the gap.”70 Notably, each of the Circuits also stress that likelihood of confusion cannot be concluded based on any single factor.71 Analysis may not begin yet, however, as presiding courts must be sure to limit any likelihood-of-confusion analysis to the points in the purchasing process that its jurisdiction has deemed relevant. The original text of the Lanham Act limited analysis to “purchasers” who patronized one company because its (implementing a much higher bar for contributory infringement). 65. See St George Exec. Shuttle, LLC v. W. Trails Charters & Tours LLC, No. 2:17-cv-900- DN, 2018 WL 3350348, at *3 (D. Utah July 29, 2018). 66. See Soter Technologies, LLC v. IP Video Corporation, 523 F. Supp. 3d 389, 398 (S.D.N.Y. 2021) (finding that the defendant’s use of the plaintiff’s trademark in a domain name to redirect traffic to the defendant’s own website was a use in commerce). 67. See Edina Realty, Inc., v. TheMLSonline.com, No. Civ. 04-4371JRTFLN, 2006 WL 737064 at *3 (D. Minn. Mar. 20, 2006) (holding that purchase of trademarked AdWord was a use in commerce); Buying For The Home, LLC v. Humble Abode, LLC, 459 F. Supp 2d 310, 321-23 (D.N.J. 2006) (holding the same); Rescuecom Corp. v. Google Inc., 562 F.3d 123, 129-30 (2d Cir. 2009) (finding sale of trademarks as AdWords to be a use in commerce). 68. Barton Beebe, An Empirical Study of the Multifactor Tests for Trademark Infringement, 94 CAL. L. REV. 1581, 1588 (2006) [https://perma.cc/N2NS-KML5] [hereinafter The Beebe Study]. 69. Id. at 1588, 1591. 70. Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2d Cir.1961), cert. denied, 368 U.S. 820 (1961). 71. See AHP Subsidiary Holding Co. v. Stuart Hale Co., 1 F.3d 611, 616 (7th Cir. 1993) (“None of the . . . factors alone is dispositive in a likelihood of confusion analysis.”); Team Tires Plus, Ltd. v. Tires Plus, Inc., 394 F.3d 831, 833 (10th Cir. 2005) (“[N]o single factor is dispositive.”); Gateway, Inc. v. Companion Prods., Inc., 384 F.3d 503, 509 (8th Cir. 2004) (“No single factor is dispositive.”); Plus Prods. v. Plus Disc. Foods, Inc., 722 F.2d 999, 1004 (2d Cir. 1983) (“No single . . . factor is determinative.”). 2023] CONFUSION MÉLANGE 291 trademark created the mistaken belief that it was another.72 Although this restrictive language was removed in 1962,73 courts persisted in limiting likelihood-of-confusion analysis to so-called “point-of-sale” confusion.74 Congress’ intent would not be put into practice until 1973 when the District Court for New York’s Southern District heard the case of Grotrian-Steingweg v. Steinway and Sons, which concerned a piano-maker’s adoption of “Grotrian- Steinweg” as a trademark with the specific intent of creating an “initial interest” among potential consumers of the famous “Steinway” brand of high-end pianos.75 Although the court acknowledged that any case of mistaken identity between the two brands would be temporary at best, it nevertheless concluded that such confusion created an injury actionable under the Lanham Act.76 The precedent of this case would develop into the doctrine of “initial interest confusion,” which holds that a consumer need not be confused at the time of purchase for trademark infringement under the Lanham Act to have occurred; as long as the consumer was at one point confused by identical or similar trademarks, the duration of the hex is not an issue.77 The doctrine is instead concerned first with whether consumers were lured into a “bait-and-switch” scenario, whereby the alleged infringer used or imitated another’s trademark in order to foster interest among that competitor’s customers;78 and second with whether the owner of that misused trademark suffered an injury due to the alleged infringer having “appropriat[ed] the goodwill . . . developed in [their trademark].”79 These concerns mirror the dual goals of trademark protection: 72. See S. Rep. No. 87-2107 (1962), reprinted in 1962 U.S.C.C.A.N. 2844, 2850. 73. Pub. L. No. 87-772, § 11, 76 Stat. 771 (1962). 74. See Am. Greetings Corp. v. Easter Unlimited, Inc., 579 F. Supp. 607, 616 (S.D.N.Y. 1983) (“[I]t is buyer confusion that the Lanham Act addresses.”); see also BRS, Inc. v. Volume Shoe Corp., No. C81-1281A, 1982 WL 52210, at *7 (N.D. Ga. June 28, 1982) (“[N]o trademark infringement suit can proceed without evidence of . . . confusion at the point of sale.”). 75. Grotrian, Helfferich, Schulz, Th. Steinweg Nachf. v. Steinway & Sons, 365 F. Supp. 707 (S.D.N.Y. 1973). 76. Id. at 716-17. 77. CALLMANN ET AL., supra note 52, at § 22:17 (“[T]he alleged-infringer does not escape liability simply because any likely initial confusion will later be rectified.”). 78. See Checkpoint Sys., Inc. v. Check Point Software Techs., Inc., 269 F.3d 270, 294 (3d Cir. 2001) (“[T]he Seventh Circuit has equated initial interest confusion to a ‘bait and switch scheme.’”); Epic Sys. Corp. v. YourCareUniverse, Inc., 244 F. Supp. 3d 878, 901 (W.D. Wis. 2017) (initial interest confusion has been “summarized . . . as a ‘bait and switch’ technique”); Mirina Corp. v. Marina Biotech, 770 F. Supp. 2d 1153, 1160 (W.D. Wash. 2011) (“‘Initial interest confusion’ has been likened to a ‘bait and switch’ scheme.”); PlayNation Play Sys., Inc. v. Velex Corp., 924 F.3d 1159, 1167 n.4 (11th Cir. 2019) (initial interest confusion “occurs when an alleged infringer pulls a sort of bait and switch”); Vail Assocs., Inc. v. Vend-Tel-Co., 516 F.3d 853, 872 (10th Cir. 2008) (“Initial interest confusion is a ‘bait and switch’ tactic.”). 79. Brookfield Commc’ns, Inc. v. West Coast Ent. Corp., 174 F.3d 1036, 1057 (9th Cir. 1999); see BigStar Ent., Inc. v. Next Big Star, Inc., 105 F. Supp. 2d 185, 209 (S.D.N.Y. 2000) (initial interest confusion allows infringer “to profit from . . . established goodwill and reputation”); Austl. Gold, Inc. 292 INDIANA LAW REVIEW [Vol. 57:281 preventing bait-and-switch tactics speaks to consumer protection, while preserving goodwill reflects the safeguarding of trademark owner’s brand investments.80 Despite its appeal, the doctrine of initial interest confusion became something of a legal oddity in the years following the Grotrian-Steinweg decision.81 However, it was revitalized for the internet age by the Ninth Circuit Court in the case of Brookfield Communications v. West Coast Entertainment.82 The defendant in that case had used the plaintiff’s trademark in its metatags—text hidden on a website to improve its placement in organic search results—which the court felt could not create point-of-sale confusion, but would nevertheless likely cause that initial interest confusion.83 To explain initial interest confusion, the court used an analogy of a Blockbuster video store that purchases a highway billboard advertising its own location as that of its competitor, West Coast Video (the “Brookfield analogy”).84 Under the Brookfield analogy, a consumer who sees the advertisement on the highway may take the exit to patronize West Coast Video only to find the deceptive Blockbuster in its place; nevertheless, the consumer may decide to buy from the Blockbuster anyway, as she believes doing so would be better than leaving empty-handed.85 The Second and Ninth Circuits were quickly joined in adopting the initial interest confusion doctrine by the Third, Fifth, Sixth, Seventh, Eighth, and Tenth Circuits (the “initial interest Circuits”).86 However, the doctrine is not followed in the First, Fourth, Eleventh, D.C., and Federal Circuits, all of which still limit any analysis of consumer confusion to the moment of sale (collectively, the “point-of-sale Circuits”).87 3. Secondary Splits on Initial Interest Confusion.—The philosophical split between initial interest Circuits and point-of-sale Circuits might, in isolation, be considered relatively simple. But in the two decades since Brookfield, a number of secondary splits have developed among the initial interest Circuits. The Circuits’ respective opinions on these splits break down according to the following table (“Figure 1”). v. Hatfield, 436 F.3d 1228, 1239 (10th Cir. 2006) (initial interest confusion involves “us[ing] the goodwill associated with [another’s] trademarks”). 80. See discussion supra Part I.B.2. 81. See discussion infra Part I.B.3. 82. Brookfield Commc’ns, Inc., 174 F.3d at 1062. 83. Id. at 1062, 1066. 84. Id. at 1064. 85. Id. 86. See infra Figure 1. 87. See Hasbro, Inc. v. Clue Computing, Inc., 232 F.3d 1, 2 (1st Cir. 2000) (stating that the First Circuit’s “refusal to enter the ‘initial interest confusion’ thicket is well taken given the unlikelihood of ‘legally significant’ confusion”); Suntree Techs., Inc. v. Ecosense Int’l, Inc., 693 F.3d 1338, 1347 (11th Cir. 2012) (“[W]e need not reach the question whether initial interest confusion is actionable in the Eleventh Circuit.”); Weiss Assocs., Inc. v. HRL Assocs., Inc., 902 F.2d 1546, 1549 (Fed. Cir. 1990) (“[W]e are not following this concept of initial confusion.”). 2023] CONFUSION MÉLANGE 293 Figure 1 Federal Circuit Initial Interest Confusion is Unactionable if the Parties’ Products are Unrelated88 Initial Interest Confusion is Unactionable if Confusion is Brief or Disclaimers are Used89 Initial Interest Confusion is Unactionable if Consumers are Sophisticated or Careful90 Consumers on the Internet are Assumed to be Sophisticated or Careful91 Consumers of Expensive Products are Assumed to be Sophisticated or Careful92 1st Cir. 2nd Cir. T X X X T 3rd Cir. T / X? T T T T 4th Cir. 5th Cir. X T X - - - - 6th Cir. - - X X X T 7th Cir. X X T / X T / X X 8th Cir. - - T T X X 9th Cir. T X X T T 10th Cir. T X X - - - - 11th Cir. D.C. Cir. Fed. Cir. T = Rule AdoptedX = Rule Rejected- - = No Precedent: = Point-of-Sale Circuit 88. See infra notes 94-97, 99-100 and accompanying discussion in Part I.B.3.a. 89. See infra notes 101-07 and accompanying discussion in Part I.B.3.b. 90. See infra notes 110-17 and accompanying discussion in Part I.B.3.c. 91. See infra notes 120-27 and accompanying discussion in Part I.B.3.d. 92. See infra notes 128-33 and accompanying discussion in Part.I.B.3.e. 294 INDIANA LAW REVIEW [Vol. 57:281 Unlike many Circuit splits, these five secondary splits cannot be attributed to any single philosophical disagreement. If guilt had to be assigned, it would most reasonably be to the doctrine’s own underdeveloped nature at the time of the Brookfield decision. In the twenty-six years between initial interest confusion’s inception in Grotrian-Steinweg and its revitalization in Brookfield, only six cases included in the Federal Reporter even mentioned the doctrine.93 The result was that, when the doctrine began to be more widely consumed after Brookfield, its actual substance was much more akin to a raw egg than an omelet. Because the doctrine had been subject to so little legal refinement before its expansion at the turn of the millennium, the initial interest Circuits have naturally made their own attempts at filling gaps in the doctrine whenever novel issues arise. Such issues have included the interplay between initial interest confusion and consumer sophistication, uncertainty about the applicability of the Brookfield analogy to e-commerce, and concerns about the doctrine opening the doors to trivial litigation. Although a disagreement between Circuits on any of these issues alone may seem fairly inconsequential, the existence of all five secondary splits has led to complete discord; no two Circuits have adopted the exact same combination of positions. To better understand these secondary splits, it is necessary to address them individually. a. Initial interest confusion is unactionable if the parties’ products are unrelated.—Proponents of the rule wish to restrict the initial interest confusion doctrine to cases between actual competitors, where one party’s appropriation of another’s name and reputation is most likely to have a bottom-line effect. This thinking tracks well with the legacy of initial interest confusion on the internet, and the competitive relationship between the two businesses is a central aspect of the Brookfield analogy. Currently, there is support for this rule in the Second,94 Third,95 Ninth,96 and Tenth Circuits.97 Authority supporting the opposite side of this secondary split stems largely from pre-Brookfield decisions. Nevertheless, it could be viewed as a rejection of 93. Mobil Oil Corp. v. Pegasus Petroleum Corp., 818 F.2d 254, 260 (2d Cir. 1987); Weiss Assocs., Inc. v. HRL Assocs., Inc., 902 F.2d 1546, 1549 (Fed. Cir. 1990); Nike, Inc. v. Just Did It Enterprises, 6 F.3d 1225, 1231 (7th Cir. 1993); Dorr-Oliver, Inc. v. Fluid-Quip, Inc., 94 F.3d 376, 383 (7th Cir. 1996); Rust Env’t & Infrastructure, Inc. v. Teunissen, 131 F.3d 1210, 1217 (7th Cir. 1997); Elvis Presley Enterprises, Inc. v. Capece, 141 F.3d 188, 204 (5th Cir. 1998). 94. See BigStar Ent., Inc. v. Next Big Star, Inc., 105 F. Supp. 2d 185, 210 (S.D.N.Y. 2000) (explaining that initial interest confusion applies where “[t]he junior user, offering competing goods, [seeks] knowingly to benefit from public recognition of the senior company’s name”). 95. See Chatam Int’l, Inc. v. Bodum, Inc., 157 F. Supp. 2d 549, 558 (E.D. Pa. 2001) (“Where companies ‘are non-competitors, initial interest confusion . . . does not [impose] the prospect of harm.’”) (quoting Checkpoint Sys., Inc. v. Check Point Software Techs., Inc., 104 F.Supp.2d 427 (D.N.J. 2000)). 96. See Network Network v. CBS, Inc., No. CV 98-1349 NM(ANX), 2000 WL 362016 at *9 (C.D. Cal. Jan. 18, 2000) (“Dissimilarity of goods and services resolves the initial interest confusion question” in favor of rejecting the doctrine). 97. See Altira Grp. LLC v. Philip Morris Co’s. Inc., 207 F. Supp. 2d 1193, 1203 (D. Colo. 2002) (declining to find actionable initial interest confusion in a tobacco company’s use of a trademark similar to investment company). 2023] CONFUSION MÉLANGE 295 over-reliance on that case and its famous analogy. Alternatively, the jurisdictions that subscribe to this view could simply be seen as expansionist in regard to initial interest confusion, perhaps in light of Congress’ alleged endorsement of the doctrine.98 In any event, courts in the Fifth and Seventh Circuits seem to have endorsed this unencumbered vision of initial interest confusion.99 Peculiarly, there is authority in the Third Circuit also supporting this view, albeit older than that endorsing the opposition.100 b. Initial interest confusion is unactionable if confusion is brief or disclaimers are used.—Perhaps out of a concern for incentivizing de minimis infringement claims, most Circuits have further restricted initial interest confusion by allowing those who use another’s trademark to escape liability by minimizing the time that consumers are confused. In their precedent on the issue, the Fifth and Eighth Circuits have specifically mentioned disclaimers as a valid tool toward this end.101 Perhaps the broadest statement of this position comes from the Third Circuit: hearing a case on an infringing domain name, the District Court for the District of New Jersey suggested that a defendant in such an action need only prove that “a reading of the [allegedly infringing] Internet site” would “eliminate [any] likelihood of confusion” that the domain name may have caused, with no specific instruction on the required thoroughness of the reading.102 Across the proverbial aisle, however, the remaining Federal Circuits have declined to so restrict the applicability of initial interest confusion. Instead, this side has adopted the view, in the words of the Seventh Circuit, that “[w]hat is important is not the duration of the confusion, it is the misappropriation of . . . goodwill.”103 In effect, the proponents of this viewpoint argue that the harm is inflicted as soon as an initial interest has been fostered. The Seventh Circuit’s 98. See discussion supra Part I.B.2.b. 99. See Westchester Media Co. L.P. v. PRL USA Holdings, Inc., 103 F. Supp. 2d 935, 972 (S.D. Tex. 1999) (finding “a perceived affiliation” via initial confusion between magazine publisher and fashion label); Brach Van Houten Holding, Inc. v. Save Brach’s Coal. for Chi., 856 F. Supp. 472, 475 (N.D. Ill. 1994) (holding that the labor union’s use of a similarly named candy company’s logo was likely to cause confusion “based on an initial understanding”). 100. See Acxiom Corp. v. Axiom, Inc., 27 F. Supp. 2d 478, 497-98 (D. Del. 1998) (finding “likelihood of [initial interest] confusion” between trademarks of telecommunication company and marketing company). 101. See Gruma Corp. v. Mexican Restaurants, Inc., No. 4:09-cv-488-MHS-ALM, 2013 WL 12134147, at *14-15 (E.D. Tex. Sept. 27, 2013) (initial interest confusion can be rendered non- injurious “by conspicuous disclaimers placed . . . to alert the prospective customers”), and Fargo Elecs., Inc., v. Iris Ltd., No. 04-1017 (JRT/FLN), 2005 WL 1431653 at *6 (D. Minn. Mar. 8, 2005) (finding harm from initial confusion in search results can be mitigated if alleged infringer’s website contains “prominent disclaimers”). 102. Jews For Jesus v. Brodsky, 993 F. Supp. 282, 303 (D.N.J. 1998), aff’d, 159 F.3d 1351 (3d Cir. 1998). 103. Promatek Indus., Ltd. v. Equitrac Corp., 300 F.3d 808, 812-13 (7th Cir. 2002), as amended (Oct. 18, 2002); see Caterpillar Inc. v. Telescan Techs., L.L.C., No. CIV.A. 00-1111, 2002 WL 1301304, at *4 (C.D. Ill. Feb. 13, 2002) (finding that the use of disclaimers on a landing page “will not cure the problem of initial interest confusion” resulting from the use of trademarks in a domain name). 296 INDIANA LAW REVIEW [Vol. 57:281 view has seemingly found support in the Second,104 Sixth,105 Ninth,106 and Tenth Circuits.107 c. Initial interest confusion is unactionable if consumers are sophisticated or careful.—While no single factor in any of the multifactor likelihood-of-confusion tests is individually decisive, one factor in particular has become particularly consequential outside of the context of those tests: the level of consumer sophistication, alternatively called the degree of consumer care. This factor refers broadly to how much scrutiny a typical consumer will pay when purchasing the product in question. Although the status of sophistication and care is a legal fiction, courts have sought to designate certain facts as indicia of the designation, with most courts connecting it “with price, length and complexity of the purchase transaction,” “infrequency of purchase,” and whether the relevant consumers are “professional buyers.”108 Consumer care has also been tied to product type, with more careful purchasing being assumed for “important products of individual taste,” such as clothing, cosmetics, and fragrances.109 Regardless of how it is determined, most Circuits are of the opinion that a consumer’s level of sophistication or care has no effect on her susceptibility to initial interest confusion. This could be viewed as the “original” philosophy of initial interest confusion, given that the Second Circuit case originating the doctrine included a finding that discerning consumers of high-end grand pianos might still be subject to initial interest confusion.1 1 0 Under this view, a sophisticated consumer is no less likely to take the highway exit in the Brookfield analogy, nor any more willing to return home empty-handed. In the decades since that decision, this view has been co-opted by courts in the Fifth,111 Sixth,112 104. See OBH, Inc. v. Spotlight Mag., Inc., 86 F. Supp. 2d 176, 190 (W.D.N.Y. 2000) (finding that initial interest confusion caused by defendant’s domain name “cannot adequately be remedied by disclaimer”). 105. See PACCAR Inc. v. TeleScan Techs., L.L.C., 319 F.3d 243, 253 (6th Cir. 2003), abrogated by KP Permanent Make-Up, Inc. v. Lasting Impression I, Inc., 543 U.S. 111 (2004) (“A disclaimer . . . read by a consumer after reaching the web site comes too late.”). 106. See SMC Promotions, Inc. v. SMC Promotions, 355 F. Supp. 2d 1127, 1135 (C.D. Cal. 2005) (actionable initial interest confusion likely “regardless of Defendants’ use of a disclaimer on their websites”). 107. See Austl. Gold, Inc. v. Hatfield, 436 F.3d 1228, 1243 (10th Cir. 2006) (disclaimers on landing page would not “prevent Defendants from capitalizing on consumers’ initial interest confusion”). 108. Thomas R. Lee et al., Trademarks, Consumer Psychology, and the Sophisticated Consumer, 57 EMORY L.J. 575, 580-81 (2008). 109. CALLMANN ET AL., supra note 52, at § 21:74. 110. Grotrian, Helfferich, Schulz, Th. Steinweg Nachf. v. Steinway & Sons, 365 F. Supp. 707, 717 (S.D.N.Y. 1973). 111. See John Crane Prod. Sols., Inc. v. R2R & D, LLC, 861 F. Supp. 2d 792, 799 (N.D. Tex. 2012) (“[E]ven a sophisticated purchaser can be subject to initial interest confusion.”). 112. See Covantage Credit Union v. Blue Cross Blue Shield of Mich. Mut. Ins., No. 21-12559, 2022 WL 821668 at *11 (E.D. Mich. Mar. 16, 2022), appeal dismissed, No. 22-1329, 2022 WL 13638736 (6th Cir. May 11, 2022) (“Initial interest confusion afflicts both sophisticated and 2023] CONFUSION MÉLANGE 297 Seventh,113 Ninth,114 and Tenth Circuits.115 On the other hand, courts in other Circuits have rejected this philosophy, instead finding that sophisticated or careful consumer behavior acts as a natural defense against initial interest confusion, mitigating or entirely eliminating its effects. Courts in the Third and Eighth Circuits have formally accepted this view.1 1 6 Peculiarly, the Seventh Circuit appears to also have precedent—apparently still good law—endorsing this view despite its clear contradiction to other decisions in its jurisdiction.117 d. Consumers on the internet are assumed to be sophisticated or careful.—Recall that the Ninth Circuit’s Brookfield analogy presupposes that a consumer misled into visiting a deceptive Blockbuster is nevertheless likely to patronize it in light of the time and effort already expended to travel there.118 This phenomenon, whereby consumers tend to continue behavior into which they perceive an investment to have already been made, is commonly referred to outside of the legal field as the “sunk cost effect” or “sunk cost fallacy.”119 Within a matter of years after Brookfield, the Ninth Circuit had already begun to question whether internet consumers were actually susceptible to the type of sunk cost present in the Brookfield analogy.120 In the Court’s view, the consumer in the Brookfield analogy could not reasonably be expected to patronize the deceptive Blockbuster if she could return to the highway as instantaneously as an internet unsophisticated buyers.”). 113. See Epic Sys. Corp. v. YourCareUniverse, Inc., 244 F. Supp. 3d 878, 902 (W.D. Wis. 2017) (“[T]he doctrine [of initial interest confusion] can apply to more expensive items and more sophisticated consumers.”), and Miyano Mach. USA, Inc., v. MiyanoHitec Mach., Inc., 576 F. Supp. 2d 868, 885 (N.D. Ill. 2008) (finding that “initial interest confusion is likely to occur . . . [e]ven if customers are sophisticated”). 114. See 24/7 Customer, Inc. v. 24-7 Intouch, No. 5:14-cv-02561-EJD, 2015 WL 1522236 at *4 (N.D. Cal. Mar. 31, 2015) (“[N]o matter how sophisticated the customers, they are susceptible to initial interest confusion.”). 115. See SanMedica Int’l, LLC v. Amazon.com, Inc., No. 2:13-cv-00169-DN, 2016 WL 527055, at *8 (D. Utah Jan. 20, 2016) (noting that consumers who are “likely to exercise a moderate to high degree of care” may nevertheless “potentially [be] misled into an initial interest”). 116. See Checkpoint Sys., Inc. v. Check Point Software Techs., Inc., 269 F.3d 270, 296-97 (3d Cir. 2001) (“[I]nitial confusion will not likely facilitate free riding on the goodwill of another mark” when consumers exercise a high degree of care), and Sensient Techs. Corp. v. SensoryEffects Flavor Co., 613 F.3d 754, 766 (8th Cir. 2010) (declining to recognize initial interest confusion where “customers are sophisticated and exercise a relatively high degree of care”). 117. See Rust Env’t & Infrastructure, Inc. v. Teunissen, 131 F.3d 1210, 1217 (7th Cir. 1997) (declining to recognize initial interest confusion when “the relevant market of consumers is composed of sophisticated buyers”). 118. See discussion supra Part I.B.2.b. 119. See Jamie Ducharme, The Sunk Cost Fallacy Is Ruining Your Decisions. Here’s How., TIME MAG. (Jul. 26, 2018, 10:00 AM), https://time.com/5347133/sunk-cost-fallacy-decisions/ [https://perma.cc/BQ2T-RJMU]. 120. See Playboy Enterprises, Inc. v. Netscape Commc’ns. Corp., 354 F.3d 1020, 1036 (9th Cir. 2004) (Berzon, C.J., concurring) (expressing that “the minimal inconvenience in directing one’s web browser” eliminated sunk cost effect). 298 INDIANA LAW REVIEW [Vol. 57:281 user can click the “back” button on her web browser.121 Pursuing this rationale, the Ninth Circuit eventually adopted the assumption that internet users generally exercise a high degree of care.122 Precedent quickly sprung up in the Third and Seventh Circuits embracing this notion.123 This view is far from universal, however; other Circuits have embraced the assumption that internet consumers exercise the same degree of care as other consumers, and many have even recognized that the internet fosters a lower degree of care. In at least one Second Circuit case, the often spontaneous nature of online shopping was said to weigh in favor of a likelihood of confusion.124 Sixth Circuit courts have adopted the same position, formally endorsing the premise that internet users “do not undergo a highly sophisticated analysis” when browsing the web.125 Despite the Seventh Circuit’s aforementioned endorsement of the Ninth Circuit’s view, there seems to also be some recognition in that jurisdiction that the contrary position is good law.126 The Eighth Circuit Court of Appeals also recently rejected the Ninth Circuit’s assumption.127 e. Consumers of expensive products are assumed to be sophisticated or careful.—This secondary split is the least spoken-upon, as only five of the initial interest Circuits have decisions on record that could reasonably be said to endorse one side or the other. That being said, it may be the split that can most readily be understood; conventional wisdom would say that the more expensive a purchase is, the heavier the decision will weigh on the potential buyer. Applying the 121. See id. 122. See Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1153 (9th Cir. 2011) (the assumption that internet consumers are not sophisticated “may have been accurate [in Brookfield], [but] we suspect that there are many contexts in which it no longer holds true”), and Toyota Motor Sales, U.S.A., Inc. v. Tabari, 610 F.3d 117, 1178 (9th Cir. 2010) (internet consumers “are generally quite sophisticated” and “skip from site to site, ready to hit the back button whenever they’re not satisfied”). 123. See CollegeSource, Inc. v. AcademyOne, Inc., No. CIV.A. 10-3542, 2012 WL 5269213 at *19 (E.D. Pa. Oct. 25, 2012) (explaining that “[m]odern Internet users” possess a high level of care); Chatam Int’l, Inc. v. Bodum, Inc., 157 F. Supp. 2d 549, 559 (E.D. Pa. 2001) (“Internet surfers are inured to the false starts and excursions awaiting them.”), and Morningware, Inc. v. Hearthware Home Prods., Inc., No. 09 C 4348, 2012 WL 3721350 at *10 (N.D. Ill. Aug. 27, 2012) (stating that consumers possess a general “sophistication with . . . internet commerce”). 124. Edible Arrangments, LLC v. Provide Com., Inc., No. 3:14-CV-00250 (VLB), 2016 WL 4074121, at *10 (D. Conn. July 29, 2016) (“Internet purchasing is both fast-paced and rapidly evolving” and “weighs against sophistication.”). 125. Audi AG v. D’Amato, 469 F.3d 534, 554 (6th Cir. 2006) (quoting PACCAR Inc. v. TeleScan Techs., L.L.C., 319 F.3d 243, 253 (6th Cir. 2003)); see Ohio State Univ. v. Thomas, 738 F. Supp. 2d 743, 753-54 (S.D. Ohio 2010) (“[I]nternet consumers are especially unlikely . . . to take care in ascertaining the true source of the material they view.”). 126. See 3M Co. v. Mohan, No. 09-1413 ADM/FLN, 2010 WL 5095676 at *22 (D. Minn. Nov. 24, 2010), aff’d, 482 F. App’x 574 (Fed. Cir. 2012) (stating that ease of navigating web makes it “particularly unlikely that online consumers will avoid confusion through . . . due care”). 127. See Select Comfort Corp. v. Baxter, 996 F.3d 925, 937 (8th Cir. 2021) (noting that the “mix of authority regarding consumer confusion in . . . internet shopping” shows why assumed sophistication is unwise). 2023] CONFUSION MÉLANGE 299 Brookfield analogy, the diverted driver may be more hesitant to patronize the Blockbuster if the product she seeks is a costly one. Courts in the Second,128 Third,129 Sixth,130 and Ninth Circuits seem to agree that a product’s cost is of paramount importance when determining whether its consumers are sophisticated or careful.131 Not all Circuits subscribe to this notion, of course. The contrary belief is that price is just one consideration when determining the sophistication of the relevant market of consumers, and a consideration that may be easily outweighed at that. For example, the Eighth Circuit Court recently stated that, although expensive products typically warrant greater deliberation, the infrequency of their purchase makes it difficult to assume that consumers are particularly sophisticated about them.132 Similarly, some courts in the Seventh Circuit have expressed that neither the high price of competing products nor the technical expertise of their consumers makes those consumers particularly adept at identifying the products’ respective sources.133 II. THE ISSUES A. The Inefficiencies of the Multifactor Tests for Likelihood of Confusion The Circuits’ various multifactor tests have been subject to wide criticism, having been accused of being bloated unwieldy, costly, and primarily geared to benefit deep-pocket litigants.134 Even worse, there is evidence that the tests’ 128. See Tiffany & Co. v. Costco Wholesale Corp., 971 F.3d 74, 90 (2d Cir. 2020) (“[T]he greater the value of an article, the more careful the typical consumer can be expected to be.”) (quoting Manhattan Indus., Inc. v. Sweater Bee by Banff, Ltd., 627 F.2d 628 (2d Cir. 1980)). 129. See Checkpoint Sys., Inc. v. Check Point Software Techs., Inc., 269 F.3d 270, 286 (3d Cir. 2001) (finding that “consumers view the purchase [of expensive security systems] as an important investment decision” and thus exercise greater care). 130. See Covantage Credit Union v. Blue Cross Blue Shield of Mich. Mut. Ins., No. 21-12559, 2022 WL 821668 at *11 (E.D. Mich. Mar. 16, 2022) (“When . . . [products] are expensive or unusual, then the buyer can be expected to exercise greater care.”). 131. See Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1153 (9th Cir. 2011) (“[C]onsumers of expensive products online [are expected] to be even more sophisticated.”). 132. Select Comfort Corp., 996 F.3d at 937 (finding that “mattresses are relatively expensive” but purchased so infrequently that sophistication and care cannot be assumed). 133. See Unity Health Plans Ins., v. Iowa Health Sys., 995 F. Supp. 2d 874, 891 (W.D. Wis. 2014) (“Price is not . . . necessarily determinative [of consumes’ degree of care] on its own.”); see also Agdia Inc. v. Jun Quiang Xia, No. 3:15-CV-075 JD, 2017 WL 3438174 at *5-6 (N.D. Ind. 2017) (declining to find that consumers of “nuanced agricultural products” costing “several hundreds of dollars” were sophisticated). 134. See Glynn S. Lunney, Jr., Trademark’s Judicial De-Evolution: Why Courts Get Trademark Cases Wrong Repeatedly, 106 CAL. L. REV. 1195, 1236 (2018) (arguing that complex trademark litigation has created “a two-tiered trademark system: one system for those who can afford it, and a second for those who cannot”); Robert G. Bone, Taking the Confusion Out of “Likelihood of Confusion”: Toward A More Sensible Approach to Trademark Infringement, 106 NW. U. L. REV. 300 INDIANA LAW REVIEW [Vol. 57:281 complexity bestows little, if any, legitimate benefit on the act of truth-finding, and may even be used as a veneer to prop up shabby decisions. A 2006 study of preliminary injunction and bench trial opinions identified a number of “core” factors— “similarity of the marks,” “proximity of the goods,” “evidence of actual confusion,” “strength of the mark,” “intent of the defendant,” and “consumer sophistication”—which are used by all or almost all Circuits and which are highly correlated to a finding that consumer confusion is likely.135 Correlations with favorable findings on other “non-core” factors were also found.136 Legal scholars attribute these correlations in large part to the phenomenon of “stampeding”: the tendency for a judge—having found enough of the core six factors to believe consumer confusion to be likely—to decide the non-core factors such that they conform with the pre-selected outcome.137 In this manner, an injunction against a defendant may be justified on five or six factors when it may only be truly persuasive as to two or three. Some legal scholars attribute stampeding to the “overload of factors” that judges are tasked with juggling during likelihood of confusion analysis.138 Others postulate that judges knowingly stampede factors in order to efficiently complete the formalities of their analysis after their mind has already been made up.139 In either event, there is certainly case law showing a tendency for some judges to more or less abandon the multifactor analysis after having latched onto one or two pieces of persuasive circumstantial evidence.140 The result of this behavior, whether conducted knowingly or otherwise, is that the Circuits’ current tests for likelihood of confusion are not only overwrought and redundant, but also are quickly abandoned once they become inconvenient. 1307, 1336 (2012) (“[T]he vague and open-ended nature of the likelihood of confusion test makes success hard to predict even for legitimate users, and the dependence of the test on case-specific facts makes litigation costly.”). 135. The Beebe Study, supra note 68, at 1609 (finding that five of the six core factors had “win rates” between .793 and .970, with “proximity of the goods” an outlier at .716). 136. Id. 137. Id. at 1614. 138. Daryl Lim, Trademark Confusion Revealed: An Empirical Analysis, 71 AM. U. L. REV. 1285, 1323 (2022). 139. The Beebe Study, supra note 68, at 1603. 140. See Tempur-Pedic North America, LLC v. Mattress Firm, Inc., No. CV H-17-1068, 2017 WL 2957912, at *8-9 (S.D. Tex. July 11, 2017) (concluding that consumer confusion from purchase of trademarked AdWord was likely because defendant’s sponsored link connected to a “potentially confusing web page” without considering any other factors); Playboy Enterprises, Inc. v. Netscape Commc’ns. Corp., 354 F.3d 1020, 1027 (9th Cir. 2014) (expressing that disputed evidence of actual confusion “alone probably precludes summary judgment” regardless of other factors); 1-800 Contacts, Inc. v. Lens.com, Inc., 722 F.3d 1229, 1252 (10th Cir. 2013) (finding that consumer confusion from purchase of trademarked AdWord was unlikely because defendant had not used plaintiff’s trademark in its sponsored link or landing page); Alzheimer’s Disease & Related Disorders Ass’n v. Alzheimer’s Found. of Am., Inc., 307 F. Supp. 3d 260, 299 (S.D.N.Y. 2018) (limiting analysis on an AdWord infringement case to three “most important factors” of strength of mark, similarity of marks, and actual confusion). 2023] CONFUSION MÉLANGE 301 B. Circuit Court Splits on Initial Interest Confusion It should come as no surprise that numerous problems have emerged as a result of the veritable wad of precedent caused by the secondary splits between initial interest Circuits. As Figure 1 demonstrates, no two initial interest Circuits have the exact same combination of rules. In effect, these splits have taken a simple division on the basis of initial interest confusion and transformed it into a precedential battle royale. The madness of this mélange becomes apparent when applied to examples, particularly ones in the realm of digital marketing. Imagine that SmellNet.com, an online retailer of perfumes, has recently discovered that SmellX.com, a new competitor, has been purchasing “smellnet” and “smellnet.com” as AdWords in order to scoop SmellNet.com’s online consumers.141 If the case arose in the Sixth Circuit, the presiding court would likely find that perfume—being a relatively expensive item—is presumably purchased by sophisticated, discerning consumers. Nevertheless, SmellNet.com would be permitted to argue initial interest confusion since the Sixth Circuit does not restrict the doctrine’s use on the basis of consumer sophistication. But consider if the case arose in the Third Circuit: although consumers of perfume would likely once again be considered sophisticated on price alone, Third Circuit precedent would now prohibit SmellNet.com from proceeding with its case on an initial confusion theory. Instead, it would have to meet the higher evidentiary hurdle of proving a likelihood of point-of-sale confusion. One can also imagine the scenario where the operator of a famous and well- established casino—the “Ecuatorial”—finds out that a competitor has been running an online gambling website using the domain name “ecuatorialcasino.com.”142 Although the typically impulsive draw of gambling might suggest that the casinos’ relevant consumers may not be particularly discerning or careful, the mere fact that the defendant’s gambling venue exists in cyberspace would likely be enough to classify its userbase as sophisticated in the Third, Seventh, and Ninth Circuits. This finding alone would make initial interest confusion unactionable in the Third and (possibly) Seventh Circuits, and the operators of the Ecuatorial would have to pursue their case by arguing point-of- sale confusion. Further, consider the example of “Fam Car,” a website-based, family-oriented transportation business that recently learned that a former employee had started a competing business, “FamMoto.”143 Anticipating litigation, Fam Car’s counsel decides to peruse FamMoto’s website, only to find that it is plastered with disclaimers disavowing any association with Fam Car. If Fam Car’s case were heard in the Third, Fifth, or Eighth Circuits, this fact alone would effectively eliminate its ability to argue initial interest confusion.144 This is not so in the Seventh Circuit; instead, initial interest confusion might be prohibited by virtue 141. See FragranceNet.com, Inc. v. FragranceX.com, Inc., 679 F. Supp. 2d 312 (E.D.N.Y. 2010). 142. Aztar Corp. v. MGM Casino, No. NO. 00-833-A, 2001 WL 939070 (E.D. Va. Apr. 9, 2001). 143. Kid Car NY, LLC v. Kidmoto Techs. LLC, 518 F. Supp. 3d 740, 747-49 (S.D.N.Y. 2021). 144. See discussion supra Part I.B.3.b. 302 INDIANA LAW REVIEW [Vol. 57:281 of the businesses’ primarily virtual nature. Only if Fam Car were fortuitously placed within the jurisdiction of the Second, Sixth, Ninth, or Tenth Circuits would it be able to state its claim on a theory of initial interest confusion. That, in essence, is the heart of the problem: the burden of proof on any claim for trademark infringement will, under current circumstances, be inexorably tied to geographic location. It is difficult to imagine any reasonable justification for leaving a federal-level law to the foibles normally reserved for state-level issues. C. Increasing Breadth, Intricacy, and Cost of Trademark Litigation In the nearly quarter century since the Brookfield decision that revived initial interest confusion, it may still be said that the doctrine is most applicable to cases of alleged trademark infringement in online commerce and digital marketing.145 What is less certain is whether this integration has benefitted these fields, the law, or the public at large. In the opinion of many legal scholars, this proposition can be resoundingly rebuffed; the doctrine has been indicted for such varied charges as menacing litigants with inconsistent application, battering the free market with over-expansive protections, and vandalizing the consumer protection principles underpinning trademark law.146 These misgivings, along with the practical consideration that initial interest confusion weakens their position, will naturally motivate defendants to put forth arguments against applying the doctrine, making infringement litigation more prolonged and costly for all involved. Of course, trademark infringement litigation does not end once the often- arduous question of initial interest confusion has been answered; that resolution is often merely the beginning of proceedings. The parties must still thereafter argue through the likelihood-of-confusion factors, as well as the question of whether the defendants used the trademark in commerce, if that element has not yet been settled.147 These factual inquiries stretch out the timeline for trademark infringement litigation and balloon its cost. Under these conditions, deeper- pocketed trademark owners can muscle their way into favorable outcomes over 145. See discussion supra Part I.B.2.b.; Playboy Enterprises, Inc. v. Netscape Commc’ns Corp., 55 F. Supp. 2d 1070, 1074 (C.D. Cal. 1999), aff’d, 202 F.3d 278 (9th Cir. 1999) (“Initial interest confusion . . . is a brand of confusion particularly applicable to the internet.”); 1 E-COMMERCE AND INTERNET LAW § 7.08[2] (2020 Update) (“The doctrine therefore can be especially important to brand owners in Internet disputes . . .,” although it “is better suited to some Internet disputes than others.”). 146. See Christopher Ruhland, Can ‘Initial Interest Confusion’ Doctrine in Trademark Law be Clarified?, 11 No. 5 ANDREWS INTELL. PROP. LITIG. REP. 12, 1 (2004) (questioning “whether Brookfield and subsequent cases incorrectly applied the ‘initial interest confusion’ doctrine”); Priya Singh, Abolish Trademark Law’s Initial Interest Confusion and Permit Manipulative Internet Search Practices, 3 J. BUS. ENTREPRENEURSHIP & L. 15, 31 (2009) (claiming that “initial interest confusion does not necessarily require there to be a likelihood of confusion” and commerce would be better served with its removal); Ross D. Petty, Initial Interest Confusion Versus Consumer Sovereignty: A Consumer Protection Perspective on Trademark Infringement, 98 TRADEMARK REP. 757, 759 (2008) (“Initial interest confusion also has been criticized for contravening the [goal of] . . . serving consumer welfare.”). 147. See explanation supra Part I.B.1. 2023] CONFUSION MÉLANGE 303 smaller parties.1 4 8 The result is that many legitimate uses of trademarks—including those protected by free speech—have become chilled, as discontinuing such use in the face of threatened legal action is typically viewed as preferable to wagering on expensive and unpredictable litigation.149 III. PROPOSED SOLUTIONS Effectively addressing these issues requires the acceptance of certain parameters. The first is that, given the choice between resolving an issue in a case as a matter of law or leaving it for resolution by a trier-of-fact, the preference should always be for the former.150 It is true that such an approach could, if taken to extremes, decrease case-specific factual inquiry and lead to a detached class of one-size-fits-all litigation. But the current state of digital trademark litigation represents close to the opposite, with cases having become so inundated with layer upon layer of factual inquiry that it is little wonder the nation’s courts have struggled to maintain any semblance of cohesion; such circumstances necessitate drastic measures. The second parameter is that the motivations behind trademark law must be kept in mind. Recall that trademark law has two primary purposes: to protect consumers from deception and confusion, and to protect producers from misappropriation and devaluation of their brand.151 In regard to initial interest confusion, these two purposes are perpetuated in the dual goals of disincentivizing bait-and-switch tactics and preventing the appropriation of trademark goodwill.152 One must also keep in mind the current disparity between large firms with valuable brands and small firms that often forgo legitimate uses of those brands to avoid litigation.153 Given these notions, these recommendations must also attempt to strike a three-part balance: seeking rules that fairly serve the welfare of consumers, the interests of large businesses, and the concerns of smaller businesses. With these principles in mind, the issues currently facing trademark law can be properly addressed. While three recommendations are made in this Note, they 148. See Bone, supra note 134, at 1336 (in infringement litigation, “a legitimate user might well agree to a pro-plaintiff settlement, especially if the user is risk averse and less financially secure than the trademark owner”); Glynn S. Lunney, Jr., Trademark’s Judicial De-Evolution: Why Courts Get Trademark Cases Wrong Repeatedly, 106 CAL. L. REV. 1195, 1236 (2018) (claiming litigation complexity has created “a two-tiered trademark system: one system for those who can afford it, and a second for those who cannot”). 149. See Deven R. Desai, Response: An Information Approach to Trademarks, 100 GEO. L.J. 2119, 2123 (2012) (“[T]he state of the likelihood-of-confusion doctrine generates litigation and chills speech.”); Graeme B. Dinwoodie, Lewis & Clark Law School Ninth Distinguished IP Lecture: Developing Defenses in Trademark Law, 13 LEWIS & CLARK L. REV. 99, 128 (2009) (postulating “uncertainties about factual outcomes may . . . chill socially desirable conduct”). 150. Dinwoodie, supra note 149, at 128 (“Defenses that immunize conduct as a matter of law would ameliorate the[] concerns” of proper trademark usage being chilled). 151. See discussion supra Part I.B.1. 152. See discussion supra Part I.B.2.b. 153. See discussion supra Part II.C. 304 INDIANA LAW REVIEW [Vol. 57:281 are not intended to be inexorably linked. Though the latter two are mutually nonexclusive with the first and exclusive as to each other, any of the three could be adopted independently of the remaining two. A. Standardize Initial Interest Confusion Rectifying the fragmented state of the initial interest confusion doctrine requires the recognition of a specific truth: that the Circuits’ efforts to adopt hard and fast rules on the doctrine’s application, disjointed though they may be, are not entirely misguided. As noted earlier, trademark litigation has become over- complicated, costly, and intimidating to parties who may not understand the law.154 To redress these issues, the initial interest confusion doctrine should be standardized and simplified. This requires resolution of each of the five secondary splits that currently plague the doctrine. 1. Bar Initial Interest Confusion if the Parties’ Products are Unrelated.—Given the emphasis placed on bait-and-switch tactics in the discourse on initial interest confusion by postulations like the Brookfield analogy, this secondary split seems almost ridiculous. After all, if the party setting the “bait” were not a competitor to the trademark owner, then no “switch” could reasonably be expected to occur; the baited consumers would be wholly uninterested in the competitor’s offering, and the trademark owner would not be deprived of a lost sale. It is also worth mentioning that, for the Circuits refusing to limit initial interest confusion to the competitive business context, the precedential case in each predates the seminal Brookfield decision, with some originating deep into the 1990s.155 Not only was the initial interest confusion doctrine in its undeveloped infancy at that time, as previously discussed,156 but it must also be said that any contemporary legal interpretations could not have anticipated the form which commerce has taken, particularly in regard to the rise in internet entrepreneurship. In this latter day of internet commerce, even small merchants can easily and continuously search for similarly-named brands. If the initial interest doctrine was available in all cases, then more aggressive parties would be granted a green light to intimidate or bring litigation against any whom they suspect may weaken their own trademarks, even if there is no chance of their customers being actually diverted. Given these considerations, it would be reasonable to allow initial interest confusion only in those cases where a party’s trademark has been allegedly infringed by a competitor. In such situations, the plaintiff fits the bill for the type of party that the doctrine was invented to protect, and therefore can be justified in enjoying a lower evidentiary burden. In all other cases, point-of-sale confusion should apply. 2. Allow Initial Interest Confusion Regardless of Brevity of Confusion or Disclaimer Usage.—Under the current secondary split, the majority of Circuits currently allow initial interest confusion regardless of the brevity of confusion or 154. Id. 155. See sources and accompanying text supra notes 99-100. 156. See discussion supra Part I.B.3. 2023] CONFUSION MÉLANGE 305 the use of disclaimers.157 It would perhaps be best to explain the superiority of this approach by way of example. Imagine a modified version of the Brookfield analogy, with the part of the deceptive business being played by the Adidas footwear brand rather than by Blockbuster. Now imagine that, rather than a billboard advertising its own location as that of a Nike store, Adidas purchased the top sponsored link position for the AdWord “nike shoes,” and wrote the text of its advertisement to be generically appealing to fans of footwear. If a consumer ran a search for “nike shoes,” she would see Adidas’ advertisement and, perhaps, click on it out of the assumption it would take her to the Nike website. While the consumer’s confusion would be short-lived after arriving at the Adidas landing page, Adidas has already gotten what it wanted by prompting the misdirection. It would not be appropriate to allow infringers like Adidas in this scenario to effectively impose a higher evidentiary burden on trademark owners like Nike by simply attempting to cure the initial confusion after it has already occurred. Even if Adidas used a disclaimer on its website, Nike has still had its goodwill and reputation appropriated and the consumer has already been subject to the bait- and-switch. In addition, allowing infringers free range to divert consumers off- course, only to avoid responsibility through post-hoc disclaimers, would do a disservice to the motivation for trademark law to foster commercial efficiency.15 8 This restriction is further necessitated by the need for more efficient resolution of trademark infringement cases. If, starting tomorrow, courts begin to accept disclaimers as a defense against initial interest confusion, by next week parties would be lining up to argue what facts constitute a legally effective disclaimer. It is better to render this a non-factor rather than invite further factual inquiries into litigation which is already excessively fact-driven. 3. Bar Initial Interest Confusion if Consumers are Sophisticated or Careful.—This restriction is notably unpopular among the initial interest Circuits, with only the Third and Eighth Circuits fully endorsing it.159 In truth, both positions on this secondary split have valid reasonings behind them. Nevertheless, the more restrictive position should remain in place in its current proponent Circuits and be adopted in all others, as it represents a reasonable limitation on initial interest confusion in light of the doctrine’s goals. Visualize once again the modified version of the Brookfield analogy, with Adidas’ generically written sponsored link in the search results for “nike shoes.” While it might be true that a consumer who sees that link may click the link under the presumption that it is a Nike link, it is also true that such a consumer—having searched specifically for “nike shoes”—will be likely to discriminate on the brand of shoe she purchases and therefore unlikely to patronize Adidas after uncovering its deception. While such a hypothetical does seem to fit the classic bait-and-switch scenario to which initial interest confusion is often applied, it seems unlikely that Nike will be able to state any real harm stemming from Adidas’ misappropriation of its goodwill and reputation. On the contrary, Nike’s goodwill and reputation 157. See supra Figure 1. 158. See discussion supra Part I.B.1. 159. See supra Figure 1. 306 INDIANA LAW REVIEW [Vol. 57:281 are precisely what prevented it from losing a potential sale.160 With only one of the two motivations behind initial interest confusion present, it simply does not seem reasonable to allow Nike to take advantage of the doctrine’s lower evidentiary burden when it’s the brand sophistication of the relevant consumer that effectively mitigated any injury. Initial interest confusion is, at its heart, a doctrine intended to make likelihood of confusion easier to prove in circumstances where it makes sense to do so. The problem is that the Circuits currently determine the existence of such circumstances based on a mixture of a few consumer-focused metrics—such as the notion that “professional purchasers” are sophisticated by default—and a swath of product-focused metrics—price, complexity of transactions, frequency of purchases, and product type.161 While this latter group is well-suited for inferring a tendency for the relevant consumers to exercise sophistication and care, they are often employed as bright-line rules by courts to conclude that consumers are sophisticated or careful based on one or two of these product- focused metrics.162 This conflation of tendency and certainty is yet another contributor to the arbitrariness of modern trademark infringement litigation. Federal courts should therefore begin to recognize product-focused metrics as mere indicia of sophistication and care, and instead place the majority of the evidentiary weight on consumer-focused metrics, with specific focus on how loyal the relevant consumers are to particular brands and how susceptible those consumers are to impulse purchasing. 4. Do Not Assume Consumers on the Internet to be Sophisticated or Careful.—The Circuits that have adopted initial interest confusion should move away from the assumption that internet consumers have a high degree of care and cannot suffer from the sunk cost effect. This misguided assumption is largely based on the early-2010s-era premise that internet consumers were naturally savvy and would only become savvier as internet usage becomes more commonplace.163 In the decade since this prediction was made, internet usage has indeed expanded, but one would be hard-pressed to argue that the average internet consumer has become much more scrutinizing. Some recent studies suggest that even experienced internet users are consistently unable to discern sponsored links from organic results on a search engine results page, despite search engine providers plainly labeling such links as advertisements.164 160. See Lee et al., supra note 108, at 649 n.377 (“If a[] [sophisticated] consumer’s motivation makes her unlikely to be satisfied with a counterfeit, her initial interest confusion may be not cognizable if she can be expected to hold out for the genuine article.”). 161. Id. at 579-81. 162. See sources and accompanying text supra notes 128, 130, 131; see also Brockmeyer v. Hearst Corp., 248 F. Supp. 2d 281, 299 (S.D.N.Y. 2003) (finding consumers of magazines sophisticated due to price despite magazines generally being “purchased on impulse”). 163. Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1152 (9th Cir. 2011) (“[T]he default degree of consumer care is becoming more heightened as the novelty of the Internet evaporates and online commerce becomes commonplace.”). 164. Tom Vaughton, Almost 60% of People Still Don’t Recognize Google Paid Ads When They See Them, VARN (Jan. 18, 2018), https://varn.co.uk/01/18/varn-original-research-almost-60-people- 2023] CONFUSION MÉLANGE 307 Even if one were to assume that internet users were becoming more sophisticated, there is little reason to doubt that online retailers and service providers are becoming at least as sophisticated at fostering consumer impulsivity.165 An assumption that internet consumers are sophisticated by default not only contravenes consumer protection by unjustly raising the evidentiary bar for showing deceptive marketing practices, but also makes it more difficult for trademark owners to defend the value of their brands online.166 5. Do Not Assume Consumers of Expensive Products to be Sophisticated or Careful.—While it is not certain where the presumption originated that a high price tag will automatically invite sophisticated behavior, the Circuits who have adopted this presumption have expressed essentially the same sentiment: consumers of expensive products can be expected to be more careful since a larger amount of money hangs in the balance of their decisions.167 Regardless of its origin, such a presumption should be dismissed. As stated in subsection 3 above, the price of a product is only an indicator that the relevant consumers tend to behave carefully, not conclusive evidence of such behavior.168 In regard to e- commerce specifically, there is no evidence that expensive items have somehow avoided the aforementioned trend away from consumer scrutiny; even traditionally negotiable transactions like used car sales have seemingly become comfortable with the impersonality of e-commerce.169 There should also be some concern with the fact that this assumption essentially turns one of the “core” factors of likelihood-of-confusion analysis into an easily settled numbers game.170 While this may be a reasonable measure to promote efficient litigation, it would be better to save courts’ time in other areas with less risk of adversely affecting the overall infringement litigation. As stated in subsection 3 above, the factor of consumer sophistication and degree of care is primarily concerned with how much scrutiny consumers practice and how susceptible they are to being deceived or swayed. While a product’s high price tag may increase the tendency for a consumer to act carefully, it is not conclusory to that issue and should not be treated as if it were. still-don’t-recognise-google-paid-ads-see/ [https://perma.cc/L4HH-ANRP]; OFCOM, ADULTS’ MEDIA USE AND ATTITUDES 182 (2016), https://www.ofcom.org.uk/__data/assets/pdf_file/0026/ 80828/ 2016-adults-media-use-and-attitudes.pdf [https://perma.cc/CHQ8-GY7E] (finding only 52% of inexperienced and 62% of experienced internet users could identify sponsored links). 165. See Fiona Zeng Skovhøj, 4 Magical Ways of Driving Impulse Buying on Your E-Commerce Site, CLERK.IO (June 17, 2021), https://www.clerk.io/blog/the-magic-of-impulse-purchases-this- season [https://perma.cc/4SAZ-X3F9]. 166. See discussion supra Part I.B.1. 167. See sources and accompanying text supra notes 128, 130, 131. 168. See discussion supra Part III.A.3. 169. Langer v. Carvana, LLC, No. 8:21-cv-00303-JLS-JDE, 2021 WL 4439096 (C.D. Cal. Aug. 24, 2021) (a case involving virtual auto sales platform). 170. See discussion supra Part I.B.2.b. 308 INDIANA LAW REVIEW [Vol. 57:281 B. Abridge the Multifactor Likelihood of Confusion Tests in Digital Marketing Cases Trimming down the multifactor tests in digital marketing cases would mitigate many of the unfavorable aspects of modern infringement litigation. Removing less important factors from consideration in certain cases would give parties facing threats of infringement litigation a clearer notion of their prospects at success, therefore reducing the current “chilling” effect on legitimate trademark usage.171 The solution to fixing the contemporary issues of trademark law may actually be the adoption of a test that is nearly a hundred years old. When the American Law Institute first published the Restatement (First) of Torts in 1938, it proffered two separate tests for likelihood of confusion.172 The test, intended to be used in cases between competitive businesses,173 lists four factors for courts to consider: (a) the degree of similarity between the designation and the trade-mark or trade name in (i) appearance; (ii) pronunciation of the words used; (iii) verbal translation of the pictures or designs involved; (iv) suggestion; (b) the intent of the actor in adopting the designation; (c) the relation in use and manner of marketing between the goods or services marketed by the actor and those marketed by the other; (d) the degree of care likely to be exercised by purchasers.174 The other test, which contained nine factors, was intended to be used only in cases where the parties in question were not business competitors.175 Despite its original limitations, this test began to seep into cases of infringement between competitors, eventually developing into the Circuits’ current multifactor tests for likelihood of confusion.176 While it cannot be said that the American Law Institute circa 1938 could have predicted the nature of online commerce today, their simplified test has obvious appeal given the policy-based criticism levied on the current schema for trademark infringement litigation.17 7 This course of action would also address the issue of stampeding, as the factors recognized under the First Restatement’s approach all fall within the core factors found to be actually determinative in 171. See discussion supra Part II.C. 172. The Beebe Study, supra note 68, at 1588. 173. Id. 174. RESTATEMENT (FIRST) OF TORTS § 729 (1938). 175. The Beebe Study, supra note 68, at 1588. 176. Physicians Formula Cosms., Inc. v. W. Cabot Cosms., Inc., 857 F.2d 80, 83 (2d Cir. 1988) (“Although the Polaroid test originally was applied to noncompeting products . . . it has been expanded to apply where, as here, competing goods are involved.”). 177. See discussion supra Part II.C. 2023] CONFUSION MÉLANGE 309 findings of a likelihood of confusion.178 Further, since digital marketing primarily gives rise to infringement litigation between competitors,179 applying this abridged test to digital marketing would be within the purview of the First Restatement’s drafters. C. Replace the Multifactor Likelihood of Confusion Tests for Digital Marketing Cases Perhaps as a result of the need to fit into the prefabricated systems that make up e-commerce, cases of trademark infringement in digital marketing have tended to break down into predictable fact patterns. In broad strokes, these cases almost always involve competitors; one or both having used the other’s trademark either visibly—such as using infringing domain names or misleading sponsored links—or invisibly—such as using the trademark as an AdWord or metatag—or some combination of the two. Online service providers may be contributorily liable for selling trademarks as AdWords or facilitating infringing recommendations. It would behoove federal courts to rework the test for likelihood of confusion into a step-by-step inquiry based on the regularly recurring facts in these cases. If likelihood of confusion could be determined in this way, the predictability and efficiency of trademark infringement litigation would be greatly improved. What’s more, these reworked tests could be readily adopted by federal courts without revising the Lanham Act. These tests could be structured as follows: (a) Do the parties’ products, whether goods or services, exist in the same industry? • If yes, proceed to (b). • If no, proceed to (d). (b) Can the typical consumer of the parties’ products be said, on the basis of verifiable data or overwhelming inference, to be objectively sophisticated or careful based on attributes such as their status as a professional purchaser, their high brand loyalty, or their tendency to purchase the product impulsively? • If yes, only point-of-sale confusion is applicable; proceed to (d). • If no, initial interest confusion is applicable; proceed to (c). (c) Do the defendant’s user-facing promotional messages (i.e. sponsored links, product listings, or other advertisements) contain either: (1) the plaintiff’s trademark; (2) another trademark that is similar enough to the plaintiff’s own to foster confusion or association with the plaintiff or indicate obvious inspiration; OR (3) generic text that could be applicable to any firm in the industry, including plaintiff? • If yes to any of the above, then there is initial interest confusion. • If no to all of the above, proceed to (d). (d) Is the defendant’s landing page and website sufficiently masqueraded so as to sustain a typical consumer’s confusion or 178. See supra Part I.B.2.b. 179. See discussion supra Part I.A. 310 INDIANA LAW REVIEW [Vol. 57:281 perceived association with the plaintiff to the point of sale, including through the defendant’s use of either: (1) the plaintiff’s trademark; OR (2) another trademark that is similar enough to plaintiff’s own to foster confusion or association with the plaintiff or indicate obvious inspiration? • If yes, there is a likelihood of point-of-sale confusion. • If no, there is no likelihood of confusion. This example test incorporates only the factors that are most consequential to the current multifactor tests, and further simplifies matters by eliminating factors that can be assumed due to the nature of digital marketing. The benefits of this streamlining can be demonstrated in the case that began this Note. If this test were applied to Boost Beauty, the first inquiry under part (a) would be whether defendant Arash’s products competed with plaintiff Yasha’s.180 This can be answered affirmatively given that both products served the same purpose, were sourced from the same manufacturer, and sold for the same price.181 The inquiry can therefore move on to part (b), which asks whether the typical consumer of the products in question is sophisticated or careful, supportable either by “verifiable data or overwhelming inference.” Yasha and Arash both sold “non-medicated eyelash growth serum,”182 which was of questionable value.183 Absent verifiable data to the contrary, the most logical conclusion seems to be that the consumers of such products are not particularly scrutinizing. The inquiry can proceed to part (c), which concerns the contents of defendant Arash’s user-facing promotional messages. Arash’s product listing contains a trademark (“WooLash”) that is similar to the plaintiff Yasha’s trademark (“BoostLash”) in phonetics, appearance, and overall impression; it would be reasonable to conclude, based on the facts of this case in particular, that this indicates obvious inspiration.184 The inquiry would thus conclude with a finding of a likelihood of confusion. It should be noted, however, that even if Yasha’s claim of trademark infringement failed the inquiry and a likelihood of confusion were not found, that conclusion could be reached in a fraction of the time that it took to reach the same conclusion in the actual case. What’s more, a test such as this would also serve as a practice guide for trademark usage in digital marketing, allowing those who legitimately use another’s trademark to know their rights and litigation prospects rather than have their usage chilled by thin threats of litigation. CONCLUSION Although the current issues in trademark law have primarily developed as a result of judge-made precedent, the truly unfortunate fact is that actual fault for their existence is quite difficult to assign. The average trademark infringement 180. Boost Beauty, LLC v. Woo Signatures, LLC, No. 2:18-CV-02960-CAS-EX, 2022 WL 409957 (C.D. Cal. Feb. 7, 2022). 181. Id. at *2. 182. Id. at *3. 183. See Schroeder, supra note 14. 184. Boost Beauty, LLC, 2022 WL 409957 at *1. 2023] CONFUSION MÉLANGE 311 case requires the determination of so many facts and the application of so many rules that its intricacy can exceed that of some first-degree murder cases. What’s more, the initial interest confusion doctrine was widely adopted despite having practically zero legal refinement,185 forcing federal courts into the awkward position of having to address novel issues as they arose. And arise those issues did, as the development of initial interest confusion just happened to coincide with the explosion of e-commerce and digital marketing. In truth, it is not surprising that so many problems exist in trademark law; its dysfunction seems, in hindsight, almost inevitable. Even so, this need not always be the case. This Note argues that the rules and doctrines currently governing trademark law and infringement litigation ought to be extensively revised, standardized, and simplified. Three proposals have been offered herein for the attention of judicial decision-makers or, at the very least, as food for thought to those of us who practice in the field of trademark law. The first is to cure the secondary splits on initial interest confusion by universally adopting rules to restrict the doctrine only on the basis of non-competitive products or the sophistication and care of consumers, and to base the latter determination only on consumer-focused metrics like professionalism and brand loyalty. The second is to remedy the overly complex likelihood-of-confusion analysis in digital marketing cases by adopting the abridged test proposed in the First Restatement of Torts, which would make trademark infringement litigation more predictable and reduce its chilling effect on legitimate uses of trademarks. The third is to replace the multifactor tests for likelihood of confusion with a step-by-step factual inquiry that would allow most cases of trademark infringement in digital marketing to be handled at the summary judgment stage, in addition to serving as a guideline for legitimate trademark usage. While each of these proposals may be a useful remedy, the ultimate purpose of this Note is to bring to light the numerous complexities, inconsistencies, and disparities that currently plague trademark law. The ultimate responsibility for addressing these issues falls on the federal courts who enjoy a relatively low number of statutorily-imposed guidelines for trademark infringement. This openness means that most of the current schema around infringement litigation could be reworked by federal courts to create a more universal, predictable, affordable, and fair body of law. The biggest question is whether the initiative to do so will be taken. 185. See discussion supra Part I.B.3.