B&E | The Rise of Big Data’s Use in Advertising
IndraStra Open Journal Systems
IndraStra Global

B&E | The Rise of Big Data’s Use in Advertising

By Nick Couldry and Joseph Turow 

Image Attribute: epSos .de / Creative Commons

Advertising subsidies support publishers throughout the media system. Publishers are organizations that produce and distribute content through analog or digital means; think of newspapers, television production firms, and search engines. The advertising subsidy involves payment for the right to persuade the publisher’s audience to purchase or otherwise support a product or service. Traditionally, that has meant purchasing space or time on or near a publisher’s content—for example, a newspaper article or an Internet video. 

Most people likely think of advertising in terms of its most visible manifestation, the persuasive message. Yet the activity involves two sets of activities in addition to the creation of the ad. One part, traditionally called media planning and buying, revolves around the strategic consideration and provision of funds to pay for placement of the notice. The other part, evaluation research, involves determining whether and how the message worked. The amount of money used for the direct-subsidy aspect of the process—media buying—is huge. Industry consultants’ estimate upward of $250 billion as the global amount of money advertisers and their agents spend on placing ads on one or another medium.

The use of data to plan and evaluate these expenditures is by no means new. As far back as the 19th century, advertisers and their representatives in the nascent ad agency business bought, analyzed, and evaluated lists of individuals who might be influenced by particular direct-mail solicitations to determine whether and how the postal service was a good ad medium. In the early 20th century, advertisers worked with print media firms they subsidized—principally newspapers and magazines but also outdoor boards—to develop trusted total circulation figures based on audits. Somewhat later, they supported companies that used audience ratings panels to infer circulation data for radio and television broadcasters. By the 1960s, such circulation and ratings activities were yielding large streams of data that planners and buyers examined in advance of purchasing advertising space and time. To these numerical considerations were added the quantitative and qualitative results of depth interviews, surveys, and experiments by marketers, media firms, and advertising agencies to learn why certain ads in certain media succeeded and others did not.

From a colloquial standpoint, all these activities may well have been considered to involve big data. In response to a late 2012 trade article (Smith, 2012a) about a conference which was devoted to “Data-Driven Marketing,” a reader asked,

Whats all this latest fixation/obsession about data all about, as if we never knew it existed before, well it has, big time, and clever marketers have been using it well for years and don’t need to be reintroduced to it as if experienced marketers were schoolkids.

In fact, that article’s author (one of the conference organizers) had himself questioned the term’s use in the advertising context and had come to the more specific conclusion that-

Data has been “big” all along. What has changed now is not just scale and cross-channel inputs, but the sheer speed and accessibility of data as it moves to the cloud and becomes present on any device anywhere. Making data actionable in real time and at the point of critical need or decision-making is where data is not just big, but enormously effective. (Smith, 2012b)

In fitting this characterization, the ad industry does not merely mirror the fascination with data crunching taking place throughout society. The perspective reflects a transformation of media planning, buying, and evaluation in the advertising industry that began in the 1980s. The alterations were fundamental—institutional as well as technological. Before the 1980s, advertising practitioners considered media buying and planning as rather straightforward, unexciting components of a standard (“full-service”) agency’s offerings to clients. During the 1980s and 1990s, however, agency executives began to take a different approach to their media planning and buying divisions. Several factors were involved, but many of them centered on the fragmentation of media channels due to cable television (Turow, 1997). A clutch of new agency holding companies with international footprints (WPP of the United Kingdom, Omnicom and Interpublic from the United States, and Publicis from France) established freestanding media buying operations that, along with media buying firms Aegis in the United Kingdom and Havas in France, claimed special quantitative abilities. Using different computer models, each insisted it knew the best ways to reach increasingly dispersed audiences according to a growing number of demographic, psychographic, and geographic characteristics in the most efficient and accountable ways possible. According to a research firm that keeps track of buying firm developments, these six companies spent $224 billion advertising dollars worldwide in 2009 (RECMA, 2010). That year, the six controlled about 45% of purchasing in the U.S. advertising market; in most European countries, the share reached 80% (RECMA, 2010).

The buying firms’ emphasis on computer-driven quantitative analyses to target fragmented media audiences served as a testing ground for the coming age of ubiquitous digital media. Although advertising appeared during the 1980s on computer dial-up services such as Prodigy, the business was marginal and ad agencies did not consider that it had mainstream possibilities. The growth of commercial advertising on the World Wide Web with the introduction of the Netscape browser in 1994 pointed to a venue for marketers to reach millions of audience members. The second half of the 1990s marked a transition period during which publishers and various partners refined ways to construct the audience in greater detail than earlier decades.2 Central to their digital activities were technologies—cookies, tracking pixels, Flash cookies, and various mobile device “digital fingerprinting” methods—to trace people’s actions within and across websites, applications (apps), devices, and physical locations.

The ability to tag audience members and track what they viewed allowed publishers to create and offer up segments of inferred interests to advertisers who might conclude purchasing interests from that information. Advertising networks were doing the same thing, though across websites, and they and a growing number of data collection firms such as Axiom, Experian, BlueKai, and eXelate often matched their cookie-like trackers with those of other firms to enhance advertisers’ ability to target very specific types of individuals—and often even very specific (though still anonymous) individuals. By the late 2000s, audience data exchanges owned by Google, Yahoo, Microsoft, Interpublic, Facebook, and other major players facilitated the auction of individuals with particular characteristics, often in real time. It is, therefore, now possible to buy the right to deliver an ad with a message tailored to a person with a specific profile at the precise moment that that person loads a Web page. In fact, via an exchange, a publisher can sell an advertiser the ability to instantly reach and tailor a message for someone the advertiser knows from previous contacts and may even have followed around the Web.

This article is an excerpt taken from a technical paper, titled - "Advertising, Big Data, and the Clearance of the Public Realm: Marketers’ New Approaches to the Content Subsidy" published at International Journal of Communications under Creative Commons License.

Download The Paper - LINK

Cite This Article: 

Couldry, N., & Turow, J. (2014). Advertising, Big Data, and the Clearance of the Public Realm: Marketers’ New Approaches to the Content Subsidy. International Journal of Communication, 8 1710-1726. Retrieved from http://repository.upenn.edu/asc_papers/413

Copyright © 2014 (Nick Couldry & Joseph Turow). Licensed under the Creative Commons Attribution Noncommercial No Derivatives (by-nc-nd). Available at http://ijoc.org.