INTRODUCTION
The combination of different media, platforms, and channels in brand communication is changing advertising theory, research, and practice. The various players—advertisers, advertising and media agencies, media owners, research companies, and academic advertising researchers—all are eager to contribute data, insights, and theories on uses and effects of these touchpoints in brand communication. The goal: how to effectively integrate them in advertising campaigns.
In this contribution, we have provided an overall assessment of current cross-platform issues and propose ideas for advancing both theory and practice. To this end, we have explored the most important peer-reviewed advertising and marketing journals as well as two of the most important industry sources.
We also conducted a brief survey among key players at media agencies about current practices to find out what they consider to be the major developments and challenges for the future. We conclude with a call to academia and industry for more cooperation and cumulative research on the topic.
ACADEMIC VERSUS INDUSTRY CROSS-PLATFORM-THEMED RESEARCH
We began by conducting a search for peer-reviewed academic articles on cross-media platform practices in the four most important advertising journals (Journal of Advertising Research [JAR], Journal of Advertising [JA], International Journal of Advertising [IJA], and Journal of Marketing Communications [JMC]), and three marketing journals (Journal of Marketing [JM], Journal of the Academy of Marketing Science [JAMS], and Journal of Interactive Marketing [JIM]). We also included Warc's Admap in our research as well as ESOMAR conference articles—two of the most important sources of information for the media and advertising industry.1
We found a total of 40 academic articles and 95 Admap/ESOMAR articles on cross-platform advertising. By far, the largest number of academic articles (18) was published in JAR, followed by JMC (7), and JIM (6). Admap published 40 articles, and ESOMAR published 55 in the same period.
Academic Papers
The majority of journal articles on cross-platform practices were empirical (88 percent). Of the total, 60 percent were based on real-life data and 40 percent were based on more “artificial” data such as experiments.
There was, however, a remarkable difference between the journals with regard to the type of data that were used:
Articles in JM and JIM were solely based on real-life data, as were the majority (63 percent) of the articles published in JAR. In the other journals, experiments were more common.
A large majority of the articles focused on theory development—with the noteworthy exception of JAR, where 50 percent of the articles emphasized practical advice.
The spotlight on theory development in academic journals also was reflected in the affiliations of the authors:
about 60 percent of the articles were written by academics employed at universities,
20 percent by practitioners, and
20 percent were jointly written by authors from university and industry.
A large majority of the academic articles discussed cross-media, cross-platform, or cross-channel effects. Some, for example, investigated whether campaigns that used multiple media were more effective in terms of brand responses and sales than campaigns that used only one medium. Only a few studies focused on how consumers use and combine multiple media in daily life or on their customer journey, identifying the touchpoint interactions customers have on the way to achieving their goal. Even fewer dealt with management issues related to cross-platform strategies.
Industry Papers
A very large majority of cross-platform articles in industry publications were data based. Of that majority, 70 percent appeared in Admap and 85 percent were in ESOMAR publications and conference materials.
The “nondata” articles described and analyzed cross-media research in practice, discussed the pros and cons of cross-media global deals, or reviewed the state of affairs and the reasons for the limited progress in the field.
Contrary to academic research, all the empirical articles dealt with real-life data; no articles were based on experimental research (with student participants). No article aimed at developing or testing theories, and surprisingly, none of the articles even referred to academic journal articles or theory.
One article was coauthored by an academic; all other industry articles were exclusively written by practitioners. Topics included data collection issues such as creating cross-media datasets, fusion procedures, and techniques to model the contribution of individual media to reach consumers and the effects of brand communication.
Most ESOMAR articles (63 percent) presented data on media use; the others (37 percent) analyzed the contribution of individual media to communication and brand effects.
For Admap this was different: A large majority (80 percent) of the empirical articles not only gave insight into media audiences' use of multiple media but also cross-media effects in campaigns. It was striking that there were almost no cross-references between articles, neither within the two types of publications nor between them, and the industry articles did not refer to academic journals. Some articles, in fact, gave the impression that they were primarily designed to promote an instrument offered by the company with which the author is affiliated.
On the basis of our quick scan of academic and industry literature, we offer some notable conclusions.
Attention to cross-media platforms is widespread, both in the academic literature as well as in industry publications.
The focus of articles in the two forums could not be more different. While academics are focused on theory, practitioners are focused on data collection, data modeling, and cross-media reach and effect issues.
Academics, in particular those publishing in advertising journals, use experiments, whereas practitioners always use real-life data.
Finally, but not less striking, the two strands of literature are separated: There are no cross-references between academic journals and industry outlets.
What We Learned from Academic Articles
When analyzing articles in the academic journals, four different reasons why combining multiple media in one campaign would be more effective than using only one medium or touchpoint can be noted:
Target Group Extension
By using multiple media, a campaign can reach a larger part of the target group, or drive efficiencies in audience targeting (e.g., Briggs, Krishnan, and Borin, 2005; Enoch and Johnson, 2010; Fulgoni and Lipsman, 2014; Taylor et al., 2013).
Complementary Effects
When different media are combined in a campaign, the campaign can profit from the strength of each of the media: media may complement each other (e.g., Dijkstra, Buijtels, and Van Raaij, 2005; Okazaki and Hirose, 2009; Tsao and Sibley, 2004).
Repetition
Varying the media in which ads are presented will reduce wear-out effects (e.g., Navarro-Bailon, 2012; Stammerjohan, Wood, Chang, and Thorson, 2005).
Synergy
The combined effect of multiple marketing communication activities can exceed the sum of their individual effects (e.g., Chang and Thorson, 2004; Havlena, Cardarelli, and De Montigny, 2007; Laroche, Kiani, Ekonomakis, and Richard, 2013; Naik and Peters, 2009; Naik and Raman 2003; Steele et al., 2013; Taylor et al., 2013; Vandeberg, Murre, Voorveld, and Smit, 2015; Voorveld, Neijens, and Smit, 2011; Voorveld and Van Noort, 2014; Wakolbinger, Denk, and Oberecker, 2009).
The literature also discusses several processes underlying cross-media synergy effects and proposes theoretical notions that can explain the effectiveness of campaigns, including multiple media or touchpoints. The five most important are the following:
Encoding Variability
When consumers are exposed to the same message in a variety of media rather than being repeatedly exposed to it in only one medium, information will be encoded in a more complex manner (Laroche et al., 2013; Stammerjohan et al., 2005; Tavassoli, 1998; Vandeberg et al., 2015; Voorveld et al., 2011; Voorveld and Valkenburg, 2015).
Multiple Source Credibility
When consumers are exposed to multiple media in a campaign, they might perceive these media as independent sources of information and therefore more credible and persuasive than messages from a single source (e.g., Chang and Thorson, 2004; Dijkstra, 2002; Laroche et al., 2013; Voorveld et al., 2011).
Forward Encoding
Exposure to an advertisement in one medium or on one platform stimulates interest and curiosity, which enhances processing of subsequent advertisement exposures in other media or platforms (e.g., Dijkstra, 2002; Edell and Keller, 1989; Voorveld et al., 2011).
Backward Retrieval
Mentally replaying advertisements previously viewed in one medium when exposed to advertisements in another medium adds to synergy effects (e.g., Chang and Thorson, 2004; Dijkstra, 2002; Edell and Keller 1989; Voorveld et al., 2011).
Central Processing (Elaboration Likelihood Model)
Individuals exposed to campaigns using multiple media, form their attitudes through central processing in contrast to individuals who are exposed to repeated messages in one medium who move along the peripheral route (Chang and Thorson, 2004; Laroche et al., 2013).
In addition to these main findings, the academic literature also investigated conditions that may influence the effects of cross-media campaigns. These included, among others:
the particular sequence of media use in a campaign;
fit or congruency between advertisements in the different media;
type of product;
the use of explicit versus implicit measures.
Moreover, the literature studied effects of different combinations of media, for example, the combination of Internet advertising with television, radio, or print advertising (e.g., Chang and Thorson, 2004; Havlena et al., 2007; Naik and Peters, 2009; Sheehan and Doherty, 2001; Steele et al., 2013; Taylor et al., 2013; Vandeberg et al., 2015; Voorveld, 2011; Voorveld et al., 2011; Voorveld, Neijens, and Smit, 2012; Voorveld and Valkenburg, 2015; Wakolbinger et al., 2009).
What We Learned from Industry Articles
The industry articles focused on different methods to measure use, reach, and frequency of cross-media combinations, such as:
single-source measurement of multiple media;
fusion (combining “silo” data from medium-specific syndicated audience research);
campaign tracking;
modeling.
In addition to using self-reports for data collection (recall data), articles considered passive measurement methods (registration), and digital analytics (big data), addressing their promises and pitfalls. In particular, the articles presented
methods to measure the use of different types of media, from self-reports to passive measurement, such as people meters, click registration, a media watch, or other gadgets;
methods to determine the cumulative reach of different media;
new metrics, like engagement, that go further than the simple people-with-open-eyes-or-ears-in-front-of-advertising-space definition of reach;
joint industry initiatives (such as the Institute of Practitioner's [IPA's] Touch Points study);
challenges facing market research, including
the problematic validity and reliability of recall measures,
samples that are not representative because of increasing demands on respondents,
the cost of single-source data,
complicated fusion techniques,
lack of individual-level characteristics in big data, and
problems with passive measurement of mobile and tablets using some operating systems (Barley, 2015).
On the basis of various data, articles also contributed insights into successful cross-media strategies:
Multichannel campaigns are more effective than single channel ones: the more channels used, the more effective the campaign, partly as a result of budget effects (Cox, 2011, who analyzed IPA data).
Television is (still) very important in the media mix (Cox, 2011; Field, 2007).
Advertising coupled with a direct sales channel (direct marketing or sales promotion) is the most effective route to achieve sales (Cox, 2011).
Emotional campaigns are more powerful, even in “rational” categories (Binet and Field, 2008).
The combination of paid, owned, and earned media only works if messaging and media are integrated. “At the heart of any successful POE ecosystem has to sit a communications idea that directly links Paid to Owned and/or Paid to Earned media” (Green, 2011).
The Media Agency Experience
Five of the largest media agencies in the Netherlands participated in this survey about current cross-platform practices and issues for the future. These Dutch media agencies have a high reputation and are part of international chains. It can be fairly assumed that their responses are representative for the international media planning industry. There were two types of informants—one ceo and four directors of research and insights. They were asked in particular about
the quality of their cross-platform data;
how they organize cross-platform planning;
which issues they perceive for the future;
their wishes.
All the respondents were somewhat satisfied with the quality of the cross-media planning data they have at their disposal.
They all use an integrated dataset combining syndicated audience research on television, radio, print, and online media. These data are collected on a continuous basis supervised by joint industry committees composed of representatives of media owners, advertisers, and media agencies. The integrated dataset is comparable to IPA's Touch Points.2
They are equipped with datasets that combine consumers' media use and shopping behavior into a single source, such as the Media Efficiency Panel (MEP, by GfK/Google), which is particularly strong on television and online data.3
Next to these datasets, which are available to the whole media market, the media agencies build their own datasets based on campaign effect studies, client data, passive measurement of digital media, “enrichment” of joint industry datasets, modeling, and so forth.
The media agencies also indicated that their organizations have been transformed to facilitate cross-media planning. Several aspects are important:
Agencies have allocated communication planners a much broader task than in the past when organizations had separate planners for television, radio, print, and digital specializations such as customer relationship marketing (CRM), search engine optimization (SEO), search engine advertising (SEA), social, and performance marketing.
Most agencies have replaced—at least for their most important clients—the media buying structure (the “silos” for television, print, etc.) with “client teams” or “client villages,” which are multidisciplinary teams of communication planners working together for a client, particularly in the strategy and evaluation phases of the planning process.
To further stimulate cross-media planning, media agencies have also implemented cross-media planning systems and cross-media financial structures.
The survey participants expressed concerns as well. To begin with, some traditional cross-media issues remain unanswered. For example,
How can one get single-source data on all touchpoints, measured in a valid and reliable way?
How can the value of a contact with different media, or the same medium on different devices, be compared and combined in a campaign?
How does one get insight into the relevance of various touchpoints in the different stages of the decision-making process? In other words, which touchpoints matter in which phase of the consumer's journey, and how are they connected?
Finally, although inserting multiple media in a campaign might contribute to reaching different goals, the question remains under what conditions inserting multiple media in a campaign leads to synergy effects (1 + 1 = 3), summative effects (1 + 1 = 2), or antergy effects (1 + 1 = 1.5; see also Ehrenberg-Bass, 2012).
The digitalization of media creates new—almost real-time—possibilities for mapping the customer's journey along the media touchpoints toward buying. Digitalization also generates new possibilities for media planning, such as programmatic buying, automated trading, real-time bidding, and segmentation based on location, activity, social setting, and mood.
The industry informants believed, therefore, that media planning can and should become less static and more flexible, responding to consumers' constantly changing media preferences.
CONCLUSIONS AND RECOMMENDATIONS
The Need for Collaboration
Academics and practitioners should come out of their separate worlds and make a serious effort to collaborate. Our literature study shows that academic and industry research are not connected. This is striking and would not be possible in fields like medicine, where patients expect their doctors to continuously update their knowledge using the latest academic insights. Not only, however, do practitioners need to make an effort to establish relationships with academics but also academics need to study the real issues that practitioners face.
The collaboration can be based on the tenet that academics have time but do not have the money for real-life data collection and practitioners have these data but do not have the time to analyze them beyond their immediate needs. Academia could contribute to the industry by developing theories about cross-platform advertising effects, providing tools for better and cheaper measurement of these effects, by educating new generations of media planners and serving as knowledge brokers.
The authors of this study make a plea for more collaboration between advertisers, media owners, media agencies, research companies, and academics. It is impressive to note the current initiatives and the level of cooperation that already has been established, but more should be done: The different parties in the media industry should overcome their self-interests and realize that only a joint effort can help further the profession.
It is interesting to note that this plea is supported by the Advertising Research Foundation (ARF) and the Association of National Advertisers (ANA) in the United States. As Bob Liodice, ANA president/ceo observed:
Across the ANA and the ARF, we concur that cross-platform analytics are exactly what this industry needs, because we are doing so much integrated marketing. Brands must develop a ‘far more scientific’ approach than has predominated thus far—a goal that, in turn, depends upon the major associations working closely together. We can only do that by achieving a level of collaboration across the industry.4
It is probably no surprise, then, that advertisers want the market parties to collaborate more. After all, they pay and have the biggest interest in high quality cross-platform planning.
ABOUT THE AUTHORS
Peter Neijens holds a Chair in Media and Persuasion at the Amsterdam School of Communication Research (ASCoR) at the University of Amsterdam. He has published about 200 articles in national and international journals and books. He serves on the editorial boards of the Journal of Advertising Research, International Journal of Advertising, Journal of Advertising, Journal of Marketing Communications, and Oxford Bibliographies (Communication). Neijens is a past president of the European Advertising Academy. In 2013, he was awarded the first Netherlands-Flanders Communication Association Career Award for a lifetime of scholarly achievement.
Hilde Voorveld is assistant professor of persuasive communication at ASCoR. Her research interests include cross-media strategies, media multitasking, and interactive marketing communication. Voorveld has published about 40 articles in national and international journals and books, such as the Journal of Advertising Research, Journal of Advertising, Computers in Human Behavior, and Media Psychology. She is a board member of the European Advertising Academy.
Footnotes
↵1 We searched for articles published between 2000 and 2015, using the keywords “cross-media,” “multimedia,” “cross-platform,” “multiplatform,” “cross-channel,” “multichannel,” “cross-touchpoints,” and “multitouchpoints.” We checked the abstract of each article to see whether these concepts were actually the main topic.
↵2 Touch Points, commissioned by the Institute of Practitioners in Advertising (IPA) in the United Kingdom, is a consumer-centered, multimedia database that produces information on consumers' use of all media. The project was launched in 2006 (see Smit and Neijens, 2011; http://www.ipa.co.uk/touchpoints).
↵3 MEP was also introduced in Germany and the United Kingdom.
↵4 “ARF and ANA Seek New Metrics.” Warc News, March 21, 2014.
- © Copyright 2015 The ARF. All rights reserved.