«Technische Universität München As marketing expenditures in the Internet rush, effectiveness and efficiency become increasingly relevant. ...»
Interest-Based Internet Advertising and Privacy
Concerns: How to Increase the Acceptance of a
Rising Marketing Phenomenon
Nicole Groene, Florian von Wangenheim, Jan H. Schumann
Technische Universität München
As marketing expenditures in the Internet rush, effectiveness and efficiency become increasingly relevant. Particularly online-firms offering free content need to provide powerful marketing tools to advertisers to support
their own business model. Behavioral targeting enables websites to selectively display advertisements to consumers according to their surfing profiles, making advertisements more relevant, thereby increasing advertising revenues from websites. However, targeting can cause privacy concerns and negative consumer reactions. Furthermore, there is increasing regulatory pressure for websites to inform surfers about targeting practices and provide them with opt-out functions. Proactively addressing those challenges is highly important for advertising-supported websites. Building on privacy related fairness norms, the authors develop mechanisms to increase consumers’ acceptance of targeted advertising. In a laboratory and a large-scale field experiment, they find that under certain conditions, surfers are motivated by reciprocity. When reminded that targeted online advertisements help fund free content, consumers are not only more willing to provide information but also perceive targeted advertisements as less intrusive. To further address privacy concerns, websites should allow consumers to access and edit their information.
1. Introduction The Internet already constitutes the second-largest advertising medium, and while advertising revenues from traditional media have stagnated, online advertising revenues are expected to grow continuously (Interactive Advertising Bureau, 2010). Within online marketing, behaviorally targeted advertising has emerged as a major trend that is predicted to account for one-forth of total U.S. display advertising revenues by 2012 (Hallerman, 2008). Targeting enables websites to increase advertising relevance and, thus, effectiveness by selectively displaying advertisements to surfers with specific interests (Iyer; Soberman; Villas-Boas, 2005). This is typically achieved by placing cookies on surfers’ web browsers that often track surfing behaviors across websites within an advertising network (McDonald; Cranor, 2010). Therefore, targeted advertising enables websites to charge higher prices for their advertising space as it reduces waste for the advertiser.
Research on behavioral targeting is limited and controversial. On the one hand, studies sponsored by targeting firms report substantial increases in click rates of up to 1000% through behavioral targeting (e.g., Yan et al., 2009). On the other hand 66% of American adults reject behavioral targeting (Turow et al., 2010). Thus, while several academic studies report that consumers are concerned about their privacy with regard to behavioral targeting (e.g., Alreck; Settle, 2007; McDonald; Cranor, 2010), there is a surprising lack of research on how marketers can address consumers’ privacy concerns and increase the acceptance of behavioral targeting. Such research is particularly relevant for advertising-supported websites, such as newspapers, communities, and directories in light of the recent “Do Not Track” proposal by the Federal Trade Commission (FTC). The proposal suggests the installation of a nationwide optout tool through which consumers can restrict the collection of information about their web browsing behavior (Federal Trade Commission, 2010). Whereas such a tool provides consumers with better choice regarding their privacy online, it will most likely lead to a drop in advertising revenues if websites do not find mechanisms to increase the acceptance of targeting and thus reduce the number of individuals seeking an opt-out.
Finding ways to increase the acceptance of targeting, however, is also important to avoid other harmful consequences resulting from consumer privacy concern, such as website avoidance (Wirtz; Lwin, 2009) or negative word-of-mouth (Son; Kim, 2008).
Furthermore, in a large-scale study on targeting and obtrusiveness of display advertisements, Goldfarb and Tucker (2011a) suggest that privacy concerns even negatively affect advertising effectiveness. In recent commentaries to their article, several researchers stress the importance of research on the underlying cognitive mechanisms of privacy concerns and advertising effectiveness (Goldfarb; Tucker, 2011b;
Lodish; Reed II, 2011). In the context of direct mail and e-commerce, research on factors that increase the acceptance of personalization usually assumes that consumers perform a utilitarian cost–benefit trade-off with regard to their privacy (e.g.
Xie; Teo; Wan, 2006). Several studies find that consumers are willing to provide marketers with personal information if they receive financial rewards, such as coupons or discounts (e.g., Hann et al., 2007; Hui; Teo; Lee, 2007). However, the applicability of these findings in the context of targeted advertising is limited because providing consumers with monetary benefits is hardly implementable on non-e-commerce websites. Thus, in practice, websites currently trying to increase consumers’ acceptance of targeting usually emphasize that targeting makes their advertisements more relevant instead of offering monetary benefits (e.g., Google, 2009; Yahoo, 2011).
Regarding free online content, complementing the common utilitarian perspective on factors that increase consumers’ acceptance of personalization with a normative perspective that includes considerations of fairness and reciprocity seems suitable. Research in the context of pay-what-you-want pricing mechanisms shows that consumers voluntarily pay something for a service received, even if they do not have to (Kim;
Natter; Spann, 2009). Surprisingly this finding has not yet been applied to the context of free online content and advertising acceptance. Against this background, the research objectives of this article are threefold: 1. To examine how privacy concerns related to targeting practices affect consumers’ perceptions of targeted advertisements, a proven mediator of advertising effectiveness. 2. To identify and test mechanisms that increase the acceptance of behavioral targeting based on fairness norms, and 3. To test whether the identified mechanisms improve consumers’ perceptions of targeted advertisements. Figure 1 provides an overview of the theoretical foundation of our research framework. We test these mechanisms with data from two studies: a laboratory experiment accompanied by a survey and a large-scale field experiment.
Fig. 1: Theoretical Foundation of Research Framework of Studies 2. Theoretical Background and Literature Review 2.1. Consumers’ Online Privacy Concerns and Fairness In a marketing context, information privacy concern refers to the extent to which a person is worried about the practices of an organization regarding the collection and subsequent use of his or her personal information (Smith; Milberg; Burke, 1996). Recent online privacy literature indicates that this concern pertains to two facets of the interaction with a marketer: First, consumers are concerned about potentially harmful consequences of information collection (Malhotra; Kim; Agarwal, 2004), such as monetary (e.g.; Youn, 2009) or psychological harm (e.g., Dinev; Hart, 2006). Second, consumers are concerned about the fairness of their interaction with a marketer involving their information (Ashworth; Free, 2006). With regard to procedural fairness (i.e. the fairness of procedures and how they are enacted; Thibaut; Walker, 1975), consumers’ concerns revolve around being aware of marketers’ information practices and having control over their information (Malhotra; Kim; Agarwal, 2004). Yet, so far, the role of distributive fairness (i.e. the perceived fairness of the allocation of outcomes of an exchange; Homans, 1961), in influencing privacy concern has received only limited attention (e.g., Ashworth; Free, 2006; Dinev; Hart, 2006).
Lee, 2007) and convenience (Hann et al., 2007). Yet, the existing findings are of limited applicability to our research since they are either context specific such as convenience in an online store or room to provide financial benefits is limited because free content websites often struggle to finance their operations.
3. Research Framework and Hypotheses
Similar to the theoretical framework employed by previous privacy researchers, we conceptualize behavioral targeting as a social exchange between a website and its surfers. In this exchange relationship, a website offers free content to a consumer. In return, the consumer watches advertisements and allows targeting so that the website can receive advertising revenues from a third party—the advertiser—that targets this consumer. According to social exchange theory (SET) (e.g., Homans, 1961; Thibaut; Kelley, 1959) individuals evaluate their relationships with regard to costs and rewards. Accordingly, consumers are only willing to accept targeting if their perceived benefits are greater than the corresponding privacy cost. Because consumers’ privacy concerns revolve around the fairness of their interaction with a marketer (e.g., Ashworth; Free, 2006), we assume that norms related to fairness constitute a suitable basis to derive mechanisms for website to sustain or even increase revenues through targeted advertising.
3.1. Impact of Targeting on Perceived Intrusiveness An informed consent to targeting requires that consumers are informed about the behavioral advertising practices employed by a website. However, since most consumers are currently not fully aware of targeting practices (McDonald; Cranor, 2010), transparency might result in a higher perceived intrusiveness of targeted advertisements. This is because surfers who know or suspect that an advertisement shown to them has been delivered through behavioral targeting might perceive this as a threat to their ability to avoid being profiled when surfing online. This perception should lead
to reactance, a proven mediator of intrusiveness (Edwards; Li; Lee, 2002). Furthermore, the cognitions related to a potential threat of their privacy resulting from behavioral targeting require additional mental processing and thus interrupt surfers’ cognitive processes. Thus:
H1: When informed about behavioral targeting practices employed by a website, consumers perceive its advertisements as more intrusive than when not informed.
3.2. Mechanisms Related to Distributive Justice In exchange relationships with business partners, individuals typically assess the fairness of the allocation of outcomes according to the norm of equity (Aggarwal, 2004). Consequently, in exchange relationships in which the norm of equity is particularly salient, people’s “motivation is to get something back in return, that is quid pro quo” (Aggarwal, 2004, 88).
3.2.1. Increasing benefits through advertising relevance.
The most obvious way to increase distributive justice in the context of behavioral advertising is to increase consumers’ perceived benefits from targeting. Regarding benefits, the advertising industry often claims that targeting makes advertisements more interesting and more useful to consumers (Alreck; Settle, 2007). From an academic perspective, there is some support for this claim. In a direct mail context, Milne and Gordon (1993) find that respondents prefer less mail and more targeted mail.
This is in line with Alreck and Settle (2007), who emphasize the advantage of targeting in reducing irrelevant advertisements. Also Edwards, Li, and Lee (2002) show
that advertisements deemed informative are perceived less intrusive. Thus:
H2: Informing customers that targeting makes advertisements more interesting to them a) increases consumers’ acceptance of targeting and b) reduces the perceived intrusiveness of targeted advertisements, compared with not emphasizing relevance.
3.2.2. Increasing fairness through reciprocity.
Another way to positively influence consumers’ perceptions of distributive justice is to alter the perceived input of the website. As SET assumes that individuals perform actions that are rewarding to them (Emerson, 1976), this should be done in a way that consumers are motivated to participate in the exchange involving their information. Here the norm of reciprocity (Gouldner, 1960) might have motivational power, as it entails a person’s innate desire to repay a favor, which is typically driven by a feeling of indebtedness towards the donor (Greenberg, 1980). Free content might constitute a benefit that consumers might wish to reciprocate, for example, by consenting to targeting. Yet, after many years of consuming free content, consumers have developed a “free mentality” (Dou, 2004). Therefore, the norm of reciprocity might not be focal in the context of targeted advertising. Employing cues that make
the norm of reciprocity more salient with regard to a website’s provision of free content to the consumer might evoke a desire to reward the website. Thus:
H3: An appeal to reciprocity a) increases customers’ acceptance of targeted advertising on a website and b) reduces the perceived intrusiveness of targeted advertisements on a website, compared with not appealing to reciprocity.
3.2.3. Interaction of equity and reciprocity.
If different norms are applicable in a given situation, the norm that is activated guides behavior (Cialdini; Kallgren; Reno, 1991). The two mechanisms suggested might create two forces: While the relevance mechanism aims to satisfy the desire to be