Trademark Licensing in Keyword Advertising

Internet advertising can lead to a number of complex problems related to trademark infringements, in particular concerning consumer confusion. The aim of this paper is to analyse the view taken by the Court of Justice of the European Union (CJEU) regarding the likelihood of consumers’ confusion in respect of online advertising campaigns. The paper additionally discusses the negative consequences of broadening the concept of taking unfair advantage and separation of this concept from likelihood of confusion or a likelihood of detriment to the distinctive character or the reputation of the trademark. Finally, it focuses on the possibility of introducing licensing models, which would potentially strike the right balance between conflicting interests of the trademark holders, advertisers and consumers.

Everyone who uses search engines knows that after typing a particular word the web browser engine displays two types of results. The first of them, which also takes up the most space on the website are links to websites, which are “natural” results of the search, i.e. websites indexed by web browsers as the most closely connected with the keyword (the so-called “organic” results). The order in which links are displayed is subject to complex algorithms, which takes into account a number of factors; in particular the number of views and amount of websites referring to specific keyword. The more frequently the website is visited, the higher it will be displayed within the organic results of the search, thus it would mimic an Internet user’s natural search behaviour.  [1] It is worth noting that usually web browsers strictly protect the structure of algorithms and modify them to prevent the algorithm from potential manipulations by advertisers. It should also be emphasised that advertisers have no legal way to affect the order of the organic results of the search. [2]

The other results displayed by web browsers’ engines are those which are generated by paid search services. Despite the fact that nearly every search engine has such a service, this article will refer only to the Google AdWords service as the key example of contextual advertising. [3] Links displayed in Google AdWords are displayed above organic results or next to them and are distinctly marked as “Ad”, “Sponsored” or “Sponsored links”. The advertiser may choose between a variety of formats – most commonly advertisements show text consisting of descriptive sentences, a link to the advertiser’s website, and the website’s URL address. The order of sponsored links is determined by: keywords chosen by advertiser; the amount the advertiser is willing to pay for every “click” on the link to his or her webpage; and the number of views of an advertisement. [4] More than one advertiser may reserve particular keywords. This enables the creation of “sponsored links campaigns”, which allow displaying links to the advertiser’s webpage in positions visible for web users regardless of webpage position in organic search results. The revenue generated by the sponsored ads makes up a large part of Google’s revenues. Most importantly, the advertisers may also choose words, which are registered trademarks as keywords, consequently leading to potential trademark litigations.

The protection of trademarks afforded by  [5] the Directive 2015/2436 of the European Parliament and of the Council of 16 December 2015 (hereinafter referred to as “TMD”) is twofold. Firstly, art. 10(2)(a) states, that the proprietor of a trademark may prevent other parties from using this mark without his or her consent when the third party uses a sign identical to the trademark in relation to the goods or services which are identical with those for which the trademark was registered. Additionally, art. 10(2)(b) entitles the owner of a trademark to restrain a third party from using an identical or similar sign to the trademark in relation to identical or similar goods when there exists a likelihood of confusion on the part of the public, which in particular includes the likelihood of association between the sign and the trademark. Secondly, art. 10(2)(c) allows the proprietor of the registered trademark to prevent all third parties who have not obtained consent from using in the course of trade any sign which is identical with, or similar to, the trademark in relation to goods or services which are not similar to those for which the trademark is registered, where the latter has a reputation in the Member State and where use of that sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or the repute of the trademark.

Despite strict limitations on the protection of trademarks included in art. 10(2)(a) and art. 10(2)(b) of the TMD, the CJEU pointed to the functions of the trademark when determining the scope and limits of trademark protection. [6] However, unlike the statutory limitations, these functions, which are described as “functional equivalent to limitations”,  [7] are not mentioned or defined in the Directive. Instead they should be “conceptualized as being inherent to the exclusive right and have to be carved out by case law”. [8]

By virtue of art. 10(2)(a) and (b) of the TMD, trademark owners are given control over communication concerning their marks, which covers the identification and distinction of the goods or services offered in the marketplace. Consequently, protection afforded under art. 10(2)(a) and 10(2)(b) is to be granted only if use of a conflicting sign is likely to cause confusion to the consumers. In case of the double identity rule (when an identical sign is used for identical product) embodied in art. 10(2)(a), the risk of confusion may be deemed so obvious that it can be presumed. Nevertheless, as highlighted in Senftelben,  [9] existence of such a factual presumption should not encourage the court to deviate from the general requirement of evidence that the likelihood of confusion may arise from the use at issue. As it has turned out in case law of the CJEU, the fact that the third party merely used a registered trademark does not necessarily lead to consumers’ confusion. This occurred in the Arsenal case,  [10] where the defendant offered identical goods to those of the trademark holder using a mark, which was identical to the registered trademark of the claimant. Nevertheless, the defendant displayed signs which informed clients that its products were unofficial. Consequently, consumers could not have been confused as to the origin of the goods. Despite that, the CJEU stated that such a use of a registered trademark constitutes infringement under art. 10(2)(a) if it is capable of affecting one of the functions of the registered mark. As a consequence, the CJEU has interpreted the TMD as protecting a registered trademark against non-trade mark use by a third party due to a likelihood of confusion.  [11] In this particular case, it was also held that this type of use affects the original functions of a registered trademark, as the use of the mark was likely to confuse subsequent buyers who were unaware of the defendant's information regarding whether there is a link in the course of trade between the goods and the trademark. More notably, it was the first time the CJEU referred to functional interpretation of the protection of trademarks. It would thus just be a matter of time before the court determined other functions worth protecting. In the L'Oreal  [12] decision the CJEU continued this path of extended trademarks' protection by assuming that art. 10(2)(a) protects “not only the essential function of the trade mark, which is to guarantee to consumers the origin of the goods or services, but also its other functions, in particular that of guaranteeing the quality of the goods or services in question and those of communication, investment or advertising”. As a result, even if the origin function is not threatened, the double identity rule embodied in art. 10(2)(a) of the TMD may be violated if any of the other functions of the trademark are affected.  [13]

On a different occasion, the CJEU expressed its view that the function of indicating the origin of the trademark is adversely affected when internet users are shown a third party’s advertisement on the basis of a keyword being identical to a mark. [14] The Court continued this opinion by stating that:

the function of indicating the origin of the mark is adversely affected if the ad does not enable normally informed and reasonably attentive internet users, or enables them only with difficulty, to ascertain whether the goods or services referred to by the ad originate from the proprietor of the trade mark or an undertaking economically connected to it or, on the contrary, originate from a third party. [15]

Finally, the CJEU established that:

where the ad, while not suggesting the existence of an economic link, is vague to such an extent on the origin of the goods or services at issue, that normally informed and reasonably attentive internet users are unable to determine, on the basis of the advertising link and the commercial message attached thereto, whether the advertiser is a third party vis-à-vis the proprietor of the trade mark or, on the contrary, economically linked to that proprietor.  [16]

In the Interflora case,  [17] where the defendant – Marks & Spencer used its competitor’s name “Interflora” as a keyword to advertise its own flower delivery services, the CJEU followed the path once established in the L’Oreal case. The CJEU established that the use of the word “Interflora” as a trademarked keyword may cause confusion by insinuating that the flower delivery service offered by the defendant is part of Interflora’s commercial network. The CJEU concluded that “the advertising at hand does not allow it to be determined whether M & S is a third party in relation to the proprietor of the trade mark or whether, on the contrary, it is economically linked to that proprietor”. [18] Consequently, in those circumstances it is believed that the function of the original trademark would be adversely affected. It is however important to underline, that contrary to the L’Oreal case, the defendant’s advertisement or website did not contain any reference to the plaintiff’s trademark.

Hence, it shall be considered whether an average consumer may be confused by a sponsored link, particularly if a trademark is used neither in the text of an advertisement (i.e. sponsored link) nor on the advertiser’s website. Primarily, an average consumer typing a trademarked keyword is aware that the website of the trademark holder will be displayed in organic search results or - eventually – in sponsored links. In such an event it is dubious to believe that a well informed and reasonably attentive web user would consider all or even just more than one of the results displayed after entering a specific keyword, as results referring to websites of service providers are economically connected to the owner of the trademark. [19] This may be more problematic in the case of trademark owners, which supply their goods through diverse distribution networks using several official distributors. In such a scenario, consumers may be more vulnerable to confusion as it would be more difficult to assess which distributor is an official dealer and which one is not. Nevertheless, the majority of consumers should be aware that in most cases, searching for a trademarked keyword in Google may result in the search engine displaying links to websites of the trademark holder’s competitors proposing an alternative to the products or services of the trademark holder, as well as a variety of links referring to websites that are not offering any products or services, but rather comparing the prices of services and goods of various businesses in the same industry, or reviews for specific goods or services.

It is also possible that some of the consumers who enter a trademarked keyword are searching for competitors of a trademark holder – as a result, links referring them to competitive websites would not be confusing to this particular group of web users. [20] Such an assumption is all the more convincing according to the belief that consumers pay less attention to advertisements due to prior negative experiences with them – i.e. obstructing the consumer from purchasing a product directly.  [21] However, according to the same research, ninety to ninety-one percent of the more experienced consumers look at the sponsored links that appear above the organic results. Despite this behaviour, it can be presumed that those consumers are less likely to be confused because compared to the average user, they are more aware of the way search engines work.  [22] Taking this consideration into account, the view taken by the CJEU in the Interflora case may not hold as it does not take into consideration how the consumers perceive the search engines. It can also be argued that the CJEU did not introduce a reasonable test for ascertaining whether the consumers were actually confused. The requirement to prove difficulties with assessing whether the goods or services referred to by the advertisement originate from the proprietor of the trademark, or an undertaking economically connected to it or, on the contrary, originate from a third party, is not particularly clear and possibly sets the standard of proof too low. This would be comparable to the problematic American doctrine of initial interest confusion. Consequently, simply diverting a consumer’s attention may be tantamount to the likelihood of confusion, which would be an unrealistic assumption taking into consideration the awareness of average web users. A keyword that triggers an advertisement outside the hit list, which does not contain the senior trademark, should not infer that it originates from the competitor identified by the search term, or that there is a business connection between the competitor and the advertiser.

Following the CJEU approach taken in the Interflora case, in the brick-and-mortar world, it should be concluded that deliberately placing two competing goods on the same shelf amounts to trademark infringement. In addition, it shall be noted that in response to a question regarding certain brand of shoes, a shop assistant will lead us to a shelf where shoes of other brands are also located. As a result, the consumer would encounter competitors’ shoes, despite asking for a specific brand. There is little to no likelihood that an average consumer would conclude that the producers of such products are economically connected just because they are placed on shelf next to each other. [23]

Consequently, without the need to prove any possible confusion, the CJEU’s reasoning indicates that a slight diversion of the web user’s attention has an adverse effect on the origin-indicating function of the trademark. Such an approach may undoubtedly have a negative influence on the functionality of web search engines. Consumers using Google’s web search engine and typing a trademarked keyword may not only be searching for the website of the trademark holder, but rather for a whole variety of websites related to that product or service. [24] If, because of such a search, the consumer decides to visit the website of an entity which is not the trademark holder – it would not be caused by confusion, but rather because the consumer accepted that by virtue of using the web search engine and clicking on a link referring to a competitor’s website. Evidently, it is crucial that advertisements and websites do not use the trademark – otherwise the boundaries of the legitimate use of a trademark would be crossed.

It should also be considered that even if the sponsored link would mislead a consumer, then a consumer would be aware of this whilst accessing the website – a consumer would find out that the accessed website is not what he or she was searching for. In such circumstances, the consumer may just click “back” (which is effortless and natural for every internet user) and go back to the list of results. In such a case, the competitor may of course gain advantage of an additional visitor to the website generated on the basis of keyword advertising; however, this is rather connected to the aspect of taking unfair advantage, which will be discussed below. Therefore there is likelihood diminished possibility of confusion or undermining the functionality of the web search engine, as the searching does not become much more costly or time consuming, and the mere occurrence of links to competitors’ websites hardly invokes confusion among consumers. Consumers searching online expect to find results to various websites, from which they may choose to click. [25] A consumer’s confusion may be dependent on the layout of the advertisement and the content of the website. However, no confusion shall arise if the sponsored link does not use a misleading layout and is placed in the position typical for sponsored links.  [26]

In addition, it shall be noted that there is high likelihood that the consumers perceive the organic search results as more relevant than sponsored listings. This notion is backed by the results of an eyeball tracker study, where consumers were found to look at the first couple of organic results before looking at the advertisements. [27] It is thus believed that consumers are likely to assume that the most appropriate results will appear near the top of the organic search results since the relevance of those websites qualifies them for a top spot in an unpaid ranking system.

To counter balance those critical voices, there exist opinions, which support the current view taken by the CJEU. They argue that the mere appearance of a defendant’s website on a search engine’s results list necessarily indicates consumer confusion at a certain level as comparing to the brick-and-mortar context, it has been recognized that confusion can occur at “subliminal levels” which is, according to some authors, very similar to confusion arising from search engines.  [28] It is argued that the side-by-side comparison-shopping experience should only occur if that is what the consumer requests when he or she enters a categorical term in the search engine.  [29] Consequently, it is believed that an actionable harm based on confusion of the consumer shall be permitted even if the consumer is not aware of the confusion at the time.

In regards to analysing the advertising function of a trademark, this function is adversely affected if the third party's use denies the proprietor of that mark the opportunity of using its mark effectively to inform and win over consumers. [30]

When the use of a trademark by a third party affects the proprietor’s use of its mark as an aspect of a sales promotion or as an instrument of a commercial strategy this infringes upon the advertising function. [31] The CJEU stated that the use of a keyword identical to an existing trademark in the AdWord service does not affect the advertising function of a trademark, as it does not deprive the trademark holder of the use of the trademark to inform and persuade consumers. The CJEU reasoned that there is no infringement upon the advertising function if the trademark holder has to pay more per single click on their own sponsored link than his or her competitor who also chose a trademark as a keyword in the AdWord service. [32] The basis of such a conclusion is that the trademark holder’s website would be always displayed on the top of natural results after entering the trademark word.  [33] The view taken by the CJEU is questionable due to the following reasons. Firstly, the trademark holder may not possess any website – then there is clearly an exploitation of the trademark’s expenses on its conventional advertising in the promotion of the mark. Secondly, it is worth considering whether all entities entitled to use the trademarks have a right to a high place in the natural results. Having a high ranking website among the list of natural results is not necessarily the due to a “natural” process. [34] The occurrence of websites displayed in natural results requires specific skills of a web developer in the field of search engine optimisation. [35] This applies especially to holders of trademarks with a reputation, because many websites may refer to their trademark (e.g. the word “Coca-Cola” may appear on a number of websites and all these websites affect the position of the trademark holder’s website in organic search results). There is no doubt that displaying a newly registered trademark holder’s website among the top-ranking natural results requires financial expenditure (it is worth noting that most of the keywords are descriptive terms, consisting of existing words, which results in a high likelihood of existence of websites referring to single words constituting a trademark). Furthermore, if obtaining a high position among natural results was a given for all trademark holders, then no trademark holder would decide to use services such as Google AdWords. This leads to another problem related to the advertising function of the trademark. Trademark holders must oftentimes bid on their own marks in search engine keyword auctions to prevent competitors from appearing higher than their unpaid search result. [36] If the trademark holder does not decide to do so, it may lose the opportunity to inform consumers interested in its trademark in the first place about its products. Consequently, the trademark holder is forced to additional financial expenditures in order to maintain the level it achieved through engine optimization just because other entity can freely bid on its trademark as a keyword. As discussed in academic writing, many trademark owners would probably not choose their own trademark as keywords if competitors and other entities were not allowed to do so.  [37] Bearing that in mind, one should consider whether the approach taken by the CJEU, which categorically denies any impact of contextual advertisement on a trademark’s advertising function, satisfies all the nuances and complexities of e-marketing. There are some concerns regarding whether or not the trademark holder would be entitled to some compensation when it is exposed to greater financial burdens in order to catch up with its competitors due to the use of its trademark as a keyword by those competitors. Thus, it seems that the approach taken by the CJEU requires a more thorough analysis of the technology used in online advertising campaigns.

According to the CJEU, a trademark has an investment function if its proprietor uses it to acquire or preserve a reputation capable of attracting consumers and retaining their loyalty. [38] In such circumstances, an investment function may be interrelated with an advertising function. However, it should be emphasized that the use of a trademark to acquire or preserve a reputation in the scope of an investment function may occur through other trade techniques than advertising. If the trademark already enjoys a reputation, an investment function is adversely affected if use by a trademark holder’s competitor of that mark affects that reputation negatively [39] - however, in such scenario article 10(2)(c) of the TMD, which concerns dilution applies. The CJEU took the view that there is no adverse impact on a trademark’s investment function, if a competitor is using a trademark with respect of trademark’s function as an indication of origin and if the only consequence of that use is to oblige the proprietor of that trademark to adapt its efforts to acquire or preserve a reputation capable of attracting consumers and retaining their loyalty. [40] Thus, no investment function of a trademark is affected if, as a result of a competitor’s use of trademark, some of the consumers choose competing goods and services instead of goods and services labelled with the trademark. Such analysis is - to some extent – convergent with an adverse affect of the trademark’s advertising function, which is causing some difficulties in determining the scope of both functions. The CJEU is of the opinion that neither a need to intensify advertising, nor a potential loss of clients justifies preventing competitors to use a trademark in the AdWord service.

The use of the trademarked keywords also triggered many concerns regarding the impact of such a use on the reputation of well-known trademarks. Art. 10(2)(c) of the TMD introduce a concept of anti-dilution protection to the European trademark law. Dilution occurs when an entity uses a mark identical or substantially similar to a pre-existing trademark “triggering a mental association on the part of the consumer between the two marks, thereby eroding the strength of the original mark”. [41] Art. 10(2)(c) of the TMD consists of provisions which allow the prevention of using of a senior mark where use of that sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or repute of the trademark.

The concepts of detriment to the distinctive character or the reputation are further defined by the General Court in the Intel/CPM case [42] in two manners: as blurring, which is detrimental to the distinctive character of a trademark which attempts the exclusive link between the product or service and the owner of that trademark, or tarnishment, which provokes the association of the trademark with goods or services that have a negative connotation or are incompatible with the image of the mark. The CJEU, in the Google France decision stated that taking unfair advantage of the distinctive character or repute of the trademark, referred to as the concept of free riding covers cases in particular where, “by reason of a transfer of the image of the mark or of the characteristics which it projects to the goods identified by the identical or similar sign, there is clear exploitation on the coat-tails of the mark with a reputation”.  [43] The justification for entitling the trademark right holder to ban third parties from taking advantage of a trademark is based on the assumption that the trademark right holder has spent resources on the creation of a reputation of the mark, and that third parties take unfair advantage of the distinctive character or repute of that trademark.  [44] Furthermore, the Court assumed that:

the taking of unfair advantage of the distinctive character or the repute of a mark, within the meaning of that provision, does not require that there be a likelihood of confusion or a likelihood of detriment to the distinctive character or the repute of the mark or, more generally, to its proprietor.  [45]

Consequently, any advantage taken by the third-party of a trademark with a solid reputation is by default unfair, even when this use did no damage to the well-known trademark. [46] Such an outcome met considerable criticism from academic commentators, who observed, that by setting a low criteria for taking unfair advantage, a loophole is created when evidence of detriment is impossible. [47] Furthermore, critics argue that it is not enough to demonstrate that free riding has occurred, but it shall also be illustrated that legal intervention promotes efficiency by taking account of transaction costs, imperfect information etc.  [48] Otherwise, in the brick-and-mortar world, the protection of the mere level of attention against the taking of advantage may result in a situation where “a person who invested time and money in developing an attractive garden might be entitled to compensation from the neighbouring property owners as they benefit from owning a house on a ‘well-kept street’”.  [49]

As a result, the defendant was not entitled to advertise their products as alternatives to the products offered by the trademark proprietor by simply describing them as similar to the reputed trademark, irrespective of the fact, that no detriment to the distinctiveness or repute of the famous trade mark occurred. It is of essence to highlight, that products offered by the defendant in the case at hand were described as similar and not the same - the reputed trademark acted only as a referencing label. In effect, the public could not be accurately informed about the features of the defendant's products.

This controversial trademark centric-approach was echoed in the Interflora decision. The Court assumed that the investment which a proprietor has put into the mark to attract consumers may be jeopardised if a large number of consumers using the keyword will see the competitor's advertisement and decide to purchase its goods or services rather the ones originating from a reputed brand. [50] Such reasoning relates to the arguments raised in L'Oreal, however widens the concept of riding on the coat-tails of a trademark with a reputation in order to benefit from its power of attraction to circumstances, when the selection of a sign similar or identical to a trademark with reputation refers only to keywording in Internet search engines. Accordingly, use of prominent trademarks in a descriptive way in order to indicate that defendant's products are similar to those denoted by the well-known trademark is prohibited even when the advertisement triggered by the keyword does not contain the sign. [51]

In the circumstances of the Interflora case, the advertiser undoubtedly takes advantage from the use of the interflora trademark as keyword. The choice of the term interflora and its various options by Marks & Spencer is aimed at attracting potential consumers to competing services. The factor determining the advertiser’s choice of keywords is recognition of the trademark on the market and the reputation of the mark Interflora and to enable the customers to associate it with a particular type of service while keeping in mind that a vast number of consumers associate those services only with the operation of Interflora Inc. It is thus worth considering whether the advantage taken by the advertiser as a result of the mentioned campaign should be automatically qualified as unfair advantage.

For this reason, two ways of understanding of the word Interflora by potential consumers should be analysed. The first refers to the group of people identifying the word only with the trademark owned by the particular business enterprise. These people enter the word Interflora into a web browser in order to find the website of the proprietor of the Interflora trademark and to choose its service. Even if this group of consumers accidentally enter the Marks & Spencer website, there is a very low possibility that they would decide to use the services provided by it, because they are determined to use the service provided by Interflora and are not interested in any alternatives.  [52] These kinds of consumers do not constitute Marks & Spencer’s potential customers, and the web searching engine is treated by them only as a way to reach Interflora’s website.

The second group consists of consumers who choose as a keyword a term they associate the most with flower delivery services, but are not strictly determined to use the services provided by the well-known brand. They may probably be interested in competing offers but they use the trademarked keyword to locate this kind of service on the Internet. [53] Such an approach, which allows a competitor to indicate an alternative to the services offered by the most dominant enterprise within the market, seems to be desirable from the perspective of protection of competition. In the Interflora case, the aim of using the trademark by the defendant was to make consumers recall the unique method of Interflora’s operation as florist network. In such a situation, advertisers do not use the reputation of the trademark, but rather the way in which its services are organized. Thus, the use of Interflora in this case can be treated merely as a use of a trademark in order to describe a certain class of services and offer them to the public.

If the word Interflora is associated with the method of operating of florist network and the trademark holder is so dominant on the market that the average consumer does not know any other entrepreneur providing competing services, then it should be assumed that a number of internet users, by entering word Interflora into the web search engine, are doing that in order to find providers of services of this kind.

In light of above, it would be reasonable if the CJEU draw its attention to the lack of conceptual designates for keywords within the paid search services. Keywords as such have no meaning in reference systems used in contextual advertising because the algorithms of web search engines address the signs regardless of their meaning in any language and without the context in which they were used.

As a result, the particular word reserved by the advertiser obtains designates only in the mind of the web user. [54] One should not exclude a situation in which the web user enters the word Interflora into a web search engine, e.g. to take up employment. In such a situation, the word Interflora does not refer to a trademark or description of the method of conducting business activity, but to a designated employer. The situation becomes even more complicated if one considers that some trademarks are simply natural words, or that there are cases where certain combinations of words have become trademarks as a result of their registration or widespread use in specific industries. [55] Consequently, trademarks have an idiosyncratic nature, which means they can simultaneously transmit multiple meanings.  [56] Furthermore, trademarks are being registered in different jurisdictions and for different classes of goods or services, which can cause further problems. To illustrate: it is easy to imagine that in the real world no salesman would have a problem with determining whether a consumer is asking about a coconut chocolate bar or paper towels when they ask for a product called Bounty  [57] despite the fact that for both products the same verbal trademark is reserved. In practice, consumers just do not confront trademarks in an abstract way very often, and, when they do, context usually makes the product category obvious. However context is what the web search engine lacks and thus restricting the possibility to reserve keywords within services such as Google AdWords may undermine the functionality of the web search engine. It should also be emphasized that interrelation between the use of a trademark and the context of this use may adversely impact on the ability of a trademark with a reputation to remain distinctive. [58]

The Interflora judgement clearly refers to a relation between the need for descriptive use, which allows a competitor to inform consumers that they offer an alternative to the leading brand, and the need to protect reputed trade marks from “the risk of being hurt by their own success and becoming generic”. The CJEU assessed that the use of a reputed mark in keyword advertising by a third party is detrimental to the distinctive character of this trademark if that usage may contribute to turning that trademark into a generic term. The use of a sign similar or identical to a trademark with a reputation is allowed if the advertisement message displayed on the basis of keywords corresponding to that trademark suggests that they represent an alternative to the goods or services of the proprietor of that mark. Moreover, the advertisement cannot offer a mere imitation of the goods or services of the proprietor of that trademark and may not tarnish or dilute or adversely affect the functions of a trademark.

This part of the Court’s reasoning represents the due cause defence under which the use of trademarks with a reputation as keywords may be acceptable. The use of a trademark in order to describe a certain class of services and offer them to the public shall be classified as a descriptive use. It is believed that there is a difference between descriptive use and a trademark becoming generic, although in some extreme circumstances descriptive use may cause the trademark to become a generic term.  [59] Despite that, the proprietor shall be able to effectively protect its trademark from becoming generic by virtue of various marketing techniques. Otherwise, the right balance between free competition and trademark protection may be jeopardised. This is evidenced by the fact that in most cases, owners of reputed trademarks would be chosen as reference labels for the type of goods or services due to advertising,, and as a result gain wide popularity.  [60] The consumers simply mostly know the brands which they are exposed to, and at some point they may even associate the whole class of goods or services with this brand. Consequently a well-known trademark may turn into a generic term notwithstanding its competitor’s marketing strategies. If we point to the risk of a trademark becoming generic due to competitor’s actions it must be emphasised that this well-known trademark must have already been a reference label for a certain class of goods or services at the time of launching the keyword advertising campaign. Otherwise, bidding for the trademarked keywords by the competitor would not make any sense. Furthermore, the requirement to display a message, which suggests that the goods or services advertised represent an alternative to the goods or services of the trademark holder, seems excessive, as it is unreasonable to expect that a consumer may be confused regarding the origin of the goods advertised by links triggered by the keyword. If no confusion occurs, no taking of unfair advantage shall occur as well. It is interesting how such information which suggests that the goods or services advertised represent an alternative to the goods or services of the trademark holder should look like, especially given that advertisements located under links are very short. It shall be also questioned why the holder of the junior mark cannot use the senior mark to inform consumers about their products, if it is perfectly legal to sell such products. Such an approach renders selling legal replicas of well-known products nearly impossible.

Moreover, in light of academic writing, the phenomenon of turning a trademark into a generic term is the ultimate weakening of a trademark and makes the proprietors of the most famous trademarks "victims" of their own success.  [61] However, one should bear in mind that the right to use someone else's trademark in a descriptive manner is necessary to preserve competition and to inform consumers about alternatives to products or services offered by leading brands. While it is important to educate the consumers to use descriptive terms for goods or services offered by leading brands while using web search engines in order to reach their competitors, there are trademarks which became almost synonymous with certain classes of goods or services (to name just a few – a Walkman for personal stereo, Vaseline for petroleum jelly, Tarmac for asphalt road surface, or even Google itself as a generic term for internet search engine). It would be very difficult to change consumer’s habits and request them to search for neutral descriptive terms. It is hard to think that in those situations a consumer’s search would be aided by the ability to remember the product category associated with a brand. Moreover, if a consumer is thinking about purchasing certain goods or services, it may be helpful if the consumer can remember several relevant brands so that he or she can choose among them.  [62] In some extreme situations (the abovementioned example of the Vaseline trademark) it would even be unlikely that a consumer might be aware that such a term is a trademark and not a generic term. It would also be unrealistic to expect an average consumer to know the descriptive term for a product, which holds the Vaseline trademark. Furthermore, although trademarks are in general excluded from registration if they are descriptive or devoid of a distinctive character, word signs registered as trademarks do not appear out of a vacuum as they are commonly built out of the existing words. In some situations, they are even descriptive terms that gained their secondary meaning as a trademark (e.g. the North Face for outdoor clothes), because of the use that is made of them, consumers also come to see them as a badge of origin. The other group of trademarks consists of signs, which are already known and exist as part of natural language. It is their use as a trademark that makes them unique, non-trivial, non-figurative – a perfect example of such a word sign is the Apple trademark for computer devices. Consequently, in some cases, policy of restricting the use of trademarks on the Internet may lead to absorbing natural language by trademarks, a result which would be highly undesirable.

As a side note, it shall be considered that in some rare circumstances the process of transforming a trademark into a descriptive term, whilst maintaining the meaning as a badge of origin, may bring positive effects for the trademark holder. This is encapsulated by the fact that consumers may positively associate a particular trademark with high quality products. We shall not forget that such an outcome is one of the reasons why the trademark holders invest in advertising. A good example of this process is illustrated by the fact that Mercedes trademark conveys a level of quality regarding the automobile industry and thus is often used - at least in Poland - to describe high quality goods or services. This happens because the consumers associate the Mercedes trademark with high quality in general as a result of Mercedes’ brand history and marketing techniques. However, it must be stressed that the trademark holder desires such transformation only if the connotation is positive and if the trademark maintains its character as a badge of origin. Therefore, this process may be only positive if it is under the full control of the trademark proprietor.

Firstly, it should be noted that the CJEU inconsistently applies the theory of a function of a trademark. The current approach to the function of origin indication should be reviewed as it does not consider the nature of the use of trademarks in online search engines such as Google AdWords and the level of consumer awareness when it comes to web browsing and confusion. The question of whether Internet users can easily distinguish whether or not the competitor is part of the trademark’s proprietor network is relevant in assessing whether there is unfair advantage or dilution. Thus, the approach towards the use of trademarks in the digital environment should rather be more focused on the behaviour of the web users. As a result, the trademark holders’ competitors ought to be given more flexibility when structuring their e-marketing strategies. However, the CJEU ignores the adverse impact of keyword advertising on the advertising and investment functions of the trademark.

Furthermore, the scope of protection of the trademark with a reputation afforded in the EU legal system appears to be too broad. A complete ban on use of the trademarks on the Internet due to expected problems with rebutting the presumption of taking unfair advantage of the distinctive character of the senior mark by its competitors and unclear rules on assessing consumers’ confusion may in some circumstances lead to a phenomenon of absorbing the use of natural language in the Internet by the proprietors of the strongest trademarks. It shall be underlined that there are well known and trademarked keywords which are hardly substitutable descriptive terms or consisting of natural words; therefore, introduction of any limitations, which disallow usage of those words as keywords, would be mostly detrimental for the consumers.

Thirdly, the view taken by the CJEU that taking any advantage of using the competitor’s trademark always denotes unfair advantage, contradicts the functionality of web search engines, reduces the benefits that the Internet brings to consumers, and hinders the development of the free market. Giving business enterprises absolute control over the use of their registered trademarks may, in consequence, jeopardize the essential function of the Internet, which is the free flow of information. Additionally, it may result in reducing the potential of web search engines to process commercial information. The due cause defence may only be taken into consideration if the web users have no difficulties with assuming from the competitor’s advertisement that it is not part of the senior trademark proprietor’s network. Consequently, using trademark law in its current form without any limits, may counterproductively destroy the Internet’s utility for everyone. [63]

As it appears from the judgments of the CJEU, striking the right balance between the consumer’s rights to information and trademark owner’s right to maintain its investment in the trademark on the basis of the law in force is difficult. The preceding analysis justifies the idea of introducing a mechanism within paid search programs, which would allow purchasers of other entities' trademarks as keywords to use those marks if this use would create a stream of revenue for trademark holders. Such a proposal would establish a licensing model for the use of a competitor’s trademark as a keyword in a paid search. [64] This concept originates from the similarities between the position of copyright owners of digital public performance rights for sound recordings and holders of trademarks which are used for keywording.  [65] The idea of introducing a licensing model for the use of trademarks requires taking into consideration the wording of art. 21 of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). According to this provision, states are prohibited from introducing compulsory licenses for the use of trademarks. Consequently, the licensing model requires the creation of royalty structures for the use of a competitor's trademark as keywords in paid search services dependent upon the consent of the trademark holder. The introduction of the licences would possibly require establishing bodies, which would administer the licenses between trademark owners and licensees.  [66] The other aspect that would need to be thoroughly considered is whether the trademark holder shall be free to choose the rate structure imposed on the trademark use. The royalties system may reflect the way in which the search engine providers collect revenues from keyword advertising. Consequently, the royalties model could be structured in the following ways:

In the case of the percent of the click cost of the trademarked keyword structure, the trademark owner would only be paid each time clicking on the sponsored link is triggered by the trademarked keyword. This model would reflect the cost of a trademarked keyword in the trademark owner’s fee. Consequently, if the cost of a trademarked keyword went up or down, the same would happen to the trademark owner's fee.  [67]

In the flat fee per click system, the advertiser would be charged a flat rate every time a web user clicks on the advertisement. Such a royalty structure does not take into consideration the cost of the keyword; thus it provides the trademark holder with a stable stream of income.

The flat fee per impressions structure would charge the advertiser a certain fee every time the advertisement is generated as a result of typing a trademarked keyword into the web browser. This royalties structure entitles the trademark holder to remuneration in case the consumer searches for a specific keyword and the trademark holder’s sponsored link shows up despite the fact that the consumer may divert his or her attention to a different page. As a consequence, the trademark holder would be compensated every time typing its trademark as a keyword triggers the display of the advertisement.  [68] This would allow compensation for the additional expenses incurred by the trademark holder due to the violation of the advertising function of the trademark by the advertiser.

The last of the possible royalties structure - the blanket license, would introduce an annual license (for a minimal fee) for the use of a trademark in the keyword campaigns. Such a model would be well suited for smaller brands, which may be unaware of the fact that their trademark is used as a keyword.

It is clear that there are many possible ways to tailor the royalties system in order to suit different kinds of trademark holders. Consequently, it would be reasonable to leave this choice to the advertisers and trademark holders.

The other issue regards considering whether the use of trademarks in keyword advertising shall not be treated as comparative advertising. According to Article 2(c) of the DMCA, comparative advertising is “any advertising which explicitly or by implication identifies a competitor or goods or services offered by a competitor”. In light of the recent judgement in the Belgian Electronic Sorting Technology case  [69] (BEST) it has been proposed that the use of a trademark in a domain name and that of metatags in a website’s metadata is covered by the term “advertising”, used in the Directive 2006/114/EC of the European Parliament and of the Council of 12 December 2006 concerning misleading and comparative advertising  [70] (hereinafter: comparative advertising directive).  [71] According to previous decisions of the CJEU, in order to reconcile the protection of registered marks and the use of comparative advertising, the proprietor of a registered trademark is not entitled to prevent the use, by a third party, of a sign identical with, or similar to, its mark, in a comparative advertisement which satisfies all the conditions, under which comparative advertising is permitted.  [72] The judgement in the BEST case does not answer the question of whether the use of metatags for the offer of substitutable products constitutes a form of comparative advertising. Should the CJEU resolve this issue positively, then it is highly expected that use of the trademarked keywords in line with the requirements of the comparative advertising directive would remove the use of the trademarks in keyword advertising from the scope of the trademark directive and leave the whole issue to the law on comparative advertising.

Consequently, the proposed licensing model would provide compensation to the trademark holders for the loss incurred through the infringement of the advertising and investment functions of the trademark found by the CJEU under the double identity rule, while the use of the rules related to the comparative advertising may strengthen the position of the trademark holder as regards likelihood of confusion, taking unfair advantage, and causing detriment to the distinctive character or the repute of the trademark. This would be all the more welcomed as the grounds upon which comparative advertising is permitted are the same as the principles introduced in trademark law with one exception - there is no point of reference in the comparative advertising directive to afford protection to other trademark functions.  [73] Consequently, the licensing model could compensate for the infringement of the advertising function of the trademark with a stream of revenues, whilst the law on comparative advertising would govern the issues of likelihood of confusion and unfair advantage of the reputation of a trademark protecting the trademark holders from a risk of misuse of their trademarks used under the license agreements. However, in these circumstances, the concept of taking unfair advantage of the distinctive character or the repute of the senior trademark would still need to be reconsidered.

The abovementioned concept is not flawless, however banning the use of trademarked keywords would undermine the utility of the web browser in a significant manner, and as a result would deprive the consumers of a reliable source of information. Thus, this proposal shall be treated as an attempt to balance the interests of the trademark holders, their competitors, and the consumers. The introduction of a paid license for the use of trademarked keywords and the flexibility of the royalties model shall act as solution for conflicts arising between trademark holders and advertisers, while the need for protection of consumers’ information and the protection of fair competition could be solved by the law on comparative advertising.

* Maciej Zejda is a teaching and research assistant at the Faculty of Law and Administration of the University of Gdańsk.