What Did That $2.5 Million Ad Buy Us? Cognitive Science Goes to the Super Bowl
Year: 2007
Type: article
Abstract: What Did That $2.5 Million Ad Buy Us? Cognitive Science Goes to the Super Bowl Lisa A. Haverty ([email protected]) Arnold Worldwide, 101 Huntington Avenue Boston, MA 02199 USA Stephen B. Blessing ([email protected]) Department of Psychology, University of Tampa 401 W. Kennedy Blvd., Box Q Tampa, FL 33606 USA Abstract measures cannot even take the first step of determining whether an ad can induce recall of the brand featured in the ad. Even the most revered creative minds in the industry cannot identify a concrete “magic formula” for ad memorability. (Horovitz, 2006). The industry is therefore spending billions of dollars on a trial-and-error approach that does not offer a precise measurement of its return on investment. We propose the following argument. The purpose of advertisements is to sell products. Ads are therefore designed to at least raise awareness of a brand’s existence, and at best to be persuasive and even likable. However, awareness, persuasion, and likability will have no effect if the brand being advertised is not remembered as belonging to that ad. Indeed, any skillful persuasiveness could be inadvertently attributed to a competing brand. This would not have an advantageous impact on sales. Therefore, the effectiveness of any ad should be correlated with its brand memorability. Indeed, brand memorability should be a predictive factor of the economic success of any given ad. In this study, we endeavor to assess the brand memorability for specific ads and to thereby take the evaluation of television advertising into the realm of information processing. We employ a simple model of cognition and apply it to the most sensational event of the advertising year: the Super Bowl. In testing our model, we examine memory for ads a year after the game aired. To our knowledge, no empirical study of ad recall or recognition after such a long interval has ever been attempted. However, memory of the brand associated with “that funny ad” a year later should not, perhaps, be too much to ask for $2.5 million. Our CogScore model is being developed in a modern advertising agency by the first author. By way of giving context to the development of this model, some personal observations are of interest. First, people often assume that ad agencies have psychologists on staff. While this may have been true historically, it is not generally the case today. Advertisers certainly have expertise in influencing people, but they do so without scientific expertise on human behavior. Indeed, the industry appears to pay very little attention to the academic journals that are devoted to the scientific examination of their field. Advertisers spend a lot of money each year producing commercials. Yet they have done little to incorporate sound cognitive science into ensuring that their advertisements are effective, particularly in terms of memory for the desired brand. This research develops a metric, CogScore, that attempts to predict the memorability of the brand from any given ad. We show that our metric significantly predicts the memory for brand, much more so than a traditional metric on which advertisers currently rely. Keywords: information processing; memory; advertising effectiveness; gender differences; Super Bowl Introduction Advertisers spend billions of dollars every year in an attempt to persuade the public to consume countless products (Raine, 2005). Despite this demonstrated belief in the power of advertising, the corporate world is frustrated by a general inability to determine the return on that investment (Neff, 2006). An oft-quoted remark attributed to John Wanamaker, a 19 th -century businessman and a founding father of modern advertising, is still relevant today: “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” In fact, while today’s online and direct marketing methods (e.g., direct mail, infomercials) lend themselves to measurements of effectiveness and therefore to testing and analysis of effective and ineffective features, the vast majority of advertising can be assessed only at a very broad level or with intrinsically unreliable consumer self- reports. Broad-level measures include the noisy economic data of actual sales and ad-industry “tracking studies” that purport to measure “brand awareness” (which correlates nicely with “ad spend,” thereby encouraging further advertising investment), while consumer self-reports of “liking” and “purchase intent” are found via focus groups or online surveys. None of these measures can reveal the actual economic impact of a particular ad. Nor have these measures been shown to accurately predict how effective an ad will be before it is put into market and has already cost the company an enormous amount of money, both in development costs and in “media spend.” Indeed, these
Authors: Lisa A. Haverty, Stephen B. Blessing
Institution: University of Tampa
Language: English
Subfield: Marketing
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2767271495