Consumer Decision-Making Processes in Mobile Viral Marketing Campaigns ⁎ Christian Pescher & Philipp Reichhart & Martin Spann

Available online at www.sciencedirect.com ScienceDirect

Journal of Interactive Marketing 28 (2014) 43–54 www.elsevier.com/locate/intmar

Consumer Decision-making Processes in Mobile Viral Marketing Campaigns ⁎ Christian Pescher & Philipp Reichhart & Martin Spann

Institute of Electronic Commerce and Digital Markets, Ludwig-Maximilians-Universität München, Geschwister-Scholl-Platz 1, D-80539 München, Germany

Available online 22 November 2013

Abstract

The high penetration of cell phones in today's global environment offers a wide range of promising mobile marketing activities, including mobile viral marketing campaigns. However, the success of these campaigns, which remains unexplored, depends on the consumers' willingness to actively forward the advertisements that they receive to acquaintances, e.g., to make mobile referrals. Therefore, it is important to identify and understand the factors that influence consumer referral behavior via mobile devices. The authors analyze a three-stage model of consumer referral behavior via mobile devices in a field study of a firm-created mobile viral marketing campaign. The findings suggest that consumers who place high importance on the purposive value and entertainment value of a message are likely to enter the interest and referral stages. Accounting for consumers' egocentric social networks, we find that tie strength has a negative influence on the reading and decision to refer stages and that degree centrality has no influence on the decision-making process. © 2013 Direct Marketing Educational Foundation, Inc. Published by Elsevier Inc. All rights reserved.

Keywords: Mobile commerce; Referral behavior; Sociometric indicators; Mobile viral marketing

Introduction relies on peer-to-peer communications for its successful diffusion among potential customers. Therefore, viral market- The effectiveness of traditional marketing tools appears to be ing campaigns build on the idea that consumers attribute higher diminishing as consumers often perceive advertising to be credibility to information received from other consumers via irrelevant or simply overwhelming in quantity (Porter and referrals than to information received via traditional advertising Golan 2006). Therefore, viral marketing campaigns may provide (Godes and Mayzlin 2005). Thus, the success of viral marketing an efficient alternative for transmitting advertising messages to campaigns requires that consumers value the message that they consumers, a claim supported by the increasing number of receive and actively forward it to other consumers within their successful viral marketing campaigns in recent years. One social networks. famous example of a viral marketing campaign is Hotmail, Mobile devices such as cell phones enhance consumers' which acquired more than 12 million customers in less than ability to quickly, easily and electronically exchange informa- 18 months via a small message attached at the end of each tion about products and to receive mobile advertisements outgoing mail from a Hotmail account informing consumers immediately at any time and in any location (e.g., using mobile about the free Hotmail service (Krishnamurthy 2001). In addition text message ads) (Drossos et al. 2007). As cell phones have the to Hotmail, several other companies, such as the National potential to reach most consumers due to their high penetration Broadcasting Company (NBC) and Proctor & Gamble, have rate (cf., EITO 2010), they appear to be well suited for viral successfully launched viral marketing campaigns (Godes and marketing campaigns. As a result, an increasing number of Mayzlin 2009). companies are using mobile devices for marketing activities. In general, a viral marketing campaign is initiated by a firm that Research on mobile marketing has thus far devoted limited actively sends a stimulus to selected or unselected consumers. attention to viral marketing campaigns, particularly with respect to However, after this initial seeding, the viral marketing campaign the decision-making process of consumer referral behavior for mobile viral marketing campaigns, e.g., via mobile text messages. Thus, the factors that influence this process remain largely ⁎ Corresponding author. E-mail addresses: [email protected] (C. Pescher), unexplored. The literature on consumer decision-making suggests [email protected] (P. Reichhart), [email protected] (M. Spann). that consumers undergo a multi-stage process after receiving a

1094-9968/$ -see front matter © 2013 Direct Marketing Educational Foundation, Inc. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.intmar.2013.08.001 44 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 stimulus (e.g., a mobile text message) and before taking action framework to consider social relationships (e.g., tie strength, (e.g., forwarding the text message to friends) (Bettman 1979; De degree centrality) when analyzing mobile viral marketing Bruyn and Lilien 2008). At different stages of the process, various campaigns. Thus, to understand referral behavior, we integrate factors that influence consumer decision-making can be measured psychographic (e.g., usage intensity) and sociometric (e.g., tie using psychographic, sociometric, and demographic variables as strength) indicators of consumer characteristics. We are then able well as by consumer usage characteristics. Whereas previous to determine the factors that influence a consumer's decision to studies have mainly focused on selected dimensions, our study refer a mobile stimulus and are able to identify the factors that considers variables from all categories. lead to reading the advertising message and to the decision to De Bruyn and Lilien (2008) analyzed viral marketing in an learn more about the product. online environment and discussed relational indicators of business students who had received unsolicited e-mails from Related Literature friends. This study provided an important contribution and amplified our understanding about how viral campaigns work. Viral Marketing and Factors that Influence Consumer Referral The present paper differs from the work of De Bruyn and Lilien Behavior (2008) and goes beyond their findings in four important ways: actor, medium, setting, and consumer characteristics. The first Viral marketing campaigns focusontheinformationspreadof difference is the actor involved. In viral campaigns, the customers, that is, their referral behavior regarding information or initiator, usually a company, sends the message to the seeding an advertisement. Companies are interested in cost-effective points (first level). Next, the seeding points forward the marketing campaigns that perform well. Viral marketing cam- message to their contacts (second level), and so on. Whereas paigns aim to meet these two goals and can, accordingly, have a De Bruyn and Lilien (2008) focused on the second-level actors, positive influence on company performance (Godes and Mayzlin the present study focuses on the first-level actors, e.g., the direct 2009). Companies can spread a marketing message with the contacts of the company. We believe that for the success of a objective of encouraging customers to forward the message to their campaign, additional insights into the behavior of first-level contacts (e.g., friends or acquaintances) (Van der Lans et al. 2010). actors are very important because if they do not forward the In this way, the company then benefits from referrals among message, it will never reach the second-level actors. The second consumers (Porter and Golan 2006). Referrals that result from a difference is the medium used in the campaign. Although we viral marketing campaign attract new customers who are likely to cannot explicitly rule out that participants of De Bruyn and be more loyal and, therefore, more profitable than customers Lilien's (2008) campaign used mobile devices, they conducted acquired through regular marketing investments (Trusov, Bucklin, their campaign at a time when the use of the Internet via mobile and Pauwels 2009). devices was still very uncommon. Therefore, it is reasonable to Two streams of research can be identified. The first is the assume that at least the majority of their participants used a influence of viral marketing on consumers, and the second is desktop or a laptop computer when they participated in De research that has analyzed the factors that lead to participating Bruyn and Lilien's (2008) campaign. In contrast, the present in viral marketing campaigns. First, previous research identified study explicitly uses only text messages to mobile devices. In that viral marketing influences consumer preferences and pur- addition, mobile phones are a very personal media which is chase decisions (East, Hammond, and Lomax 2008). Further, an used in a more active way compared to desktop or laptop influence on the pre-purchase attitudes was identified by Herr, computers (Bacile, Ye, and Swilley 2014). The third difference is Kardes, and Kim (1991). In addition, viral marketing also the setting in which the viral campaign takes place. Whereas the influences the post-usage perceptions of products (Bone 1995). participants in the study by De Bruyn and Lilien (2008) were Second, previous research has identified satisfaction, business students from a northeastern US university, we conduct customer commitment and product-related aspects as the most a mobile marketing campaign in a field setting using randomly important reasons for participating in viral marketing campaigns selected customers. The fourth and most important difference is (cf., Bowman and Narayandas 2001; De Matos and Rossi 2008; that De Bruyn and Lilien (2008) focused exclusively on relational Maxham and Netemeyer 2002; Moldovan, Goldenberg, and characteristics. In addition to relational characteristics, this Chattopadhyay 2011). With respect to psychological motives, paper also considers variables that describe demographic factors, self-enhancement was identified as a motive for consumers to psychographic factors, and usage characteristics. As these generate referrals (De Angelis et al. 2012; Wojnicki and Godes variables yield significant results, the study and its findings go 2008). The importance of self-enhancement in addition to social beyond the findings of De Bruyn and Lilien (2008). benefits, economic incentives and concern for others was identified The main goal and contribution of this work is, first, to as a motive behind making online referrals (Hennig-Thurau et al. analyze consumers' decision-making processes regarding their 2004). Referrals can be differentiated into positive and negative forwarding behavior in response to mobile advertising via their referrals. Anxiety reduction, advice seeking and vengeance are cell phone (i.e., text messages) in a mobile environment using a factors that contribute to negative referrals (Sundaram, Mitra, and real-world field study. To analyze consumers' decision-making Webster 1998). processes, we use a three-stage sequential response model of Within the referral process, the relationships and social the consumer decision-making process. Additionally, we inte- network position of the consumer are also influential. For grate consumers' egocentric social networks into a theoretical example, Bampo et al. (2008) found that network structure is C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 45 important in viral marketing campaigns. Furthermore, it has group-person connectivity, commitment to the brand, and been determined that consumers are more likely to activate relationship with the mobile device. In contrast to our study, strong ties than weak ties when actively searching for Okazaki (2008) did not analyze whether referrals were made, information (Brown and Reingen 1987) because strong ties nor did he analyze the referrals that were directly made via a tend to be high-quality relationships (Bian 1997; Portes 1998). mobile device by forwarding the mobile text message. Instead, he In addition, targeting consumers who have a high degree analyzed the general viral effect in the form of telling or centrality (e.g., quantity of relationships) leads to a higher recommending the mobile advertising campaign. Further, our number of visible actions, such as page visits, than do random field study analyzes the entire consumer decision-making process seeding strategies (Hinz et al. 2011). Kleijnen et al. (2009) for a mobile viral marketing campaign via text messages across analyzed the intention to use mobile services using sociometric the three stages: from stage one, reading, to stage two, interest, to variables and evaluated how consumers' network positions stage three, decision to refer. influence their intentions to use mobile services. However, the To summarize, in contrast to the existing studies in the field previous study contributes to the literature by analyzing a different of mobile viral marketing, we analyze consumers' egocentric research question than is examined in our paper. Specifically, networks via measures such as tie strength and degree Kleijnen et al. (2009) focusedontheintention to use services, centrality. These sociometric factors are analyzed jointly while our study focuses on consumers' decision-making processes with psychographic constructs across the three stages in the until they make a referral. In summary, previous research focused decision-making process. Thus, our study uses a real-world on the consumers' psychographic constructs or relationships and mobile viral marketing campaign and enables us to test the social networks to explore why consumers participate in viral relative importance of social embeddedness and consumer marketing campaigns and why they make referrals, two constructs characteristics with respect to consumers' decision to forward that are rarely analyzed together. Iyengar, Van den Bulte, and mobile messages. Valente (2011) used both constructs jointly and found that correlations between the two are low. However, this study did Decision-making Process and Specifics of the Mobile Environment not take place in an online or mobile context but rather in the context of referrals for new prescription drugs between specialists. Consumer decision-making is a multiple-stage process In contrast, our study analyzes both aspects together within a (Bettman 1979; De Bruyn and Lilien 2008; Lavidge and mobile viral marketing campaign. Steiner 1961). In a viral marketing campaign, the final goal is to In addition to offline- or online-based viral marketing activities, generate a high number of referrals. Therefore, our model of an increasing number of companies are conducting marketing consumer forwarding behavior is designed for the specific campaigns using mobile phones, and promising approaches situation of mobile viral marketing campaigns. include mobile viral marketing campaigns. Research on mobile The process and first stage begin with the consumer reading viral marketing is relatively unexplored because most research in a mobile advertising text message on his or her mobile phone. the field of mobile marketing analyzes marketing activities such as If this text message sparks the consumer's interest and the mobile couponing (Dickinger and Kleijnen 2008; Reichhart, consumer wants to learn more about the offered product, he/she Pescher, and Spann 2013), the acceptance of advertising text enters the interest stage, which is the second stage of the model. messages (Tsang, Ho, and Liang 2004) or the attitudes toward If the consumer finds the product interesting after learning (Tsang, Ho, and Liang 2004) and the acceptance of mobile about it, he or she makes a referral, which is the third stage of marketing (Sultan, Rohm, and Gao 2009). In the context of mobile our model (decision to refer). viral marketing research, Hinz et al. (2011) studied mobile viral In this study, we analyze the stages of the consumer marketing for a mobile phone service provider and determined that decision-making processes within a mobile environment, i.e., the most effective seeding strategy for customer acquisition is to within a mobile viral marketing campaign. There are several focus on well-connected individuals. In contrast to our study, their differences between mobile viral marketing and online or referrals were conducted via the Internet (i.e., the companies' offline viral marketing. A mobile text message is more online referral system) rather than via a mobile device (i.e., intrusive than an e-mail because it appears immediately on forwarding the text message immediately). Nevertheless, generat- the full screen. Consumers usually carry their mobile phone ing referrals using a mobile device can affect referral behavior. with them and a mobile message may also reach them in a Palka, Pousttchi, and Wiedemann (2009) postulated a grounded private moment. Contrary, consumers may need to purposely theory of mobile viral marketing campaigns and found that trust look into their e-mail accounts to receive e-mails. Therefore a and perceived risk are important factors in the viral marketing mobile message can be more personal compared to an e-mail. process. In comparison to our study, they used qualitative methods In comparison to offline face-to-face referrals, mobile referrals and did not conduct a real-world field study. Okazaki (2008) do not possess this personal aspect and can be transmitted identified, for Japanese adolescents, consumer characteristics such digitally within a few minutes to several friends in different as purposive value and entertainment value are the main factors places simultaneously. This is not possible in the offline world. in mobile viral marketing campaigns and that these factors Additionally, a mobile referral can reach the recipient faster significantly influence the adolescents' attitudes toward viral than an e-mail or an offline referral. Thus, the mobile device marketing campaigns. Furthermore, both purposive value may influence the referral behavior due to its faster digital and entertainment value are influenced by the antecedents' transmission of information. 46 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54

Development of Hypotheses influence on a consumer's attitude toward a mobile viral marketing campaign and that this attitude significantly influences the While the factors that influence the stages of the decision- intention to participate in mobile marketing campaigns (Okazaki making process can be divided into two groups, we analyze 2008). For some consumers, forwarding a (mobile) advertisement them jointly in this study. The first group consists of the in a viral marketing campaign can have a personal and a social psychographic indicators of consumer characteristics, thus meaning (e.g., doing something good for friends by forwarding the focusing on each consumer's motivation to participate in the ad). Thus, we hypothesize that consumers who place high campaign and his or her usage behavior. The second group of importance on the purposive value of exchanging messages will factors includes sociometric indicators of consumer character- display a greater likelihood to enter the reading and interest stages. istics, thus providing information about the type of relationship We also hypothesize that consumers who place high importance on that the consumer has with his or her contacts and his or her the purposive value of a message are more likely to make the resulting social network. decision to forward the message. H2. Consumers who place high importance on the purposive Psychographic Indicators of Consumer Characteristics value of a message are more likely to a) enter the reading stage, b) enter the interest stage and c) enter the decision to refer stage. As mentioned in the related literature section, according to Okazaki (2008), in viral marketing campaigns, purposive value The intensity of usage (e.g., a high quantity of written text and entertainment value are the primary value dimensions for messages) positively influences the probability of trial and consumers. This insight is based on the finding that consumers adoption (Steenkamp and Gielens 2003). Thus, consumers with gain two types of benefits from participating in sales promotions: high usage intensities are more likely to actively participate in a hedonic and utilitarian benefits (Chandon, Wansink, and Laurent mobile viral marketing campaign. As mobile viral marketing 2000). Hedonic benefits are primarily intrinsic and can be campaigns are a fairly new form of advertising, consumers with associated with entertainment value. Consumers participate high usage intensities are more likely to participate in mobile voluntarily and derive value from the fun of interacting with viral marketing campaigns and are more likely to forward peers by forwarding a referral (e.g., an ad might be of interest to messages than consumers with low usage intensities. Therefore, peer recipients) (Dholakia, Bagozzi, and Pearo 2004). A previous we propose that usage intensity has an effect on the decision to study found that the entertainment factor influences intended use forward a mobile advertising text message. The likelihood of in mobile campaigns (Palka, Pousttchi, and Wiedemann 2009). deciding to forward the mobile advertisement increases with Okazaki (2008) found that in a mobile viral marketing campaign, the usage intensity of mobile text messages. This proposition is the entertainment value directly influences the recipient's attitude consistent with Neslin, Henderson, and Quelch (1985), who toward the campaign, which, in turn, influences the recipient's found that the promotional acceleration effect is stronger for intention to participate in a mobile viral campaign. Phelps et al. heavy users than it is for other consumers. Godes and Mayzlin (2004) showed that the entertainment value is a factor that (2009) analyzed the effectiveness of referral activities and increases consumers' forwarding behavior in viral marketing argued that the sales impact from less loyal customers is campaigns conducted via e-mail. Thus, we may presume that greater, but they also highlighted that this greater sales impact consumers who place high importance on the entertainment value does not mean that the overall referrals by less loyal customers of exchanging messages are more likely to enter the reading and have a greater impact than those by highly loyal customers. interest stages than consumers who do not value entertainment to They concluded that companies who want to implement an the same degree. Additionally, the entertainment value can exogenous referral program to drive sales should focus on both also influence the decision to refer (i.e., forwarding) behavior less loyal and highly loyal customers because focusing only on because a text message that addresses consumers who place highly loyal or less loyal customers is not necessarily the high importance on entertainment value causes the recipient to cornerstone of a successful viral marketing campaign. In the think about forwarding the text message and motivates them online context, a previous study found that experience with the to forward the mobile advertisement to friends (i.e., decision Internet influences channel usage behavior (Frambach, Roest, to refer stage). and Krishnan 2007). Thus, as consumers with high usage intensity are used to communicating with mobile phones, they H1. Consumers who place high importance on the entertainment know how to write, read and forward mobile text messages. value of a message are more likely to a) enter the reading stage, Accordingly, it is likely that the threshold to forward a text b) enter the interest stage and c) enter the decision to refer stage. message is lower for consumers with high usage intensity than As utilitarian benefits are instrumental and functional, they it is for other consumers and that such consumers are thus more can be associated with purposive value (Okazaki 2008). inclined to refer. Further, the minimal effort required to directly Dholakia, Bagozzi, and Pearo (2004) analyzed the influence forward a mobile text message via a cell phone increases the of purposive value in network-based virtual communities and decision to refer. Thus, we hypothesize that heavy mobile users found that purposive value is a predictor of social identity and will be more likely to refer than will light users. a key motive for an individual to participate in virtual communities. With respect to the mobile context, previous H3. The usage intensity of the referral medium has a positive research found that purposive value has a direct, significant influence on the likelihood of making the decision to refer. C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 47

Sociometric Indicators of Consumer Characteristics reward increases the referral intensity and has a particular impact on weak ties (Ryu and Feick 2007). Brown and Reingen 1987 Sociometric indicators describe the interaction structure of found that while strongly tied individuals exchange more an individual consumer with his or her surroundings. When information and communicate more frequently, weak ties play consumers receive an interesting mobile advertising message, it an important bridging role. Additionally, Granovetter (1973) is likely that they want to find out more about it. Once the stated that one is significantly more likely to be a bridge in the consumer has visited the product homepage, he or she then case of weak ties than in strong ties. In job search, when using considers not only whether the message is worth forwarding but personal networks, it was found that weak ties have a higher rate also to whom it should be forwarded. of effectiveness when addressing specialists for jobs compared to Sociometric indicators provide information about the social strong ties (Bian 1997) and that the income of people using weak network of each individual consumer. This individual network ties was greater than those who used strong ties (Lin, Ensel, and influences the likelihood of knowing someone who may be Vaughn 1981). At the information level, consumers who are interested in the offered product. Thus, social networks have a connected via strong ties tend to share the same information significant impact on the decision-making process in a viral that is rarely new to them, while consumers obtain important marketing campaign. The decision to forward the mobile information from weak ties who tend to possess information that advertising message depends on two factors: the quality and the is “new” to them (Granovetter 1973). Consistent with this quantity of relations, i.e., the tie strength and the degree centrality. finding, Levin and Cross (2004) found that novel insights and Tie strength is an important factor in viral marketing and new information are more likely to pass along weak ties than increases with the amount of time spent with the potential between strong ties. As in our study, viral marketing information recipient and with the degree of emotional intensity between can be perceived as a novel insight or new information. Given the sender and the potential recipient (Marsden and Campbell that consumers are more likely to send a message to someone if 1984). Consumers perceive strong ties to be more influential the content is new to the receiver, it is more likely that consumers than weak ties (Brown and Reingen 1987) because the strong will forward the text message to a weak tie than to a strong tie. ties seem more trustworthy (Rogers 1995). Therefore, because Thus, we presume that consumers prefer to forward mobile consumers are more motivated to provide high-value informa- advertising text messages, such as the one used in our study (for a tion to strong ties (Frenzen and Nakamoto 1993), tie strength is new music CD), to other customers with whom they are connected an indicator of the quality of the relationship. throughweakties. Reagans and McEvily (2003) studied how social network Similarly, Frenzen and Nakamoto (1993) postulate that the factors influence knowledge transfer at an R&D firm. To motivationtoshareinformationorreferaproductisdrivenbythe measure the tie strength, they used two items that are analogous value of the information and the cost of sharing. They identified to those that we used (Burt 1984). Their results indicated that (though only for weak ties) an influence of word-of-mouth that is tie strength positively influences the ease of knowledge spread by value and opportunity costs. In our case, when customers transfer. Thus, network ties increase a person's capability to forward a mobile advertising message, the opportunity cost is low send complex ideas to heterogeneous persons. Overall, they because forwarding can be done easily and without any effort (e.g., highlighted the importance of tie strength with respect to the compared to meeting the friend personally in the city). In the case knowledge transfer process, and they postulate that tie strength of strong ties, the preferences of the recipient (e.g., for products) are holds a privileged position. Other studies found that weak ties well known to the sender, whereas these preferences are unclear for make non-redundant information available (Levin and Cross weak ties. Thus, people who want to do something beneficial for 2004). In an online setting, participants were more likely to their contacts know the likelihood that it will benefit the recipient in share information with strong ties than with weak ties (Norman the case of strong ties, but they do not know the benefit it may bring and Russell 2006). With respect to viral marketing conducted to weak ties. In our study, this benefit involved sharing information via e-mail, previous research has found that tie strength has a toacquireafreenewmusicCD.Becausethecostsofsharingare significant influence on whether the recipient examines an low using cell phones, people are more inclined to forward such e-mail message sent from a friend (i.e., opens and reads the information. With respect to strong ties, people know whether the message) (De Bruyn and Lilien 2008). Tie strength was also information is relevant. Furthermore, relationships to strong ties are determined to be less relevant in an online setting compared to an more important than relationships to weak ties. Importantly, offline setting (Brown, Broderick, and Lee 2007). In a non-mobile people do not want to displease strong ties by sending irrelevant or non-online context, stronger ties are more likely to activate the information or cause information overload with unsuitable referral flow (Reingen and Kernan 1986). Furthermore, tie information. In the case of receiving annoying information, the strength is positively related to the amount of time spent receiving recipient could ask the sender not to forward text messages positive referrals (van Hoye and Lievens 1994). anymore. This sanction is more painful when received from strong As previously mentioned, research on word-of-mouth behav- ties (e.g., good friends) than from weak ties (e.g., acquaintances) ior has shown that people engage in word-of-mouth for reasons because the weak ties are less important to the sender. This is such as altruism (Sundaram, Mitra, and Webster 1998). However, similar to the finding that people who are dissatisfied (e.g., with a Sundaram, Mitra, and Webster (1998) did not control for the product or service) are more likely to advise against the purchasing quality of a relationship between sender and recipient. Research of the product to strong ties rather than to weak ties (Wirtz and concerning referral reward programs has identified that offering a Chew 2002), which may also be due to the sanction issue. In 48 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 mobile viral marketing campaigns, because the sharing of phones (opt-in program). We sent a text message to 26,137 information is easy and not costly, factors such as knowledge randomly chosen customers that included a link to a website about the preferences of the recipient or the fear of annoying and the notice that they could download a recently released strong ties become more important when deciding whether and to music CD for free. The only purpose of this website was to give whom to refer the message. the campaign's participants the option to download the music Thus, we hypothesize the following: CD for free. In the text message, they were also asked to forward the message to their contacts. The exact text of the H4. Tie strength has a negative influence on the likelihood of message stated, “Amazing! You & your friends will receive a making the decision to refer. new CD as an MP3 for free! No subscription! Available online: Diffusion occurs via replication or transfer, e.g., of used goods URL.com -N Forward this text message to your friends now!” (Borgatti 2005). Within the group of replication processes, One week later, we sent another text message to all of the replications can occur one at a time (serial duplication), participants who received the initial mobile advertising text e.g., gossip or viral infections, or simultaneously (parallel message with a link to an online survey. This second text duplication), e.g., e-mail broadcasts or text messages on mobile message contained the request to participate in an online phones. Degree centrality is a measure that is suitable for survey. We provided one 100 Euro and two 50 Euro prizes as analyzing processes in which a message is duplicated simulta- incentives to participate in the survey. The winners were drawn in a neously because it can be interpreted as a measure of immediate lottery. In the questionnaire, the participants provided information influence — the ability to “infect” others directly (Borgatti 2005, about their behavior during different stages of the referral process p62). Hubs are identified using degree centrality because they are and about their psychological constructs and egocentric networks actors who possess a high number of direct contacts (Goldenberg (Burt 1984). Egocentric networks are networks that analyze the et al. 2009) and because they know a high number of people to focal actor and the actor's direct friends as well as the relations that whom they can forward a message and can thus influence more exist between them (Burt 1984). people (Hinz et al. 2011). Hubs also adopt earlier in the diffusion process. In detail, innovative hubs increase the speed of the Measures adoption process, while follower hubs influence the size of the total market (Goldenberg et al. 2009). Further, hubs tend to be The psychographic constructs “purposive value” and “enter- opinion leaders (Kratzer and Lettl 2009; Rogers and Cartano tainment value” were adapted from Dholakia, Bagozzi, and Pearo 1962) because they have a high status and serve as reference (2004) and Okazaki (2008). Concerning the operationalization of points in the information diffusion process. Small groups of both psychographic constructs, i.e., purposive value and entertain- opinion leaders often initiate the diffusion process of innovations ment value, we addressed the consumers' characteristics to identify (Van den Bulte and Joshi 2007). Czepiel (1974) analyzed the consumers who place high importance on the purposive value whether centrality in opinion networks influenced the adoption of messages in general. We analyzed the consumer's characteristics and found no significance for this, a finding that is contrary to concerning both constructs, i.e., the importance of the purposive other literature in the field (e.g., Goldenberg et al. 2009). value and the entertainment value for respondents. Items were Furthermore, targeting central customers leads to a significant measured on a seven-point Likert scale, using scale points from increase in the spread of marketing messages (Kiss and Bichler “do not agree at all” (1) to “totally agree” (7). We operationalized 2008). In a viral marketing campaign for a mobile service where the “usage intensity” by surveying the number of text messages referrals are conducted via the Internet (i.e., online referral each participant wrote per day (see Appendix A for details). system) and not via forwarding a mobile text message, the results To measure consumers' social networks, we surveyed their showed that high centrality increases the likelihood of participa- egocentric networks (Burt 1984; Fischer 1982; McCallister and tion (Hinz et al. 2011). Therefore, we hypothesize that degree Fischer 1978; Straits 2000). Egocentric networks are defined as centrality has a positive influence on the decision to refer. the direct relationships between an individual consumer (or ego) and other consumers (or alters) and the relationships that H5. Degree centrality has a positive influence on the likelihood exist between the alters. These are small networks of one focal of making the decision to refer. actor called the “ego”, the participant in the survey, and his or her contacts, called “alters”. The difference between a regular Empirical Study network and an egocentric network is that in the latter, all of the necessary information is obtained from one actor, which makes Goal and Research Design it a feasible method for obtaining samples that are representa- tive of a large-scale population. To access the respondents' core The goal of this empirical study is to test the hypotheses networks, we used the term “generator,” which was taken from derived above using a three-stage model that represents the Burt (1984), and adapted it to the specific characteristics of this stages of a consumer's decision-making process in a mobile study (see Appendix A for details). The participants first generated referral context. their list of alters by identifying their most frequent contacts and In our field study, we conducted a mobile marketing were then asked a series of questions, which helped us to gain campaign. The randomly chosen participants had previously additional insights in their social network, including the strength agreed to receive mobile advertising text messages on their cell of the relationship with each contact. Because each egocentric C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 49 network is calculated based on information from a single Results respondent, such networks are usually treated as undirected. Based on this information, we calculated the degree centrality, Descriptive Statistics and Bias Tests which is the number of ties for a node, and average tie strength for each network. Marsden (2002) showed that the egocentric In all, 943 subjects responded to the survey. Of these, 634 read centrality measures are generally good proxies for sociometric the initial message (reading), 440 visited the homepage (interest) centrality measures. and 144 consumers forwarded the message (decision to refer). Table 1 shows the correlations and descriptive statistics among the variables in this study. Model First, we compare the demographic characteristics of the survey respondents with the demographic characteristics of the The model consists of a funnel of three successive decisions: entire customer sample. Of the 943 survey respondents, 28% (i) reading the message, (ii) visiting the homepage (interest) and are female, which is in line with the entire sample. In addition (iii) forwarding the message (decision to refer). In each stage, we observed the age distribution of the survey respondents as the number of observations diminishes because only consumers well as the entire customer sample and found that the groups who took action in the last stage can take another action in the are essentially consistent (see Table 2). next stage, i.e., only consumers who read the message can access Second, we compared consumers who read the text message the homepage. Fig. 1 shows that the model is hierarchical, nested, (Nread = 634) with those who did not read the initial text message and sequential, which leads to desirable statistical properties. (Nnoread = 309) with respect to their surveyed demographics and Therefore, we use Maddala's (1983, p 49) sequential response cell phone usage behavior. We found no significant difference model, which is also known as a model for nested dichotomies between the groups for demographics (female: Mread =27.0%, (Fox 1997). We follow the argumentation of De Bruyn and Lilien Mnonread =30.1%, p N .31; age: Mread =29.2yrs,Mnonread = (2008), who adapted a model of Maddala (1983, pp 49) to the 30.6 yrs, p N .06), monthly cell phone usage (Mread =29.28€, context of consumer decision-making (see Appendix B for our Mnonread =29.62€, p N .95) or age and gender. model). We fit the model by simultaneously maximizing the likelihood functions of the three dichotomous models in Stata 12. Three-stage Decision-making Model Each likelihood function incorporates the estimated probabilities of the preceding stages. We include all explanatory variables in the estimation of De Bruyn and Lilien (2008) argued that consumers do not every stage to avoid an omitted variable bias. Table 3 shows the drop out at random in this process because it is a process of results of our sample selection model. self-selection, which may raise statistical concerns. However, the parameter estimates for the structure of the model used in this paper have been shown to be unaffected by changes in the Entertainment Value and Purposive Value (H1 and H2) marginal distributions of the variables (Bishop, Fienberg, and We find significant influence concerning consumers who place Holland 1975; Mare 1980). high importance on the entertainment value of exchanging mobile

Consumer receives message n=943 UNAWARE

Consumer Consumer does not read reads message message n=309 n=634 STAGE READING

Consumer Consumer does not visit visits homepage homepage n=194 n=440 STAGE INTEREST

Consumer Consumer does not for- forwards ward message message n=296 n=144 STAGE TO REFER TO DECISION DECISION

Fig. 1. Consumers' decision-making process in the viral campaign. 50 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54

Table 1 Table 3 Descriptive statistics. Results of the three-stage model. Mean StD 1 2 3 4 5 Stage: Reading Stage: Interest Stage: Decision to refer 1. Entertainment value 4.19 1.53 2. Purposive value 3.81 2.01 .46 ⁎⁎ Coefficient SE Coefficient SE Coefficient SE 3. Usage intensity 7.53 19.73 .14 ⁎⁎ .08 ⁎⁎ Entertainment value .116 ⁎⁎ .054 .181 ⁎⁎ .070 .204 ⁎⁎ .101 4. Tie strength 3.10 .80 .11 ⁎⁎ .05 .01 Purposive value .062 .045 .167 ⁎⁎ .063 .600 ⁎⁎ .095 5. Degree centrality 3.04 1.46 .08 ⁎⁎ .03 .05 .02 Usage Intensity −.003 .003 .008 .008 .010 ⁎ .006 6. Age 29.69 11.08 −.10 ⁎⁎ .05 −.14 ⁎⁎ .02 −.17 ⁎⁎ Tie strength −.219 ⁎⁎ .094 −.156 .114 −.461 ⁎⁎ .153 Notes: means, standard deviations and correlations, N = 943. Degree centrality −.021 .050 −.010 .064 .114 .085 ⁎⁎ p b .05. Age −.012 ⁎ .006 −.014 .009 .036 ⁎⁎ .012 Gender a .098 .160 .580 ⁎⁎ .200 −.229 .283 Intercept 1.069 ⁎⁎ .461 −.100 .570 −4.264 ⁎⁎ .814 n 943 − text messages with regard to the reading stage (coefficient: Log likelihood 1174.81 Wald chi2 180.58 .116/H1a), the interest stage (coefficient: .181/H1b) and the Prob N chi2 b.01 decision to refer stage (coefficient: .204/H1c). Furthermore, a we find a significant increase in the likelihood of entering the Dummy coding (0 = female, 1 = male). ⁎ p b .10. interest stage (coefficient: .167/H2b) as well as the decision to ⁎⁎ p b .05. refer stage (coefficient: .600/H2c) for consumers who place high importance on purposive value. However, no support can be found for H2a. Overall, consumers who place higher importance on the receivers of unsolicited e-mails tend to pay more attention to entertainment value of exchanging text messages and who messages from close contacts (i.e., high-quality contacts) (De place higher importance on the purposive value demonstrate an Bruyn and Lilien 2008). However, their study focused on increased likelihood of entering the interest and decision to consumers who are actively searching for relevant information refer stages. and are thus receivers of information. In contrast, the present study addresses mobile advertising text messages that are sent Usage Intensity (H3) from a consumer to the consumer's contacts without being Our results show that the influence of usage intensity on the solicited. In other words, we focus on the sender of the message. decision to refer a mobile text message is only significant at the The difference between the sender and the receiver of a message 10% level (coefficient .010/H3). Thus, consumers who are used provides a solid explanation for the differences in the results to writing mobile text messages are more likely to forward between the two studies. Further, we find a negative influence of mobile text messages as well as the advertising text message. tie strength on the likelihood to read the message, which may be This result is consistent with previous findings that usage explained by the low tendency of consumers with strong tie intensity is positively related to user behavior (Gatignon and relationships to read firm-initiated text messages. Robertson 1991). To gain deeper insight into the negative impact of tie strength on the decision to forward a message, we conduct an additional analysis and find that the average tie strength Tie Strength (H4) between senders and receivers of forwarded messages is 2.99. Our results support H4: Tie strength significantly decreases This tie strength is significantly less than the average tie the likelihood that consumers decide to send and forward a strength of 3.09 for connections where no text messages were mobile text message (coefficient: −.461/H4). Thus, lower levels forwarded (p b .05). of tie strength increase the likelihood of the decision to send advertising text messages. A previous study found that the Degree Centrality (H5) In testing H5, we focus on the influence of degree centrality on the decision to refer and find that the number of contacts (i.e., degree centrality) has no influence on the decision to refer. Table 2 Thus, H5 is not supported. Demographics of survey respondents and the entire customer sample. In addition to testing the hypotheses, we examine the Percentage influence of gender on each stage of the decision-making Survey respondents Entire customer sample process. The results show that gender has a significant influence only on the interest stage (coefficient: .580). No significant Age b20 20% 11% 20–29 37% 36% influence of gender is identified for the reading or decision to 30–39 22% 28% refer stages. Further, we test the effect of age in the 40–49 15% 17% decision-making process and find that age has a significant 50–59 4% 6% positive influence on the decision to refer stage (coefficient: ≥ 60 1% 2% .036). C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 51

Discussion more influential because of their communication habits. An interesting finding is that degree centrality has no influence on This paper develops a three-stage model to analyze the the decision-making process. decision-making process of consumers in mobile viral marketing Our results indicate that one reason for the failure of mobile campaigns. This is an important area to study because companies viral marketing campaigns is that consumers tend to forward actively approach only a small number of consumers in viral messages to consumers on whom they have limited impact. marketing campaigns. Therefore, additional information about Consumers with low-quality contacts (i.e., weak ties) are more these consumers, their decision-making processes and the factors likely to forward the message because the perceived risk, that influence the consumers may help determine the success or which can include social sanctions, of forwarding a message is failure of viral marketing campaigns. Only a few extant studies low. Accordingly, the results of this study indicate a behavior focus on mobile viral marketing campaigns. Due to the increased that might limit the impact of viral campaigns. Specifically, use and penetration rate of cell phones and smartphones, most customers who predominantly possess weak ties are more consumers can be addressed via this new medium and channel. likely to read the message, and they are more likely to pass it Furthermore, cell phones combine unique characteristics such as on to other customers with predominantly weak ties. In ubiquitous computing, always on and immediate reactions, thus contrast, the receivers of the message tend to make their making mobile phones an attractive marketing channel for viral purchase decisions based on the content that they receive from marketing campaigns. high-quality (i.e., strong tie) contacts because such ties are The three-stage model approach allows us to gain a deeper perceived to be more trustworthy when making a purchase understanding of consumers' decision-making processes in decision (Rogers 1995).Thus,theviralprocessmayleadtoa mobile viral marketing campaigns. This study has produced high number of referrals that are ignored by their receivers, several key findings. thus potentially limiting the success of the campaign. Future The first key finding is the important role of consumers research should therefore examine the circumstances under who place high importance on the purposive value and which consumers make their purchase decisions and the products entertainment value of a message during the decision-making that they choose based on recommendations from strong and process. We find that the entertainment value significantly weak ties. influences all three stages. Purposive value significantly This study has several limitations that open additional influences the interest stage and the decision to refer stage, avenues for future research. First, because of our setting and but it does not influence the reading stage. This finding is because we conducted a viral marketing campaign in a mobile relevant when conducting a mobile viral marketing campaign. environment via text messages, we were not able to directly To attract customer attention, and thus to successfully pass the observe the referrals made. In our study, this variable is reading stage, companies should develop campaigns that self-reported via a survey, which is a limitation of the study. entertain customers. To have a significant influence on the Further research should conduct an experimental setting where decision to refer, it is important to address consumers who it is possible to observe the forwarding behavior using place importance on both purposive and entertainment value. behavioral data rather than survey data. Second, the partici- However, Dholakia, Bagozzi, and Pearo (2004) found that in pants of this study were members of an opt-in program online communities, purposive value has no direct effect on and had already agreed to receive mobile advertising text participation behavior. The differences between their results and messages for advertising reasons. Therefore, it remains our results can potentially be attributed to the differences between unclear how consumers who have not consented to receiving online technology and cell phones. We suggest that purposive messages would react to unsolicited mobile advertising value is mobile-specific because the mobile setting differs from messages. Third, mobile marketing is still an emerging field the Internet. People tend to spend a considerable amount of in comparison to e-mail marketing. Therefore, it is likely that time in online communities and therefore have potentially large consumers will pay attention to these (mobile marketing) amounts of content to read. This makes it difficult to identify the campaigns because they are rather novel. It remains to be seen content that may be meaningful to the contacts. On the contrary, at how these results may change with the increasing prevalence least to date, customers receive a limited amount of mobile text of mobile marketing campaigns in the future. Fourth, we messages. Thus, the purposive value can be judged, and if studied only one product category, a new music CD. Future customers believe that the text message is useful and has purpose, research should analyze what products or services are (more or they will be interested in it and will eventually decide to forward it. less) suitable for mobile viral marketing campaigns. Fifth, in We find that tie strength plays an important role in the last this study, we only focused on data from the responses of stage of the decision-making process and has a negative consumers who were seeded by the company. To further influence on both the reading stage and the decision to refer generalize the results, future research could also analyze stage. In other words, consumers are more likely to pass consumers who receive a message from a friend. Finally, we messages on to weak ties (i.e., low-quality contacts). Further, only analyzed mobile viral marketing via text messages. Future usage intensity has a positive influence at the 10-percent level, studies could analyze and compare the findings with mobile which indicates that heavy users may be influential in mobile viral marketing campaigns using media-rich formats such as viral marketing campaigns because they have a lower threshold multimedia messages because these formats can offer more to pass information to their contacts. Thus, heavy users are entertaining content to recipients. 52 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54

Appendix A. Measurement Scales Bampo, Mauro, Michael T. Ewing, Dineli R. Mather, David Stewart, and Mark Wallace (2008), “The Effects of the Social Structure of Digital Networks on Entertainment value I want to entertain my friends by sending enjoyable Viral Marketing Performance,” Information Systems Research, 19, 3, (Cronbach's α = .804) information. 273–90. Information exchange is fun by itself. Bettman, James R. (1979), An Information Processing Theory of Consumer It is a good way to spend time in an enjoyable way. Choice. Reading, MA: Addison-Wesley. Purposive value I feel like spreading information I have discovered Bian, Yanjie (1997), “Bringing Strong Ties Back In: Indirect Ties, Network (Cronbach's α = .738) to my friends. Bridges, and Job Searches in China,” American Sociological Review, 62, 3, I want to send information to my friends who may 366–85. be interested in it. Bishop, Yvonne M.M., Stephen E. Fienberg, and Paul W. Holland (1975), Usage intensity Number of mobile text messages each survey Discrete Multivariate Analysis. Cambridge, MA: MIT Press. participant wrote per day. Bone, Paula Fitzgerald (1995), “Word-of-Mouth Effects on Short-term and Network items Degree centrality (=number of contacts the respondent Long-term Product Judgments,” Journal of Business Research, 32, 3, (Burt 1984) names): Looking back over the last six months, who are 213–23. the people with whom you discussed an important Borgatti, Stephen P. (2005), “Centrality and Network Flow,” Social Networks, personal matter? Please write their first names or initials 27, 55–71. in the boxes below. Bowman, Douglas and Das Narayandas (2001), “Managing Customer-initiated Tie strength: How close do you feel to these people? Contacts with Manufacturers: The Impact on Share of Category Require- ments and Word-of-Mouth Behavior,” Journal of Marketing Research, 38, Note: Items for entertainment value & purposive value on seven-point Likert 3, 281–97. scale. Tie strength measured on five-point Likert scale. Brown, Jacqueline Johnson and Peter H. Reingen (1987), “Social Ties and Word-of-Mouth Referral Behavior,” Journal of Consumer Research, 14, 3, 350–62. Appendix B. Sequential Logit Model Specification Brown, Jo, Amanda J. Broderick, and Nick Lee (2007), “Word of Mouth Communication Within Online Communities: Conceptualizing the Online Social Network,” Journal of Interactive Marketing, 21, 3, 2–20. Consider the following model (cf. De Bruyn and Lilien 2008): Burt, Ronald S. (1984), “Network Items and the General Social Survey,” Social Networks, 6, 4, 293–339. Y = 1 if the recipient has not read the text message. Chandon, Pierre, Brian Wansink, and Gilles Laurent (2000), “A Benefit Y = 2 if the recipient has read the text message but has not Congruency Framework of Sales Promotion Effectiveness,” The Journal of – visited the website. Marketing, 64, 4, 65 81. Czepiel, John A. (1974), “Word of Mouth Processes in the Diffusion of a Y = 3 if the recipient has visited the website but has not Major Technological Innovation,” JournalofMarketingResearch,11,2, forwarded the message. 172–80. Y = 4 if the recipient has forwarded the message. De Angelis, Matteo, Andrea Bonezzi, Alessandro M. Peluso, Derek D. Rucker, and Michele Costabile (2012), “On Braggarts and Gossips: A Self- Enhancement Account of Word-of-Mouth Generation and Transmission,” The probabilities can be written as follows (Amemiya 1975): Journal of Marketing Research, 49, 4, 551–63. De Bruyn, Arnaud and Gary Lilien (2008), “A Multi-stage Model of Word-of- Mouth Influence Through Viral Marketing,” International Journal of P1(Y=1)=F(β1′x) − β ′ β ′ Research in Marketing, 25, 3, 151–63. P2(Y=2)=[1 F( 1 x)]F( 2 x) “ − β ′ − β ′ β ′ De Matos, Celso A. and Carlos A.V. Rossi, (2008), Word-of-Mouth P3(Y=3)=[1 F( 1 x)][1 F( 2 x)]F( 3 x) Communications in Marketing: A Meta-analytic Review of the Antecedents − β ′ − β ′ − β ′ P4(Y=4)=[1 F( 1 x)][1 F( 2 x)][1 F( 3 x)]. and Moderators,” Journal of the Academy of Marketing Science, 36, 4, 578–96. Dholakia, Utpal M., Richard P. Bagozzi, and Lisa Pearo (2004), “A Social The parameters β1 are estimated for the entire sample by dividing the sample into those who read the text message and those Influence Model of Consumer Participation in Network- and Small-group- β based Virtual Communities,” International Journal of Research in Marketing, who did not. The parameters 2 are estimated from the subsample 21, 3, 241–63. of recipients who read the text message by dividing it into two Dickinger, Astrid and Mirella Kleijnen (2008), “Coupons Going Wireless: groups: those who visited the website and those who did not. The Determinants of Consumer Intentions to Redeem Mobile Coupons,” Journal of Interactive Marketing, 22, 3, 23–39. parameters β3 are estimated from the subsample of recipients who visited the website by dividing the subsample into two groups: Drossos, Dimitris, George M. Giaglis, George Lekakos, Flora Kokkinaki, and Maria G. Stavraki (2007), “Determinants of Effective SMS those who forwarded the message and those who did not. Advertising: An Experimental Study,” Journal of Interactive Marketing, The likelihood functions for the above sequential logit model 7, 2, 16–27. can be maximized by sequentially maximizing the likelihood East, Robert, Kathy Hammond, and Wendy Lomax (2008), “Measuring the Impact functions of the three dichotomous models (De Bruyn and Lilien of Positive and Negative Word of Mouth on Brand Purchase Probability,” – 2008; Maddala 1983). International Journal of Research in Marketing, 25, 3, 215 24. EITO (Feb. 2012), “More Than Five Billion Mobile Phone Users,” http://www. eito.com/press/Press-Releases-2010/More-than-five-billion-mobile-phone-users References Accessed 13. Feb. 2012. Fischer, Claude S. (1982), To Dwell Among Friends — Personal Networks in Amemiya, Takeshi (1975), “Qualitative Response Models,” Annals of Town and City. Chicago: The University of Chicago Press. Economic and Social Measurement, 4, 3, 363–72. Fox, John (1997), Applied Regression Analysis, Linear Models, and Related Bacile, Todd J., Christine Ye, and Esther Swilley (2014), “Consumer Co- Methods. Thousand Oaks: Sage. production of Personal Media Marketing Communication: The Case of Frambach, Ruud T., Henk C. Roest, and Trichy V. Krishnan (2007), “The Mobile Coupons,” Journal of Interactive Marketing (forthcoming). Impact of Consumer Internet Experience on Channel Preference and Usage C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 53

Intentions Across the Different Stages of the Buying Process,” Journal of Moldovan, Sarit, Jacob Goldenberg, and Amitava Chattopadhyay (2011), Interactive Marketing, 21, 2, 26–41. “The Different Roles of Product Originality and Usefulness in Generating Frenzen, Jonathan and Kent Nakamoto (1993), “Structure, Cooperation, and the Word-of-Mouth,” International Journal of Research in Marketing,28,2, Flow of Market Information,” The Journal of Consumer Research, 20, 30, 109–19. 360–75. Neslin,ScottA.,C.Henderson,andJ.Quelch(1985),“Consumer Promotions Gatignon, Hubert and Thomas S. Robertson (1991), “Innovative Decision and the Acceleration of Product Purchases,” Marketing Science,4, Processes,” in Handbook of Consumer Behavior,” Thomas S. Trobertson 147–65. and Harold H. Kassarjian, editors. NJ: Prentice-Hall: Englewood Cliffs. Norman, Andrew T. and Cristel A. Russell (2006), “The Pass-along p. 316–48. Effect: Investigating Word-of-Mouth Effects on Online Survey Godes, David and Dina Mayzlin (2005), “Using Online Conversations to Study Procedures,” Journal of Computer-Mediated Communication, 11, 4, Word-of-Mouth Communication,” Marketing Science, 23, 4, 545–60. 1085–103. ——— and Dina Mayzlin (2009), “Firm-created Word-of-Mouth Commu- Okazaki, Shintaro (2008), “Determinant Factors of Mobile-based Word-of- nication: Evidence From a Field Test,” Marketing Science, 28, 4, Mouth Campaign Referral Among Japanese Adolescents,” Psychology and 721–39. Marketing, 25, 8, 714–31. Goldenberg, Jacob, Sangman Han, Donald R. Lehmann, and Jae Weon Hong Palka, Wolfgang, Key Pousttchi, and Dietmar G. Wiedemann (2009), “Mobile (2009), “The Role of Hubs in the Adoption Process,” Journal of Marketing, Word-of-Mouth — A Grounded Theory of Mobile Viral Marketing,” 73, 3, 1–13. Journal of Information Technology, 24, 2, 172–85. Granovetter, Mark (1973), “The Strength of Weak Ties,” American Journal of Phelps, Joseph E., Regina Lewis, Lynne Mobilio, David Perry, and Niranjan Sociology, 78, 6, 1360–80. Raman (2004), “Viral Marketing or Electronic Word-of-Mouth Advertising: Hennig-Thurau, Thorsten, Kevin P. Gwinner, Gianfranco Walsh, and Dwayne D. Examining Consumer Responses and Motivations to Pass Along Email,” Gremler (2004), “Electronic Word-of-Mouth via Consumer-opinion Plat- Journal of Advertising Research, 44, 4, 333–48. forms: What Motivates Consumers to Articulate Themselves on the Internet?,” Porter, Lance and Guy J. Golan (2006), “From Subservient Chickens to Brawny Journal of Interactive Marketing, 18, 1, 38–52. Men: A Comparison of Viral Advertising to Television Advertising,” Herr, Paul M., Frank R. Kardes, and John Kim (1991), “Effects of Word-of- Journal of Interactive Advertising,6,2,26–33. Mouth and Product-attribute Information on Persuasion: An Accessibility- Portes, A. (1998), “Social Capital: Its Origins and Applications in Modern Diagnosticity Perspective,” Journal of Consumer Research, 17, 4, Sociology,” Annual Review of Sociology, 24, 1–24. 454–62. Reagans, Ray and Bill McEvily (2003), “Network Structure and Knowledge Hinz, Oliver, Bernd Skiera, Christian Barrot, and Jan Becker (2011), “Seeding Transfer: The Effects of Cohesion and Range,” Administrative Science Strategies for Viral Marketing: An Empirical Comparison,” Journal of Quarterly, 48, 2, 240–67. Marketing, 75, 6, 55–71. Reichhart, Philipp, Christian Pescher, and Martin Spann (2013), “A Iyengar, Raghuram, Christophe Van den Bulte, and Thomas W. Valente (2011), Comparison of the Effectiveness of E-mail Coupons and Mobile Text “Opinion Leadership and Social Contagion in New Product Diffusion,” Message Coupons for Digital Products,” Electronic Markets, 23, 3, Marketing Science, 30, 2, 195–212. 217–25. Kiss, Christina and Martin Bichler (2008), “Identification of Influencers — Reingen, Peter H. and Jerome B. Kernan (1986), “Analysis of Referral Measuring Influence in Customer Networks,” Decision Support Systems, Networks in Marketing: Methods and Illustration,” Journal of Marketing 46, 1, 233–53. Research, 23, 4, 370–8. Kleijnen, Mirella, Annouk Lievens, Ko de Ruyter, and Martin Wetzels (2009), Rogers, Everett M. and David G. Cartano (1962), “Methods of Measuring “Knowledge Creation Through Mobile Social Networks and its Impact on Opinion Leadership,” Public Opinion Quarterly, 26, 3, 435–41. Intentions to Use Innovative Mobile Services,” Journal of Service Research, ——— (1995), Diffusion of Innovations. New York: The Free Press. 12, 1, 15–35. Ryu, Gangseog and Lawrence Feick (2007), “A Penny for Your Thoughts: Kratzer, Jan and Christopher Lettl (2009), “Distinctive Roles of Lead Users and Referral Reward Programs and Referral Likelihood,” Journal of Marketing, Opinion Leaders in the Social Networks of Schoolchildren,” Journal of 71, 1, 84–94. Consumer Research, 36, 4, 646–59. Steenkamp, Jan-Benedict E.M. and Katrijn Gielens (2003), “Consumer and Krishnamurthy, Sandeep (2001), “Viral Marketing: What Is It and Why Should Market Drivers of the Trial Probability of New Consumer Packaged Every Service Marketer Care?,” Journal of Services Marketing, 15, 6, 422–4. Goods,” Journal of Consumer Research, 30, 3, 368–84. Lavidge, Robert J. and Gary A. Steiner (1961), “A Model for Predictive Straits, Bruce C. (2000), “Ego's Important Discussants or Significant People: Measurements of Advertising Effectiveness,” Journal of Marketing,25,6, An Experiment in Varying the Wording of Personal Network Name 59–62. Generators,” Social Networks, 22, 2, 124–40. Levin, Daniel Z. and Rob Cross (2004), “The Strength of Weak Ties You Can Sultan, Fareena, Andrew J. Rohm, and Tao Gao (2009), “Factors Trust: The Mediating Role of Trust in Effective Knowledge Transfer,” Influencing Consumer Acceptance of Mobile Marketing: A Two- Management Science, 50, 11, 1477–90. country Study of Youth Markets,” Journal of Interactive Marketing, Lin, Nan, Walter M. Ensel, and John C. Vaughn (1981), “Social Resources and 23,4,308–20. Strength of Ties: Structural Factors in Occupational Status Attainment,” Sundaram, D.S., Kaushik Mitra, and Cynthia Webster (1998), “Word-of-Mouth American Sociological Review, 46, 4, 393–405. Communication: A Motivational Analysis,” Advances in Consumer Maddala, G.S. (1983), Limited Dependent and Qualitative Variables in Research, 25, 527–31. Econometrics. Cambridge: Cambridge University Press. Trusov, Michael, Randolph E. Bucklin, and Koen Pauwels (2009), Mare, Robert D. (1980), “Social Background and School Continuation Decisions,” “Effects of Word-of-Mouth Versus Traditional Marketing: Findings Journal of the American Statistical Association, 75, 370, 295–305. From an Internet Social Networking Site,” Journal of Marketing,73,5, Marsden, Peter V. and Karen E. Campbell (1984), “Measuring Tie Strength,” 90–102. Social Forces, 63, 2, 482–501. Tsang, Melody M., Shu-Chun Ho, and Ting-Peng Liang (2004), “Consumer ——— (2002), “Egocentric and Sociocentric Measures of Network Central- Attitudes Toward Mobile Advertising: An Empirical Study,” International ity,” Social Networks, 24, 4, 407–22. Journal of Electronic Commerce,8,3,65–78. Maxham, James G. and Richard G. Netemeyer (2002), “A Longitudinal Study Van den Bulte, Christophe and Yogesh Joshi (2007), “New Product Diffusion of Complaining Customers' Evaluations of Multiple Service Failures and with Influentials and Imitators,” Marketing Science, 26, 3, 400–21. Recovery Efforts,” Journal of Marketing, 66, 4, 57–71. Van der Lans, Ralf, Gerrit Van Bruggen, Jehoshua Eliashberg, and McCallister, Lynne and Claude S. Fischer (1978), “A Procedure for Berend Wierenga (2010), “A Viral Branching Model for Predicting Surveying Personal Networks,” Sociological Methods & Research,7,2, the Spread of Electronic Word-of-Mouth,” Marketing Science,29,2, 131–48. 348–65. 54 C. Pescher et al. / Journal of Interactive Marketing 28 (2014) 43–54 van Hoye, Greet and Filip Lievens (1994), “Tapping the Grapevine: A Closer Philipp Reichhart: Research interests include e- and m-commerce, mobile Look at Word-of-Mouth as a Recruitment Source,” Journal of Applied marketing, consumer behavior, word of mouth, social network and location- Psychology, 94, 2, 341–52. based services. Wirtz, Jochen and Patricia Chew (2002), “The Effects of Incentives, Deal Proneness, Satisfaction and Tie Strength on Word-of-Mouth Behaviour,” Martin Spann is a professor of electronic commerce and digital markets International Journal of Service Industry Management, 13, 2, 141–62. at the Munich School of Management, Ludwig-Maximilians-University Munich, Wojnicki, Andrea C. and David B. Godes (2008), “Word-of-Mouth as Self- Germany. His research interests include e-commerce, mobile marketing, Enhancement,” Working Paper. University of Toronto/Harvard Business prediction markets, interactive pricing mechanisms, and social networks. He has School. published in Management Science, Marketing Science, Journal of Marketing, Information Systems Research, MIS Quarterly, Journal of Product Innovation Christian Pescher: Research interests include B2B and B2C e-commerce, Management, Journal of Interactive Marketing, Decision Support Systems,and innovation, and social networks in marketing and forecasting. other journals.