Trademark Infringement: Can I Bid on My Competitors’ Trademarks as Adwords?

By Vincent Allen and PJ Putnam

With the use of search engines by consumers to find products and websites becoming more prominent, search engine optimization has become an important part of the marketing plan for large and small businesses alike. For those companies that do not want to take the time or put in the effort to improve their SEO for certain keywords organically, purchasing keywords, also known as Adwords, from Google or other search engines can put a website on the first page of search results for a particular keyword overnight. By “purchasing,” we mean the submission of the high bid for the use of a keyword or campaign for a specific time period and geographic location.

In most cases, an Adwords customer “purchases” a word or a series of words that are used to describe the services or goods to be advertised and/or sold on the customer’s website. Recently, Google created a “campaign” of words that can be purchased in any given industry. A purchaser can specify which areas of the country and at what times those Adwords would be purchased. The scope of the geographic area and the times chosen determine the purchase price.

For example, a campaign that begins at midnight for three hours only on a Monday and a Wednesday will cost less than a campaign that runs seven days a week, twenty-four hours a day. If the customer decides to purchase a Google Campaign for a specific industry, the purchaser does not get to choose the words in the campaign, but is allowed to see the words. Adwords campaigns may contain the names of competitors or competitive products to increase the number of hits resulting from the campaign. As a result, a trademark owner’s competitors may be listed in ads located above the normal search results for the trademark.

Prior to 2004, Google’s trademark usage policy prevented the use of trademarks in the text of a sponsored advertisement and also prevented the use of trademarks as keywords if so requested by the trademark owner. But in 2004, Google loosened its policy to allow use of trademarks as keywords even if objected to by the trademark owner. The underlying reason for making the change was financial as Google’s research showed that about 7% of its total revenue was driven by trademarked keywords.

Google understood that the risk of litigation for both itself and its partners would increase as a result of the change in policy, but decided the risk was worth the additional revenue that was sure to come. Google even introduced a trademark-specific keyword tool that suggests relevant trademarks to clients to bid on as keywords. However, Google continued to block use of trademarks in the text of the advertisement because of internal studies suggesting that the unrestricted use of trademarks in the text of an advertisement might confuse users.

In 2009, Google further relaxed its policy to permit limited use of trademarks in advertising text. In particular, Google began permitting the use of trademarks in advertising text if 1) the sponsor is a reseller of a genuine trademarked product, 2) the sponsor makes or sells component parts for a trademarked product, 3) the sponsor offers compatible parts or goods for use with the trademarked product; or 4) the sponsor provides information about or reviews of a trademarked product.

As a result of this change in policy, use of trademarks by competitors in Adwords campaigns became even more controversial. Some companies, including American Airlines, Rosetta Stone, and Geico, filed trademark infringement suits against Google in an attempt to prevent Google from selling their trademarks as Adwords. These cases have been settled, but Google continues to allow the practice in the United States. And recently, Google reversed the practice of allowing trademark owners to object to others bidding on their trademarks as Adwords in Australia, Brazil, China, Hong Kong, Macau, New Zealand, South Korea, and Taiwan.

The Rosetta Stone case deserves some attention as a result of the evidence that was presented by Rosetta Stone to support its claim of trademark infringement. Although the district court granted summary judgment in favor of Google, the Fourth Circuit ruled last year that summary judgment was improperly granted given the evidence presented, noting that while summary judgment on a likelihood of confusion in a trademark infringement case is permissible in appropriate cases, this issue is “an inherently factual issue that depends on the facts and circumstances of each case.”

The evidence presented by Rosetta Stone in opposition to summary judgment included:

  • Testimony by five consumers who purchased bogus software mistakenly thinking it to be the authentic product;
  • Evidence that Rosetta Stone’s customer care center received 262 complaints over a one-year period from individuals who had mistakenly purchased counterfeit software;
  • Internal study by Google recommending that the only effective policy is to allow trademark usage for keywords but not allow trademark usage in ad text (either title or body);
  • Survey evidence by an expert opining that 17% of respondents were confused by Google’s sponsored links after conducting a search for Rosetta Stone; and
  • Internal study by Google reflecting that even well-educated, seasoned Internet consumers are confused by the nature of Google’s sponsored links and are sometimes unaware that sponsored links are, in actuality, advertisements.

The Fourth Circuit held that this evidence created a genuine issue of material fact that must be decided by a jury. Indeed, this is a common theme in trademark cases with respect to the likelihood of confusion issue, and in most cases summary judgment can be avoided because of the inherently factual nature of the inquiry.

Although the Rosetta Stone case was remanded for trial, the parties settled the case late last year. The settlement terms are confidential, but a joint statement by the parties provides that they will “meaningfully collaborate to combat online ads for counterfeit goods and prevent the misuse and abuse of trademarks on the Internet.”

Google has been aggressive and successful in defending its revenue stream from Adwords, and it appears that Google’s relaxed policy is here to stay. You may be thinking, “If Google allows the practice of purchasing trademarks as Adwords, then we can do it without worrying about a trademark infringement claim.” But that is not the case.

Although the purchase of Adwords, in and of itself, is generally not enough to subject the purchaser to a trademark infringement claim, the text of the ad could give rise to such a claim. Adwords are a bit unique in the trademark context, but the Fourth Circuit’s decision in the Rosetta Stone case demonstrates that the question of whether there is trademark infringement is still determined based on the “digits of confusion” like any other trademark infringement case.

In the Fifth Circuit, the digits of confusion that are considered include: 1) the type of trademark, 2) mark similarity, 3) product similarity, 4) outlet and purchaser identity, 5) advertising media identity, 6) defendant’s intent, 7) actual confusion, and 8) care exercised by potential purchasers. In College Network v. Moore Educational Publishers, decided in 2010, the Fifth Circuit upheld a jury finding that there was no likelihood of confusion caused by the purchase of the trademark “The College Network” as a keyword from Google and Yahoo to summon sponsored links for advertising competitive study guides for nursing students. In that case, experts were presented by both sides on the likelihood of confusion issue, and the jury was permitted to view the keyword-search process and visually compare the companies’ websites. The Fifth Circuit held that the evidence was not so strongly and overwhelmingly in favor of the plaintiff that a reasonable jury could not arrive at a verdict in favor of the defendant. But the opinion is short on analysis of the issues presented by sponsored advertising cases.

Under the initial interest confusion doctrine, if a consumer is confused as to the source of the goods or services by the text of the advertisement in a sponsored link, this could be enough to constitute trademark infringement even if the confusion is dispelled after clicking on the link and going to the website of the sponsor. The 2011 Network Automation v. Advanced System Concepts case is a sponsored advertising case in which the Ninth Circuit provides a thorough analysis of the likelihood of confusion issue in the context of sponsored advertising. In vacating a preliminary injunction by the district court, the Ninth Circuit found the following factors to be most important in a sponsored advertising case: 1) the strength of the mark; 2) the evidence of actual confusion; 3) the type of goods and degree of care likely to be exercised by the purchaser; and 4) the labeling and appearance of the advertisements and the surrounding context on the page displaying the results.

Of course, if there is a trademark infringement action, the court will look at all relevant factors. But the factor that advertisers should use to minimize their risk of litigation when purchasing trademarks as keywords is the labeling and appearance of the advertisements and the surrounding context on the screen displaying the search results. This factor was addressed at several points in the Network Automation decision, with the court stating that in the keyword advertising context, “likelihood of confusion will ultimately turn on what the consumer saw on the screen and reasonably believed, given the context.”

Therefore, if the advertisement is confusingly labeled or not labeled at all, there is a greater chance of initial interest confusion. However, clearly labeling the advertisement so that the consumer knows the source of the goods or knows that a competitive product is being offered before clicking the link could eliminate the likelihood of initial interest confusion.

Part of any likelihood of confusion analysis involves the consideration of the degree of care and sophistication of the consumer, which will vary depending on the product or service offered. It is assumed that the more expensive the products or services, the greater the degree of care that will be exercised by the consumer. Although the Ninth Circuit had once held that internet users as a whole exercise a low degree of care, the court suspects that there are many cases where that is no longer the case. “[T]he default degree of consumer care is becoming more heightened as the novelty of the Internet evaporates and online commerce becomes commonplace.”

Recently, the Wisconsin Court of Appeals heard an Adwords case where trademark infringement was not claimed, but the case centered on a law firm’s “right to privacy.” The plaintiff, Cannon & Dunphy, purchased Adwords containing the name of a competitor’s law firm, Habush & Rottier. The lawsuit centered on whether the Adwords purchase was barred under Wisconsin’s privacy law. In its ruling, the court held that “locating an advertisement or business near an established competitor to take advantage of the flow of potential customers or clients to the established business is not a practice the [Wisconsin] legislature intended to prohibit. Further, we fail to discern a meaningful distinction between competitors simply selecting locations in proximity to each other and using a third party [such as Google or Yahoo, as in this case] to obtain the same result.”

After the court’s ruling, the defendant did not remove its Adwords campaign. However, the plaintiff was able to win after all by simply outbidding the defendant for the Adwords, and therefore, gained control of them.

In conclusion, advertisers should conduct an analysis regarding how a typical consumer will interpret the advertisement that is used in connection with the purchase of trademarks as Adwords. Clearly identifying the name of the sponsor in the advertisement can help eliminate initial interest confusion. Likewise, comparing the sponsored product to the trademark owner’s product will also avoid confusion. But, failing to label the advertisement, or using misleading text, could subject the sponsor to trademark infringement liability. As with many purchases, Adwords are no different – Caveat Emptor!

Vincent Allen is a partner at Carstens & Cahoon. He focuses on the litigation of intellectual property disputes and the prosecution of patent and trademark applications.

P.J. Putnam is General Counsel and Executive Vice President at Calvet Companies, LLC.

The First Amendment Extends to Dillinger Tommy Guns

There has been significant activity over the past several years in the courts with regard to the First Amendment and how far it goes to protect the use of trademarks and the likeness of celebrities in video games.  I previously posted about the Madden NFL game and how the defendants were successful in a First Amendment Defense.

The Southern District of Indiana recently dismissed a case involving the use of “Dillinger” as the name of a weapon in the Godfather I and Godfather II video games.

The John Dillinger heirs sued EA under trademark infringement/right of publicity theories for using “Dillinger” and “Modern Dillinger” to refer to the Tommy Gun weapons available in the Godfather I and Godfather II games.  Although there were a number of issues raised in connection with a motion to dismiss/motion for summary judgment, the ultimate issue was whether EA had a First Amendment right to use “Dillinger” to refer to the weapons in the game.

In granting summary judgment for EA, the court applied the Second Circuit’s Rogers test, which requires an analysis of 1) whether the name used has any artistic relevance to the underlying work, and if so, 2) whether it explicitly misleads the public as to the source or content of the work.

In this case, the Godfather games were based on the novel and movie of the same name.  Neither the novel, nor the movie depicted John Dillinger.  However, it was undisputed that John Dillinger and Tommy Guns were closely related as that was Dillinger’s weapon of choice.  Because EA used the name to refer to Tommy Gun weapons in the games and the plaintiff presented no evidence of any confusion as to the source or content of the work, the First Amendment defense applies, the court held.  The court noted that even if it accepted plaintiff’s contention that the relationship between Dillinger and the games is attenuated, the threshold for “artistic relevance to the underlying work” is very low—it just needs to be above zero.

This is a good decision for game developers in that it presents a case where the artistic relevance of the use of the name to the underlying work is not as strong as it has been in other cases.  But the case will likely to be appealed to the Seventh Circuit according to plaintiff’s counsel.

It is important when developing a game to consider the artistic relevance of the use of celebrity names, likeness, and trademarks in video games.  Where there is no relevance, the courts have found that the First Amendment defense does not apply.

Hall of Famer Jim Brown Fails to Break Through Electronic Arts’ First Amendment Defense

 

Note of use of Brett Favre's likeness on front cover

Judge Florence-Marie Cooper of the Central District of California recently dismissed Jim Brown’s claim of false endorsement against Electronic Arts.  The decision continues the courts’ trend of finding that the First Amendment provides a complete defense to a claim of unfair competition when source identifiers are used within video games.

Jim Brown is a retired professional football players who is revered as one of the best football players of all time.  Electronic Arts (EA) develops and publishes video games, including the popular Madden NFL series.  The Madden NFL game is a virtual football game that contains up to 170 virtual teams and 1,500 virtual players.  The virtual players include players that wear the names and numbers of current real-life players playing on real-life teams.  

EA has an exclusive license from the NFL to use the names and numbers of current players.  The virtual players also include players from historical teams that are anonymous, represented only by number and roster positions.  The players compete in virtual stadiums with virtual fans and virtual coaches, all of which are designed by graphic artists.

Jim Brown alleges in his Complaint that EA used his name, identity, and likeness by including him in two of the historic teams used in the game, namely the 1965 Cleveland Browns team and the All Browns team.  The character in the game that purportedly represents Brown is anonymous and wears the number 37.  Brown wore number 32.  Brown’s statistics and the character’s statistics are “nearly identical.”

Section 43(a) of the Lanham Act provides a federal cause of action for the unauthorized use of a celebrity’s identity (such as visual likeness, vocal imitation, or other uniquely distinguishing characteristic) in a manner which is likely to confuse consumers as to sponsorship or approval of a product.  In the instant case, the Court assumed for purposes of deciding EA’s Motion to Dismiss that the game did use Brown’s likeness.  EA argued that the First Amendment provides a complete defense to Brown’s claim of false endorsement.

In analyzing the First Amendment defense, the courts first determine what type of speech is at issue.  Commercial speech is not entitled to as much protection as non-commercial expressive speech.  However, the courts have found that video games are expressive works that are entitled to as much protection as the most profound literature.  

To evaluate the First Amendment defense, the courts have applied the Rogers balancing test established by the Second Circuit, which provides that a Lanham Act claim against the creator of an expressive work can survive only if “the public interest in avoiding consumer confusion outweighs the public interest in free expression.”  

The Rogers test has two prongs that must be satisfied to find that the First Amendment bars a Lanham Act claim.  First, the use of the trademark (or in this case, the use of the likeness of Jim Brown) must have at least some relevance, however minimal, to the underlying work.  Second, the use of the trademark must not explicitly mislead consumers about the source or content of the work.  Even though there may be a risk that a consumer will wrongly believe that a celebrity endorsed a work, that risk may be outweighed by the public interest in artistic expression.

In Brown, the court found that the Madden NFL series of sports-based games contain numerous creative elements.  Specifically, the game contains virtual stadiums, athletes, coaches, fans, sound effects, music, and commentary created or compiled by the games’ designers.  The court found that these elements represent the designers’ creative interpretation of real-world NFL game play and that the designers’ use of a realistic sports theme “does not change the fact that the Madden NFL games manifest their designers’ creative interpretation of real-world NFL game play.”  The court held that the “Madden NFL video games are expressive works, akin to an expressive painting that depicts celebrity athletes of past and present in a realistic sporting environment.”

Upon finding the video game to be an expressive work, the court went on to apply the Rogers balancing test.  The Madden NFL games are about football. Brown is a legendary NFL football player, the best ever according to some.  Thus, the court found that the use of a legendary NFL football player’s likeness in a game about NFL football is clearly relevant, satisfying the first prong of Rogers.  

In regard to the second prong, the Court found that the use of Brown’s likeness in the game does not explicitly mislead consumers as to the source or content of the work.  Significant to this finding is the fact the Brown’s character is one of thousands of virtual athletes used in the Madden NFL games.  Unlike most of the other characters, Brown’s character is anonymous, identified only by his number and roster position as a running back.  Brown’s character is not depicted on the games’ packaging, advertisements, or promotional materials.  The court held that the mere presence of an anonymous, mis-numbered character in the game does not constitute an explicit attempt to convince consumers that Brown endorsed the game.

Although the court discussed the fact that Brown was not expressly identified in the games either by name or number in reaching its decision, the result should not have been any different even if his name and number had been used  as the Court “assumed” that Brown’s likeness was used in the game.  Thus, it follows that, based on this decision, game developers could use players’ names, numbers, and likeness with impunity in a video game so long as the use of the characters is relevant to the game itself and there is no explicit claim of endorsement by the player.

Such a result would pave the way for development of more professional football games as currently the NFL has granted an exclusive license to EA to use current players in the video games.  EA enjoys a monopoly and would probably like to keep it that way.  However, EA’s successful argument in this case could result in a competitor using the same argument to avoid the need for a license from the NFL, thus taking away some of EA’s market share.  

Can EA have its cake and eat it too?  The Brown decision does not make a player’s likeness free for all purposes.  While the players’ likeness can be used in the video game itself without permission, use of a player’s likeness or name on the packaging for the video game or in promotional materials would probably cross the line and fall outside the protection of the First Amendment because in that case, the name or likeness is used in a manner so as to cause a consumer to believe that the video game is sponsored or endorsed by that player.

In conclusion, the First Amendment may give game developers and publishers more leeway than they think in using the name or likeness of a celebrity without paying for a license. However, it is important that the use of the name or likeness have relevance to the underlying work.  The determination of when a use crosses the line and becomes an explicit attempt to mislead consumers into thinking the product is endorsed by the celebrity will be made on a case by case basis.  Yet, the Brown decision solidifies the principle that mere use of a celebrity’s likeness in and of itself does not prove a Lanham Act violation.

Remember the Alamo: Federal Trademark Dilution v. Texas Trademark Dilution

Trademark infringement occurs when a third party uses the same mark or a mark similar to a trademark owned by a trademark owner in such a manner that consumers of the products are likely to be confused as to the source of the products.  Generally speaking, a claim of trademark infringement does not exist when the competing products are not the same and are so unrelated that consumers will not be confused as to the source of each product.  That is, unless the mark is famous.

If the mark is famous, then the trademark owner may bring a federal trademark dilution claim against any party using the mark regardless of whether the trademark is being used for similar goods or services.  The rationale behind this theory of recovery is that the use of a famous mark, even if on differnt goods or services, weakens the commercial value of the mark. 

In 2006, Congress passed the Trademark Dilution Revision Act (TDRA) in response to Moseley v. V Secret Catalogue, Inc. in which the Supreme Court held that the plaintiff must prove “actual dilution” rather than “likelihood of dilution.”  Among other things, the TDRA amended the Federal Trademark Dilution Act to change the standard from “actual dilution” to “likelihood of dilution.”  Thus, proof of actual dilution is no longer required to prevail on a federal dilution claim.

To prove a federal dilution claim, a trademark owner must show (1) that the mark is famous, (2) that the mark is distinctive, (3) that the defendant began using a mark after it became famous, and (4) that defendant’s mark is likely to cause blurring or tarnishment of the mark.  Typically, the most difficult element of proof is showing that the mark is truly famous.  Under the TDRA, a mark is famous “if it is widely recognized by the general consuming public of the United States as a designation of source of the goods or services of the mark’s owner.” 

Most Texans, and indeed many people outside of Texas would recognize the longhorn silhouette used by the University of Texas in connection with the university’s athletic programs.  In Board of Regents v. KST Electric, Ltd., the university filed suit against the defendant for using the longhorn silhouette in connection with the logo for its business. 

In response to a motion for summary judgment by the defendant on the university’s federal dilution claim, the university presented evidence of how its football and other sports programs had acquired fame.  The court held that the university had only shown evidence of “niche fame” because evidence of fame among sports fans is not evidence of fame among the general consuming public.  The TDRA was enacted, in part, to get rid of “niche fame” by requiring a mark be “widely recognized by the general consuming public of the United States.”  “One of the major purposes of the TDRA was to restrict dilution causes of action to those few truly famous marks like Budweiser beer.”  Therefore, the court held that the university had not created a genuine issue of material fact as to whether the longhorn silhouette logo is “a household name.”

It is interesting to note that in KST Electric, the defendant did not move for summary judgment on the state dilution claim.  Why the defendant only moved for summary judgment on the federal dilution claim and not the state dilution claim in all likelihood relates to the broader scope of a Texas dilution claim. 

To prove a claim of dilution under Texas law, the trademark owner must show (1) that the mark is distinctive and (2) that defendant is making significant use of a very similar mark in a manner that is likely to cause dilution of the trademark owner’s mark by blurring or tarnishment.  Whether a mark is distinctive is determined by considering  (1) whether the mark is arbitrary, (2) the length of time a user has employed the mark, (3) the scope of the user’s advertising and promotions, (4) the nature and extent of the first user’s business, and (5) the scope of the first user’s reputation. 

Thus, there is no requirement that the mark be famous, and while it may be necessary to show that the mark is a strong mark, it is not necessary to prove that a mark is a household name to prevail on a Texas dilution claim.

By Vincent J. Allen