The Unseemly Web of Keyword Advertising

Despite the surface simplicity of keyword advertising disputes (typically entailing unwanted use of the exact trademark of a direct competitor promoting competing goods or services) the web the courts have spun addressing such Web-based advertising has been anything but. Fortunately, the Second Circuit's April 3 decision in Rescuecom Corp. v. Google, Inc., 562 F.3d 123 (2d Cir. 2009) (on the one-year anniversary of oral argument), straightens at least some of the tangled seams by recognizing that keyword ads tied to a trademark do constitute a use in commerce of the subject mark. The Second Circuit thus agreed with arguments made by this author in numerous articles (including Virtual Trademark Use ' The Parallel World of Keyword Ads, 98 Trademark Rptr. 873 (May-June 2008)) that New York district courts, in a cluster of cases recognizing no such “use in commerce,” had mistakenly focused on the statutory definition of how an owner acquires bona fide rights in a mark in the first instance, rather than the statutory test of infringement. The previous week, a Massachusetts decision adopted much the same analysis. Hearts on Fire Co. v. Blue Nile, Inc., 2009 WL 794482 (D. Mass. March 27, 2009).

In rejecting the broad conceptual rule precluding liability for lack of use in commerce, the Second Circuit simply remanded without assessing substantive liability. It thus remains to be seen whether or to what extent the New York courts may now adopt any of the equally broad conceptual rules embraced by various other courts supporting findings of liability, including the theory of “initial interest confusion” or a pared down “Internet-only” version of the traditional test of infringement, under which keyword ads might well be deemed per se infringements of the marks to which they are keyed. See e.g., Storus Corp. v. Aroa Marketing, Inc., No. C-06-2454 (MMC), 2008 WL 449835 (N.D. Cal. Feb. 15, 2008) (discussed below).

In their respective focuses to date on broad conceptual grounds to decide keyword advertising cases, few courts have carefully considered meaningful empirical evidence as to whether or when such advertising is indeed likely to cause consumers to make mistaken purchasing decisions. Although beyond the scope of the article, the jurisprudence in Europe is similarly fractured. Puzzling too is how different has been the treatment of Internet advertising as seen through the lens of other legal theories ' particularly privacy law. For instance, although courts consistently have held that pop-up advertising does not violate the Lanham Act, see, e.g., 1-800 Contacts, Inc. v. WhenU.com Inc., 414 F.3d 400 (2d Cir. 2005), the delivery of such advertising has been regularly regarded unlawful as an unfair and deceptive trade practice (either by the FTC or under state law). See In the Matter of DirectRevenue LLC, et al., Docket No. C-4194, FTC File No. 052-3131 (June 26, 2007); In the Matter of Advertising.com, Inc., 2005 ILRWeb (P&F) 2405, FTC File No. 042-3196; In the Matter of Zango, Inc. f/k/a 180solutions, Inc., 2006 ILRWeb (P&F) 3005, FTC File No. 052-3130; People of New York v. Intermix Media, Inc. (Tentative Agreement), No. 401394/2005 (NY Sup Ct, 2005)

And while the courts have split on theoretical grounds as to the lawfulness of keyword and pop-up advertising, the FTC has instead suggested an entirely different analytical approach in regulating Web-based advertising. In a recent report on behavioral advertising (i.e., tracking consumers' online activities to deliver ads targeted to the users' interests), the FTC has thus “attempted to balance the privacy concerns raised by online behavioral advertising against the potential benefits of the practice.” (Emphasis added.) While acknowledging that “consumers have genuine and legitimate concerns about how their data is collected, stored, and used online,” the FTC finds that consumers “may also benefit ' from the free content that online advertising generally supports, as well as the personalization of advertising that many consumers appear to value.” FTC Staff Report: Self-regulatory Principles For Online Behavioral Advertising, February 2009.

Eschewing the middle ground, courts confronting keyword controversies have used few such analytical tools. Indeed, a simple focus on the facts as analyzed under traditional theories of infringement, rather than broader conceptual constructs, may not only help ease the judicial conflicts but may better position courts to confront more subtle unseen uses of trademarks to drive Internet traffic as ever more obscure Internet optimization techniques are developed.

The Prior New York Rule: No Use in Commerce

In 1-800 Contacts, Inc., supra, the Second Circuit held that use of trademarks in an automated, machine-linking function to generate pop-up ads (in many ways similar to the way keyword ads are bought and sold) was not a “use in commerce” of the subject trademarks within the meaning of the Lanham Act. Accord U-Haul Int'l, Inc. v. WhenU.com, Inc., 279 F.Supp.2d 723 (E.D. Va. 2003); Wells Fargo & Co. v. WhenU.com, Inc., 293 F.Supp.2d 734 (E.D. Mich. 2003). Following this ruling, and relying as the Second Circuit had on the statutory definition
of use in commerce under 15 U.S.C. ' 1127, numerous New York district courts had held that, because keyword ads similarly entail no visible use of trademarks, they too did not constitute use in commerce of the subject marks. See, e.g., Merck & Co. v. MediPlan Health Consulting, Inc., 431 F. Supp.2d 425 (S.D.N.Y. 2006); S&L Vitamins, Inc. v. Australian Gold, Inc., 521 F. Supp. 2d 188, 199 (E.D.N.Y. 2007); Fragrancenet.com, Inc. v. FragranceX.com, Inc., 493 F. Supp. 2d 545 (E.D.N.Y. 2007); Site Pro-1, Inc. v. Better Metal, Inc., 2007 WL 1385730 (E.D.N.Y. 2007).

Without overruling 1-800 Contacts, the Second Circuit nonetheless ruled in Rescuecom that the district courts had misapplied its pop-up precedent to keyword cases. The error lay in focusing solely on the statutory definition under 15 U.S.C. ' 1127 of what sort of “bona fide” “use in commerce” is necessary for a claimed trademark owner to establish rights in its mark in the first place rather than the more relevant tests of commercial use constituting infringement under 15 U.S.C. ' 1114 (1) or ' 43(a), irrespective of whether any such word, term, name, symbol, or device appears directly on a product or service so as to satisfy the statutory test of what is a trademark.

In so holding, the Second Circuit engaged in statutory analysis, focusing on the structure of the two different tests of commercial use ' in particular, the language of ' 1127 requiring that the mark be placed “on the goods or their containers or the displays associated therewith” so as to show the “bona fide” use of the mark. Infringers, by contrast, have no such bona fides. In a lengthy appendix, the court also traced the history of the Lanham Act, which (as this author previously argued) shows the plainly different purposes of the two commercial use provisions.

On Beyond the Hudson

Where does this leave us? In Rescuecom, the Second Circuit expressly declined to hazard a guess, on appeal of a Rule 12 motion to dismiss, whether Rescuecom ultimately could prove infringement or whether, alternately, Google could show that its sale of keyword advertising was no different from product placement of store-brand products shelved with branded goods. (Another comparison often made is to the practice of aggregating advertising for like products or services in one place in Yellow Pages directories). The precedents give little guidance as to what circumstances are or are not likely, in fact, to cause consumer confusion. Indeed, although there has been much litigation, there have been few cases finally decided on the merits, and the decisions have tended either to accept or call into question (without finally deciding) broad conceptual or theoretical grounds for condemning keyword ads as a matter of principle.

Exemplifying one extreme, the recent decision, Storus Corp. v. Aroa Marketing, Inc., No. C-06-2454 (MMC), 2008 WL 449835 (N.D. Cal. Feb. 15, 2008), granted summary judgment that a keyword ad for defendant's money clip infringed the trademark “Smart Money Clip” ' notwithstanding the somewhat descriptive nature of the mark and notwithstanding that the defendant's ad identified its own brand name (Steinhausen) for its own “smart money clip.” Invoking a pared down special Internet version of the familiar multipart test of likelihood of confusion, Storus concluded instead that, “in the context of the Web,” the focus need be confined (perhaps entirely) to a so-called “Internet trinity” or “controlling troika” of three factors: similarity of the marks, relatedness of the goods or services, and the simultaneity of the parties' use of the Internet for marketing. Id. at *3. Because keyword ads arise only “in the context of the Web,” the third element can be assumed, and because parties are most likely to purchase keyword ads tied to their own or their competitors' marks (or to descriptive product terms that happen also to be trademarks), the remaining two elements of similarity of the marks and relatedness of the goods or services will also likely be a foregone conclusion. The only stated basis for downplaying or omitting such factors as 1) user sophistication; 2) customer care and 3) evidence or not of actual confusion was a summary order in Comp Examiner Agency, Inc. v. Juris, Inc., No. 96-0213, 1996 WL 376600, at *1 (C.D.Cal. Apr. 26, 1996). That that order contained no analysis has not prevented numerous other cases from similarly relying on it to enshrine the same new trinity. See GoTo.com, Inc. v Walt Disney Co., 202 F.3d 1199 (Fed Cir. 2002); Brookfield Commun's, Inc. v. West Coast Entm't Corp., 174 F.3d 1036, 1034 n. 16 (9th Cir 1999).

Storus and many other cases have further sidestepped deep factual analysis based on the theory of “initial interest confusion” derived from Brookfield Commun's, Inc., a case decided relatively early in the online era involving a use of metatags that, it was feared, might divert parties searching for a trademark owner's site to an infringer's site instead. In Playboy Enterprises, Inc. v. Netscape Commun's, Corp., 354 F.3d 1020 (9th Cir. 2004) for instance, the majority concluded:

Even if they realize 'immediately upon accessing' the competitor's site that they have reached a site 'wholly unrelated to' [Playboy's], the damage has been done. Through initial interest confusion, the competitor 'will have gained a customer by appropriating the goodwill that [Playboy] has developed in its mark.

Id. at 1025 (emphasis added), quoting Brookfield, 174 F.3d at 1057. Because of the ease with which Internet users can toggle back and forth between sites, the theory has hardly been without its critics. See Gov't Employees Ins. Co. v. Google, Inc., 77 U.S.P.Q.2d 1841, 1845 (E.D.Va. 2005); Google, Inc. v. American Blind & Wallpaper Factory, Inc., 2007 WL 1159950 (N.D.Cal. 2007). Accord, J.G. Wentworth, S.S.C. v. Settlement Funding LLC, 2007 WL 30115 at *7 (E.D.Pa. 2007) (distinguishing between the choices created by multiple keyword and natural search results, each clearly labeled even if on a single page, and the sort of bait and switch marketing that might lead to mistaken purchasing decisions.) Questioning the concept, however, is not the same as making fundamental factual findings whether or when keyword ads cause confusion.

Although keyword cases tend to present at least superficially simple fact patterns in which the accused party has used the plaintiff's exact mark to promote directly competing products or services, courts have yet to develop any consistent analytical approach to assessing the risks of confusion caused by keyword ads and appear to be struggling to establish some simple ' perhaps too simple ' theoretical framework for deciding these cases, yea or nay. In the process, many cases have been decided at a conceptual level rather than on solid factual grounds. Rather than drawing up broad theoretical edifices favoring or disfavoring such advertising, a return to familiar principles and solid empirical evidence may hold the key, not only to resolving these cases fairly, but to adjudicating future disputes involving similar but perhaps more subtle uses of trademarks to drive Internet commerce.

Jonathan Moskin, a member of this newsletter's Board of Editors, is a partner in the New York office of White & Case LLP. ' 2009 Jonathan Moskin.

Despite the surface simplicity of keyword advertising disputes (typically entailing unwanted use of the exact trademark of a direct competitor promoting competing goods or services) the web the courts have spun addressing such Web-based advertising has been anything but. Fortunately, the Second Circuit's April 3 decision in Rescuecom Corp. v. Google, Inc. , 562 F.3d 123 (2d Cir. 2009) (on the one-year anniversary of oral argument), straightens at least some of the tangled seams by recognizing that keyword ads tied to a trademark do constitute a use in commerce of the subject mark. The Second Circuit thus agreed with arguments made by this author in numerous articles (including Virtual Trademark Use ' The Parallel World of Keyword Ads, 98 Trademark Rptr. 873 (May-June 2008)) that New York district courts, in a cluster of cases recognizing no such “use in commerce,” had mistakenly focused on the statutory definition of how an owner acquires bona fide rights in a mark in the first instance, rather than the statutory test of infringement. The previous week, a Massachusetts decision adopted much the same analysis. Hearts on Fire Co. v. Blue Nile, Inc., 2009 WL 794482 (D. Mass. March 27, 2009).

In rejecting the broad conceptual rule precluding liability for lack of use in commerce, the Second Circuit simply remanded without assessing substantive liability. It thus remains to be seen whether or to what extent the New York courts may now adopt any of the equally broad conceptual rules embraced by various other courts supporting findings of liability, including the theory of “initial interest confusion” or a pared down “Internet-only” version of the traditional test of infringement, under which keyword ads might well be deemed per se infringements of the marks to which they are keyed. See e.g., Storus Corp. v. Aroa Marketing, Inc., No. C-06-2454 (MMC), 2008 WL 449835 (N.D. Cal. Feb. 15, 2008) (discussed below).

In their respective focuses to date on broad conceptual grounds to decide keyword advertising cases, few courts have carefully considered meaningful empirical evidence as to whether or when such advertising is indeed likely to cause consumers to make mistaken purchasing decisions. Although beyond the scope of the article, the jurisprudence in Europe is similarly fractured. Puzzling too is how different has been the treatment of Internet advertising as seen through the lens of other legal theories ' particularly privacy law. For instance, although courts consistently have held that pop-up advertising does not violate the Lanham Act, see, e.g. , 1-800 Contacts, Inc. v. WhenU.com Inc. , 414 F.3d 400 (2d Cir. 2005), the delivery of such advertising has been regularly regarded unlawful as an unfair and deceptive trade practice (either by the FTC or under state law). See In the Matter of DirectRevenue LLC, et al., Docket No. C-4194, FTC File No. 052-3131 (June 26, 2007); In the Matter of Advertising.com, Inc., 2005 ILRWeb (P&F) 2405, FTC File No. 042-3196; In the Matter of Zango, Inc. f/k/a 180solutions, Inc., 2006 ILRWeb (P&F) 3005, FTC File No. 052-3130; People of New York v. Intermix Media, Inc. (Tentative Agreement), No. 401394/2005 (NY Sup Ct, 2005)

And while the courts have split on theoretical grounds as to the lawfulness of keyword and pop-up advertising, the FTC has instead suggested an entirely different analytical approach in regulating Web-based advertising. In a recent report on behavioral advertising (i.e., tracking consumers' online activities to deliver ads targeted to the users' interests), the FTC has thus “attempted to balance the privacy concerns raised by online behavioral advertising against the potential benefits of the practice.” (Emphasis added.) While acknowledging that “consumers have genuine and legitimate concerns about how their data is collected, stored, and used online,” the FTC finds that consumers “may also benefit ' from the free content that online advertising generally supports, as well as the personalization of advertising that many consumers appear to value.” FTC Staff Report: Self-regulatory Principles For Online Behavioral Advertising, February 2009.

Eschewing the middle ground, courts confronting keyword controversies have used few such analytical tools. Indeed, a simple focus on the facts as analyzed under traditional theories of infringement, rather than broader conceptual constructs, may not only help ease the judicial conflicts but may better position courts to confront more subtle unseen uses of trademarks to drive Internet traffic as ever more obscure Internet optimization techniques are developed.

The Prior New York Rule: No Use in Commerce

In 1-800 Contacts, Inc., supra , the Second Circuit held that use of trademarks in an automated, machine-linking function to generate pop-up ads (in many ways similar to the way keyword ads are bought and sold) was not a “use in commerce” of the subject trademarks within the meaning of the Lanham Act. Accord U-Haul Int'l, Inc. v. WhenU.com, Inc. , 279 F.Supp.2d 723 (E.D. Va. 2003); Wells Fargo & Co. v. WhenU.com, Inc. , 293 F.Supp.2d 734 (E.D. Mich. 2003). Following this ruling, and relying as the Second Circuit had on the statutory definition
of use in commerce under 15 U.S.C. ' 1127, numerous New York district courts had held that, because keyword ads similarly entail no visible use of trademarks, they too did not constitute use in commerce of the subject marks. See, e.g., Merck & Co. v. MediPlan Health Consulting, Inc., 431 F. Supp.2d 425 (S.D.N.Y. 2006); S&L Vitamins, Inc. v. Australian Gold, Inc. , 521 F. Supp. 2d 188, 199 (E.D.N.Y. 2007); Fragrancenet.com, Inc. v. FragranceX.com, Inc. , 493 F. Supp. 2d 545 (E.D.N.Y. 2007); Site Pro-1, Inc. v. Better Metal, Inc., 2007 WL 1385730 (E.D.N.Y. 2007).

Without overruling 1-800 Contacts, the Second Circuit nonetheless ruled in Rescuecom that the district courts had misapplied its pop-up precedent to keyword cases. The error lay in focusing solely on the statutory definition under 15 U.S.C. ' 1127 of what sort of “bona fide” “use in commerce” is necessary for a claimed trademark owner to establish rights in its mark in the first place rather than the more relevant tests of commercial use constituting infringement under 15 U.S.C. ' 1114 (1) or ' 43(a), irrespective of whether any such word, term, name, symbol, or device appears directly on a product or service so as to satisfy the statutory test of what is a trademark.

In so holding, the Second Circuit engaged in statutory analysis, focusing on the structure of the two different tests of commercial use ' in particular, the language of ' 1127 requiring that the mark be placed “on the goods or their containers or the displays associated therewith” so as to show the “bona fide” use of the mark. Infringers, by contrast, have no such bona fides. In a lengthy appendix, the court also traced the history of the Lanham Act, which (as this author previously argued) shows the plainly different purposes of the two commercial use provisions.

On Beyond the Hudson

Where does this leave us? In Rescuecom, the Second Circuit expressly declined to hazard a guess, on appeal of a Rule 12 motion to dismiss, whether Rescuecom ultimately could prove infringement or whether, alternately, Google could show that its sale of keyword advertising was no different from product placement of store-brand products shelved with branded goods. (Another comparison often made is to the practice of aggregating advertising for like products or services in one place in Yellow Pages directories). The precedents give little guidance as to what circumstances are or are not likely, in fact, to cause consumer confusion. Indeed, although there has been much litigation, there have been few cases finally decided on the merits, and the decisions have tended either to accept or call into question (without finally deciding) broad conceptual or theoretical grounds for condemning keyword ads as a matter of principle.

Exemplifying one extreme, the recent decision, Storus Corp. v. Aroa Marketing, Inc., No. C-06-2454 (MMC), 2008 WL 449835 (N.D. Cal. Feb. 15, 2008), granted summary judgment that a keyword ad for defendant's money clip infringed the trademark “Smart Money Clip” ' notwithstanding the somewhat descriptive nature of the mark and notwithstanding that the defendant's ad identified its own brand name (Steinhausen) for its own “smart money clip.” Invoking a pared down special Internet version of the familiar multipart test of likelihood of confusion, Storus concluded instead that, “in the context of the Web,” the focus need be confined (perhaps entirely) to a so-called “Internet trinity” or “controlling troika” of three factors: similarity of the marks, relatedness of the goods or services, and the simultaneity of the parties' use of the Internet for marketing. Id. at *3. Because keyword ads arise only “in the context of the Web,” the third element can be assumed, and because parties are most likely to purchase keyword ads tied to their own or their competitors' marks (or to descriptive product terms that happen also to be trademarks), the remaining two elements of similarity of the marks and relatedness of the goods or services will also likely be a foregone conclusion. The only stated basis for downplaying or omitting such factors as 1) user sophistication; 2) customer care and 3) evidence or not of actual confusion was a summary order in Comp Examiner Agency, Inc. v. Juris, Inc., No. 96-0213, 1996 WL 376600, at *1 (C.D.Cal. Apr. 26, 1996). That that order contained no analysis has not prevented numerous other cases from similarly relying on it to enshrine the same new trinity. See GoTo.com, Inc. v Walt Disney Co., 202 F.3d 1199 (Fed Cir. 2002); Brookfield Commun's, Inc. v. West Coast Entm't Corp. , 174 F.3d 1036, 1034 n. 16 (9th Cir 1999).

Storus and many other cases have further sidestepped deep factual analysis based on the theory of “initial interest confusion” derived from Brookfield Commun's, Inc., a case decided relatively early in the online era involving a use of metatags that, it was feared, might divert parties searching for a trademark owner's site to an infringer's site instead. In Playboy Enterprises, Inc. v. Netscape Commun's, Corp. , 354 F.3d 1020 (9th Cir. 2004) for instance, the majority concluded:

Even if they realize 'immediately upon accessing' the competitor's site that they have reached a site 'wholly unrelated to' [Playboy's], the damage has been done. Through initial interest confusion, the competitor 'will have gained a customer by appropriating the goodwill that [Playboy] has developed in its mark.

Id. at 1025 (emphasis added), quoting Brookfield, 174 F.3d at 1057. Because of the ease with which Internet users can toggle back and forth between sites, the theory has hardly been without its critics. See Gov't Employees Ins. Co. v. Google, Inc. , 77 U.S.P.Q.2d 1841, 1845 (E.D.Va. 2005); Google, Inc. v. American Blind & Wallpaper Factory, Inc., 2007 WL 1159950 (N.D.Cal. 2007). Accord, J.G. Wentworth, S.S.C. v. Settlement Funding LLC, 2007 WL 30115 at *7 (E.D.Pa. 2007) (distinguishing between the choices created by multiple keyword and natural search results, each clearly labeled even if on a single page, and the sort of bait and switch marketing that might lead to mistaken purchasing decisions.) Questioning the concept, however, is not the same as making fundamental factual findings whether or when keyword ads cause confusion.

Although keyword cases tend to present at least superficially simple fact patterns in which the accused party has used the plaintiff's exact mark to promote directly competing products or services, courts have yet to develop any consistent analytical approach to assessing the risks of confusion caused by keyword ads and appear to be struggling to establish some simple ' perhaps too simple ' theoretical framework for deciding these cases, yea or nay. In the process, many cases have been decided at a conceptual level rather than on solid factual grounds. Rather than drawing up broad theoretical edifices favoring or disfavoring such advertising, a return to familiar principles and solid empirical evidence may hold the key, not only to resolving these cases fairly, but to adjudicating future disputes involving similar but perhaps more subtle uses of trademarks to drive Internet commerce.

Jonathan Moskin, a member of this newsletter's Board of Editors, is a partner in the New York office of White & Case LLP. ' 2009 Jonathan Moskin.