Uncovering patterns in mobile advertising opt-in behaviour: a decision hierarchy approach

Transcription

1 Int. J. Mobile Communication, Vol. X, No. X, xxxx 1 Uncovering patterns in mobile advertising opt-in behaviour: a decision hierarchy approach Stuart J. Barnes Norwich Business School, University of East Anglia, Norwich NR4 7TJ, UK Fax: stuart.barnes@uea.ac.uk AU: Please specify the corresponding author. Eusebio Scornavacca School of information Management, Victoria University of Wellington, PO Box 600, Wellington 6001, New Zealand Fax: eusebio.scornavacca@vuw.ac.nz Abstract: The aim of this paper is to investigate the weighting of specific criteria in the consumer s decision to opt-in to a mobile advertising campaign. We analysed the decision hierarchy for 111 respondents and tested for significant differences between the global priorities of various groupings. The results demonstrate considerable disparity between the decision-making criteria of a number of groups including by income, gender, average volume of advertising messages received and the influence of mobile advertising on previous purchases. The paper concludes with an examination of the implications of the results for practice and suggestions for future research in this area. Keywords: campaign; decision hierarchy; global priorities; mobile advertising; mobile marketing; opt-in. Reference to this paper should be made as follows: Barnes, S.J. and Scornavacca, E. (xxxx) Uncovering patterns in mobile advertising opt-in behaviour: a decision hierarchy approach, Int. J. Mobile Communication, Vol. x, No. x, pp.xx xx. Biographical notes: Stuart J. Barnes is a Chair and Professor of Management in the Norwich Business School at the University of East Anglia. Previously he worked at Victoria University of Wellington, New Zealand and the University of Bath. He holds a PhD from Manchester Business School. His primary research interests focuses on the successful utilisation of new information and communications technologies by businesses, governments and consumers. He has published five books and more than 100 articles including those in journals such as Communications of the ACM, the International Journal of Electronic Commerce, European Journal of Marketing and Information and Management. Copyright 200x Inderscience Enterprises Ltd.

2 2 S.J. Barnes and E. Scornavacca Eusebio is a Lecturer of Electronic Commerce in the School of Information Management at Victoria University of Wellington, New Zealand. Before moving to Wellington, Eusebio spent two years as a Researcher at Yokohama National University, Japan. He has published and presented more than 50 papers in conferences and academic journals and is currently on the editorial boards of the International Journal of Mobile Communications and the International Journal of Electronic Finance. His research interests include mobile business, electronic business, e-surveys and IS teaching methods. He was awarded as the MacDiarmid Young Scientist of the Year awards in Introduction The tremendous global penetration of digital wireless telephony estimated at more than 2.5 billion users worldwide has provided an unprecedented platform for new services delivered via mobile handsets (Scornavacca, Barnes and Huff, 2006; Usability News, 2006). Since the average revenue per user from voice services has been declining for several years, operators have been keenly looking to data communications services as the new revenue stream. One key area is that of mobile marketing. Mobile or wireless marketing a subset of electronic marketing is defined as All activities required to communicate with customers through the use of mobile devices in order to promote the selling of products or services and the provision of information about these products and services. (Ververidis and Polyzos, 2002) The most popular mobile advertising model involves push-based text messages (Barwise and Strong, 2002; Dickinger, Scharl and Murphy, 2005). The wireless channel benefits from the potential for detailed user information and personalisation; the message can be tailored for each customer to enable better targeting. Since mobile phones are personal objects, marketers can specifically address the person targeted, as well as recognising their social context, individual preferences, time and location (Barnes, 2002; Barnes and Scornavacca, 2004; Dickinger et al., 2004; Hristova and O Hare, 2004; Okazaki, 2004; Jelassi and Enders, 2006; Scornavacca and McKenzie, 2006). Further, text message ads have been found to boost consumers inclination to purchase considerably, via a high response rate (Enpocket, 2005a,b). The aim of this paper is to investigate the consumer s decision to opt-in to a mobile advertising campaign. The research question is: What are the priorities for consumers when deciding to opt-in to a mobile advertising campaign and how do they differ between respondent groups? We were interested in the weighting of specific criteria in the consumer s opt-in decision. The research approach is based on multi-criteria decision-making. The format of the paper is as follows. In the next section, we provide a very brief background to the research problem. This is followed by the methodology and the findings from the research. Finally, we round off with conclusions and implications for research and practice.

3 Uncovering patterns in mobile advertising opt-in behaviour 3 2 Literature review In this section, we provide a brief introduction to the general and specific areas for investigation mobile advertising and the mobile advertising opt-in decision. 2.1 Background to mobile advertising Mobile advertising has typically been categorised into push- and pull-models (Barnes, 2002). In the pull-model campaign, the marketer sends the information requested by the consumer; whereas in the push-model campaign, the marketer takes the initiative to send messages to the consumer, such as via text messaging. A third type of campaign, as suggested by Jelassi and Enders (2006), revolves around the mobile dialogue model, where the marketer tries to build a long lasting relationship with the consumer. The most prevalent mode of mobile advertising is Short Messages Service (SMS) to handheld devices, notably mobile phones. SMS, known as text messaging, is a store-and-forward communication system for the mobile phone. Recent variants, such as Multimedia Message Service (MMS) have added multimedia capabilities. Barwise and Strong (2002) identify six ways of using SMS for advertising: brand building, special offers, timely media teasers, competitions, polls/voting, products, services and information requests. Text message ads have been found to boost consumers inclination to purchase by 36%, which partly explains its growing popularity among marketers (Enpocket, 2005b) and the growing amount of related research (Kavassalis et al., 2003; Bauer et al., 2005; Haghirian, Madlberger and Tanuskova, 2005; Leppäniemi and Karjaluoto, 2005; Nysveen, Pedersen and Thorbjornsen, 2005; Virtanen, Bragge and Tuunanen, 2005; Carroll et al., 2006; Karjaluoto and Kautonen, 2006; Merisavo et al., 2006; Scornavacca and McKenzie, 2006; Bamba and Barnes, in press; Haghirian and Inoue, 2007; Heinonen and Strandvik, 2007). According to Enpocket (2005a), text message campaigns also deliver a 15% response rate, which they estimate is twice as much as direct mail or campaigns; apparently, text messages are 50% more successful at building brand awareness than TV and 130% more than radio (Enpocket, 2005b). Mobile advertising has clear potential for precise targeting and personalisation, recognising social context, individual preferences, time and location. The acceptance of SMS advertising has been examined by Bauer et al. (2005). Using a large sample and structural equation modelling they find that the most important factors that affect attitude toward mobile marketing are: consumers attitudes toward advertising in general; perceived utility (in terms of information, entertainment and social aspects); perceived risk (in terms of privacy and data security); consumers knowledge about the technology; social norms that impact on consumers behaviour. 2.2 Background to the opt-in problem A core issue for advertising is privacy: an individual s control over information held about them by third parties (Petty, 2000; Phelps, Wowak and Ferrell, 2000; Tezinde, Smith and Murphy, 2002). In the UK, under Privacy and Electronic Communication Regulations, permission is a requirement of SMS ads, as is opt-out and data protection from misuse and inaccuracy. Similar legislation is found in other parts of the EU;

4 4 S.J. Barnes and E. Scornavacca e.g. in Scandinavia you cannot approach clientele with SMS in any way before obtaining permission and so other media must be used to attract attention. The principal way of obtaining permission is by signing a contract. Bamba and Barnes (in press) found that the preferred form of receiving such a contract is mainly divided between online and SMS-based (both 41.8%) versions. They also found that freedom to opt-out (95.2%), message frequency (89.2%) and control over third parties (95%) were the most important aspects of the contract. Time and location are less important although still notable. Bamba and Barnes (in press) also found that permission to receive SMS ads only tends to occur when consumers have a high control over opt-in conditions, when the SMS ad is relevant and when brand is familiar. Using a scenario-based analysis they discovered that brand familiarity is significantly less important than either ad relevance or control over opt-in conditions, which share similar importance. This supports the earlier research by Carroll et al. (2006). Thus, even big brand names need to be wary of the way advertising campaigns are approached. Barnes and Scornavacca (2004) suggested that Wireless Service Provider (WSP) control is found to increase the probability of user acceptance to mobile advertising. In this situation, consumers trust the WSP to monitor and control mobile communications. Trust as a generalised expectation (Rotter, 1967) is defined as a willingness to rely on an exchange partner in whom one has confidence (Moorman, Deshpande and Zaltman, 1993). Trust conveys a confidence in the behaviour of the other or an ability to depend on the other or to predict the others behaviour (Thorelli, 1990; Gundlach and Murphy, 1993). If users are likely to have high levels of trust with their WSP (Ericsson, 2000; Enpocket, 2002) this transfers into their willingness to accept messages though this WSP. This new channel of advertising clearly has potential. However, it is not without problems. To appeal to consumers, marketers must adapt their ads to individual use of text messaging and to areas of interest. Successful mobile advertising starts with a consumer base of interested consumers who have already given permission to receive certain advertisements. 3 Methodology This research is predominantly based on a multi-criteria decision-making approach. This is the first phase of a multi-phase research project that is attempting to model consumer opt-in decisions in a mobile advertising context and that will ultimately apply decision models to real-life situations in a practical context using the Analytical Hierarchy Process (AHP). However, in this study, there is no specific decision that is being made between products or services, e.g. between buying product A and B; the decisions being evaluated are between various criteria and their relative importance in deciding whether to opt-in to receiving advertising messages on the respondent s mobile phone. In particular, we were interested in finding the general basis for an individual s decision to opt-in to a mobile advertising campaign with a view to developing a clearer picture of the decision model (in this case the decision hierarchy) in operation. Based on a review of the salient research in this area and on concurrent studies by the authors (e.g. Kavassalis et al., 2003; Bamba and Barnes, in press; Virtanen, Bragge and Tuunanen, 2005; Leppäniemi and Karjaluoto, 2005; Haghirian, Madlberger and Tanuskova, 2005; Nysveen, Pedersen and Thorbjornsen, 2005; Bauer et al., 2005;

5 Uncovering patterns in mobile advertising opt-in behaviour 5 Karjaluoto and Kautonen, 2006; Scornavacca and McKenzie, 2006 Carroll et al., 2006) a broad set of decision criteria were developed. The decision criteria included: A The context of receiving a message, e.g. time, location, frequency and social context. B The opinion of others, e.g. friends, family and work colleagues. C The brand contained in the message (in terms of awareness, trust and loyalty). D The characteristics of the message, e.g. content, media and uniqueness. E The degree to which messages complement other channels of advertising. F The value of the message (information, entertainment or for social interactions). G The control of the advertising channel by the operator to filter content. A number of second and third order sub-criteria were also included in questionnaire, but not for all first-order categories, since one of these (message context) would be the focus of another, related study. We wanted to make the size of the questionnaire more manageable. The questionnaire was developed for use in the Expert Choice software package. For each parent node in the hierarchy, starting with the top node, Goal: The decision to opt-in to a mobile advertising campaign, the respondent was required to rate every permutation of decision criteria options. Thus, the respondent would be asked to rate the decision between criterions A and B on a scale from 1 (=equal, the mid-point) through 3 (=moderate), 5 (=strong), 7 (=very strong) and 9 (=extreme) preference for A or B. The questionnaire also collected information about gender, age, personal income, mobile replacement plans, average text messages sent and received per day, average advertising messages received per week and whether mobile advertising has influenced a purchase decision in the past. The questionnaire was administered in a major New Zealand University in May Only respondents using a mobile phone that can send and receive text messages were eligible to complete the survey. In all, 111 useable responses were received. The responses were input into Expert Choice for analysis. Thus, the decision priorities of the sample and of various groups were examined, compared and contrasted. Unfortunately, the software does not allow statistical testing of decision priorities between groups, which we were very interested in. To enable this, we printed the global decision hierarchy for each respondent and manually input the data into Excel, along with demographic and other details from Expert Choice. The data was then imported into SPSS for statistical testing. This enabled us to use ANOVA to compare mean global priorities and for additional graphical output such as radar charts. 4 Findings In this section, we summarise details of the sample and the overall decision model, and illustrate and test differences in the decision models of different groups.

6 6 S.J. Barnes and E. Scornavacca 4.1 Description of the sample The sample collected was 63.1% female, 36% male and with one non-response. Approximately 90% of the sample was under 25 years of age and three-quarters had an income of less than NZ$2,500 per month (approximately 1,320 or US$1,670). Just over half of the sample planned to replace their mobile phone within a year. 35.1% of the respondents sent and received more than 10 text messages per day and 79.3% received Zero or one advertisement on their mobile phone per week. Overall, 18.9% of respondents stated that they had previously bought a product or service as a result of mobile advertising. 4.2 Summary of the decision sample The overall decision hierarchy of the sample (n = 111) is given in Figure 1 (where L = local priority and G = global priority), while Table 1 gives a summary of the decisions between criteria in a matrix format (see the list of criteria above for key to the coding of A G). Table 1 shows an acceptably low level of inconsistency (1%). Figure 1 and Table 1 together give a picture of how the sample decided between characteristics. Message context has the highest global priority (G: 0.213) and its preference was expressed over all other characteristics in Table 1, with preferences over channel complementarity (E), message value (F) and message characteristics (D) being the highest. At the other end of the spectrum, message value (F) and operator control (G) had the lowest first-order priorities (G: and L: 0.099, respectively) in Figure 1, and all other characteristics were preferred to them in Table 1, particularly message context (A), message characteristics (D), message brand (C) and social influence (B). Overall, there appears to be a little to choose between the second-order priorities of message value (B). The first-order priorities of other characteristics fall in between, with social influence rating at G: 0.156, followed by message brand (G: 0.153), message characteristics (G: 0.137) and channel complementarity (G: 0.135). Trust and loyalty for a brand had the highest local priorities for C, while television and website advertising were the preferred channels for E. There appears to be a balance in preferred message characteristics, although personal message filtering is slightly ahead in priority (L: 0.382). Video and graphical outputs were by far the most preferred forms of advertising media. Table 1 Matrix summary of preferences for first order characteristics A B C D E F G A B C ( ) D ( ) E F G Incon: 0.01

7 Uncovering patterns in mobile advertising opt-in behaviour 7 Figure 1 Decision hierarchy for the sample (n = 111) 4.3 Differences in priorities between respondent groups We used some different methods to analyse decision hierarchies than is typically the case. The respondent decision data was compared for various groupings in Excel and SPSS. Figures 2 5 show radar charts of four groups compared in terms of their first-order priorities. For each figure, the blue line represents the overall sample. Alongside, ANOVA was used to statistically test the differences in priorities for these groups, plus messages sent, messages received, frequency of phone replacement and respondent age. There were no significant differences in the high and low groupings of age and volumes of messages sent and received. There was third-level item that was different (p = 0.014) between those that planned to replace their phone in under or over a year: the use of video content. Presumably this means people wish to upgrade their phones sooner to receive better video capabilities. Figures 2 5 show the groupings for which the differences were most pronounced. In Figure 2, the biggest differences between high and low income groups appear to be in terms of receiver s context, brand and message characteristics. The first two are

8 8 S.J. Barnes and E. Scornavacca significantly different (p = and 0.03 levels, respectively), with the second-order priority of brand loyalty being significantly different (p = 0.02). Other notable differences were second-order, namely message uniqueness (p = 0.023) and information value of the message (p = 0.038). For gender (see Figure 3), the biggest differences in priority were again in message brand (p = 0.045), driven by brand trust (p = 0.042) and in message value (p = 0.01), driven by information value (p = 0.006). Figure 2 Global priorities income Figure 3 Global priorities gender Looking at buying behaviour (see Figure 4), the biggest second-order differences were in brand trust (p = 0.023), message uniqueness (p = 0.017) and websites as a complementary advertising channel (p = 0.027). Finally, Figure 5 looks at differences in advertising messages received. The biggest second-order differences were in brand trust (p = 0.038) and websites as a complementary advertising channel (p = 0.017).

9 Uncovering patterns in mobile advertising opt-in behaviour 9 Figure 4 Global priorities buyers Figure 5 Global priorities ads received 5 Conclusions The results demonstrate significant differences between the decision-making criteria of various groups including by income, gender, volume of advertising messages received and the influence of mobile advertising on previous purchases. The implications for practice are that every group of consumers for a mobile marketing message must be understood not only in terms of consumption patterns, but also in terms of decision patterns involved in the initial decision to opt-in to a mobile advertising campaign. Every

10 10 S.J. Barnes and E. Scornavacca advertising campaign must be packaged appropriately to achieve opt-in to the right audience. In the present environment, no mobile advertising will succeed unless it overcomes the opt-in hurdle; before mobile customers even view advertising messages they must choose to receive them. For example, female respondents had significantly less brand trust and more preference for message value, particularly information; opt-in signals for female respondents are likely to be more informative since brand may be less influential. Similarly, those who received more advertising messages trust brand more probably as a filter and link more to other electronic channels, namely websites. Thus links to and from branded websites may be more likely to obtain opt-in for this group. Among higher income groups, message context was preferred less and brand loyalty, message uniqueness and information value were more pertinent; those with higher income appear more attuned to branded, unique and valuable content. Similarly, those who have bought from mobile ads before are brand sensitive and value message uniqueness more. They also see websites as a more complementary channel. This last group provides some valuable guidance; those who tend to buy from mobile advertising appear influenced by brand trust, message uniqueness and multiple electronic channels. The research has utilised some fairly novel techniques for examining decision priorities. However, there are a number of limitations to the research. First, the sample is limited, both in terms of the sample size and types of respondents; although young persons are a core target group for mobile advertising, a broader sample would be preferred. Secondly, the example is a hypothetical one, since no real products or services are mentioned. The research will be extended in the future to apply the AHP to specific opt-in mobile advertising decision choices for selected campaigns. Thirdly, the decision hierarchy may be further developed in the future. Message context, e.g. can be extended in the second-order priorities to include time, location, message frequency and social context. We recommend further focus groups to identify additional influences on the decision process. References Bamba, F. and Barnes, S.J. (in press) SMS advertising, permission and the consumer: a study, Business Process Management Journal, Special Issue on Mobile CRM. Barnes, S.J. (2002) Wireless digital advertising: nature and implications, Int. J. Advertising, Vol. 21, pp Barnes, S. and Scornavacca, E. (2004) Mobile marketing: the role of permission and acceptance, Int. J. Mobile Communications, Vol. 2, pp Bauer, H., Barnes, S.J., Neumann, M. and Reichardt, T. (2005) Driving consumer acceptance of mobile marketing: a theoretical framework and empirical study, Journal of Electronic Commerce Research, Vol. 6, pp Barwise, P. and Strong, C. (2002) Permission-based mobile advertising, Journal of Interactive Marketing, Vol. 16, pp Carroll, A., Barnes, S.J., Scornavacca, E. and Fletcher, K. (2006) Consumer perceptions and attitudes toward SMS advertising: recent evidence from New Zealand, Int. J. Advertising, Vol. 26, pp Dickinger, A., Haghirian, P., Murphy, J. and Scharl, A. (2004) An investigation and conceptual model of SMS marketing, Paper presented in the Proceedings of the 37th Hawaii International Conference on System Sciences, Hawaii, January.

11 Uncovering patterns in mobile advertising opt-in behaviour 11 Dickinger, A., Scharl, A. and Murphy, J. (2005) Diffusion and success factors of mobile marketing, Electronic Commerce Research and Applications, Vol. 4, pp Enpocket (2002) Consumer Preferences for SMS Marketing in the UK. Available at: (accessed 13 March 2003). Enpocket (2005a) Direct Response Report. Available at: (accessed 10 February 2006). Enpocket (2005b) Brand Performance of SMS Advertising. Available at: (accessed 10 February 2006). Ericsson (2000) Wireless Advertising. Stockholm: Ericsson Ltd. Gundlach, G.T. and Murphy, P.E. (1993) Ethical and legal foundations of relational marketing exchanges, Journal of Marketing, Vol. 57, pp Haghirian, P., Madlberger, M. and Tanuskova, A. (2005) Increasing advertising value of mobile marketing: an empirical study of antecedents, Paper presented in the Proceedings of the 38th Hawaii International Conference on System Sciences, Hawaii, January. Haghirian, P. and Inoue, A. (2007) An advanced model of consumer attitudes toward advertising on the mobile internet, Int. J. Mobile Communications, Vol. 5, pp Heinonen, K. and Strandvik, T. (2007) Consumer responsiveness to mobile marketing, Int. J. Mobile Communications, Vol. 5, pp Hristova, N. and O Hare, G. (2004) Ad-me: wireless advertising adapted to the user location, device and emotions, Paper presented in the Proceedings of the 37th Hawaii International Conference on System Sciences, Hawaii, January. Jelassi, T. and Enders, A. (2006) Mobile advertising: a European perspective, in S.J. Barnes and E. Scornavacca (Eds), Unwired Business: Cases in Mobile Business (pp.82 95). Hershey: IRM Press. Karjaluoto, H. and Kautonen, T. (2006) Trust as a factor affecting consumers participation in mobile marketing, Paper presented in the Proceedings of the 35th European Marketing Academy Conference, Athens, Greece, May. Kavassalis, P., Spyropoulou, N., Drossos, D., Mitrokostas, E., Gikas, G. and Hatzistamatiou, A. (2003) Mobile permission marketing: framing the market inquiry, Int. J. Electronic Commerce, Vol. 8, pp Leppäniemi, M. and Karjaluoto, H. (2005) Factors influencing consumer willingness to accept mobile advertising: a conceptual model, Int. J. Mobile Communications, Vol. 3, pp Merisavo, M., Vesanen, J., Arponen, A., Kajalo, S. and Raulas, M. (2006) The effectiveness of targeted mobile advertising in selling mobile services: an empirical study, Int. J. Mobile Communications, Vol. 4, pp Moorman, C., Deshpande, R. and Zaltman, G. (1993) Factors affecting trust in market research relationships, Journal of Marketing, Vol. 57, pp Nysveen, H., Pedersen, P.E. and Thorbjornsen, H. (2005) Intentions to use mobile services: antecedents and cross-service comparisons, Journal of the Academy of Marketing Science, Vol. 33, pp Okazaki, S. (2004) How do Japanese consumers perceive wireless advertising? A multivariate analysis, Int. J. Advertising, Vol. 23, pp Petty, R.D. (2000) Marketing without consent: consumer choice and cost, privacy and public policy, Journal of Public Policy and Marketing, Vol. 19, pp Phelps, J., Wowak, G. and Ferrell, E. (2000) Privacy concerns and consumer willingness to provide personal information, Journal of Public Policy and Marketing, Vol. 19, pp Rotter, J. (1967) A new scale for the measurement of interpersonal trust, Journal of Personality, Vol. 35, pp Scornavacca, E. and McKenzie, J. (2006) Perceived critical success factors of SMS-based campaigns: a managerial perspective, Paper presented in the Proceedings of the 14th European Conference on Information Systems, Gothenburg, Sweden, June.

12 12 S.J. Barnes and E. Scornavacca Scornavacca, E., Barnes, S.J. and Huff, S.L. (2006) Mobile business research published in : emergence, current status, and future opportunities, Communications of the Association for Information Systems, Vol. 17, pp Thorelli, H. (1990) Networks: between markets and hierarchies, in D. Ford (Ed.), Understanding Business Markets. London: IMP Group. Usability News (2006) Mobile Phones Hit 2.5 Billion Mark. Available at: (accessed 10 November 2006). Tezinde, T., Smith, B. and Murphy, J. (2002) Getting permission: exploring factors affecting permission marketing, Journal of Interactive Marketing, Vol. 16, pp Ververidis, C. and Polyzos, G. (2002) Mobile marketing using location-based services, Paper presented in the Proceedings of the First International Conference on Mobile Business, Athens, Greece. Virtanen, V., Bragge, J. and Tuunanen, T. (2005) Barriers for mobile marketing and how to overcome them, Paper presented in the Proceedings of the Hong Kong Mobility Roundtable, Hong Kong, May.