Some Interesting Findings about Super Bowl Advertising

While a uniquely American (U.S.) event, analyzing Super Bowl advertising can provide some lessons for advertising in other big events. Without doubt, the Super Bowl is the “marketing event of the year” in the U.S. characterized by heightened attention to ads. A now classic Nielsen study found that just over half of Super Bowl watchers in the U.S. report enjoying the ads more than the game itself (Nielsen 2010). It is also the case that the Super Bowl ads receive intense media coverage before, during, and after the game (e.g., Anderson 2016; Jones 2016). Teasers are put out for some ads, and most are pre-released to the public before the game. So coverage of which ads are the most liked and which generate the most buzz is intense. Moreover, industry analysts (e.g. Ace Metrix, Yougov), the popular press, and some academic institutions (e.g. Kellogg School at Northwestern University) weigh in on what they think the best ads were.

While there is a growing stream of research on the Super Bowl ads (e.g., Raithel, Taylor and Hock 2016), more research needs to be done to delve into the factors related to the success of ads in the Super Bowl and other big events such as the Olympics, FIFA World Cup Football/Soccer (including the highly viewed final match) and other events with very high viewership in an individual country or multiple countries. With the cost of a Super Bowl ad running at $5 million for a 30 second ad and today's organizations demanding that marketers provide evidence of return on investment (Taylor 2010) it is important that advertisers be knowledgeable about whether an ad they run can provide that return. Of course, ads can have different objectives. While web development site GoDaddy had great success raising awareness of their brand from under 10 percent of the U.S. population to approximately two-third in large part due to Super Bowl ads, they pulled out this year, most likely due a realization that this objective had been maximized and other objectives might not justify the high investment. Other companies may focus on improving their image. This can be product based or can focus on corporate social responsibility as did Colgate's ad that focused on water conservation. Still others may have the goal of increasing sales. Regardless, having some idea of what tends to work is very important.

The remainder of this editorial will focus on some interesting tidbits that have come up in some work I'm doing to lay the groundwork for a larger forthcoming study of Super Bowl advertising. The focus is on correlations of the effectiveness of various types of advertising executions and/or strategies for Super Bowl ads over the past five years, with comparison to what has been found in the academic research over a longer time span. Some of these preliminary results are surprising and warrant more investigation.

1) First time Super Bowl advertisers are having a very hard time being successful

2016 saw another large influx of first-time Super Bowl advertisers, including 13 advertisers who had never bought time in the game before (Poggi 2016a). Collectively, the 13 advertisers plus Death Wish Coffee, the winner of Inuit's contest for a small company, scored well below the average on likeability ratings as well as the Kellogg School's rankings. The Kellogg School ratings saw none of these new ads graded at A, 6 given a B, 3 a C and 5 a D. As the median ad ranking was a B, this is clearly not very good experience. Our research shows that this has been consistent over the past five years, suggestive of new entrants having a difficult time breaking through. More research is warranted on the factors underlying this phenomenon. One possibility is that too many of these advertisers follow what they view to be “tried and true” approaches that really are not that effective (see below).

2) Celebrity ads are far from a guarantee of success

Another characteristic of 2016's Super Bowl 50 was an increase in the number of ads featuring celebrities, which have even been referred to as the “Celeb Bowl” (Poggi 2016b). Consistent with academic studies dating some time back, including a celebrity in an ad is not necessarily a recipe for success. Tomkovic, Yelkur and Christians (2001), for example, found that celebrities did not enhance ad likeability. An examination of the past five years show that ads featuring celebrities actually slightly underperform an average Super Bowl ad both in likeability rankings and expert panel rankings of ad effectiveness. Interestingly, show business performers tend to outperform athletes as endorsers in recent Super Bowls. Of course this is not to say that an ad with a celebrity cannot be effective; rather the strategy needs to be thought out in a holistic manner while thinking about advertiser objectives, target audience, and message. The use of celebrities in Super Bowl ads is in need of further study, but clearly, new advertisers should not think that just throwing a celebrity into an ad will boost effectiveness.

3) Using humor also does not necessarily boost an ad's effectiveness

Over the past five years, a majority of Super Bowl ads have used humor. After a year (2015) characterized as the “Somber Bowl” by some, based on some high profile ads, including Nationwide Insurance's “Boy” ad focused on accidental childhood deaths, 2016 saw a return to humor (Kulp 2016). Based on history, this really should not be characterized as a long-term “return” to humor, as humorous ads have been more common than serious ads. While Tomkovic, Yelkur and Christians (2001) did find that Super Bowl ads of years past using humor got a slight likeability boost, Kim and Cheong (2011) found that serious or transformational ads significantly outperformed humorous ads in more recent years. Moreover, Siefert et al., (2009) found that emotionally engaging Super Bowl ads enhance social media activity and online buzz. Our analysis of the last five years of data show that humorous ads appear to be underperforming against serious ads, albeit by a small margin. Still, this does not mean humorous ads can't be slightly successful; rather, they should not be seen as part of a “formula” for Super Bowl success.

4) Ads with animals can work, but some seem to fare better than others

While not based on a large sample (a mean of about 8 or 9 per year over the last 5 years), our analyses are showing that an average Super Bowl ad featuring an animal as a central character performs average or slightly above average compared to other ads. This observation is consistent with earlier research that found that including an animal as a major character in ad tend to lead to above average likability (by Tomkovic, Yelkur and Christians (2001). Notably, however, our analyses document that ads featuring dogs seem to outperform those featuring other animals.

5) Sexual appeals have become relatively rare

Compared to the past, the number of Super Bowl ads using overt sexual appeals has been quite limited in recent years. While GoDaddy's run from 2005 to about 2013 shows that sex can help to raise awareness, this is likely a function of sexual appeals having limits in many product categories in terms of building brand image and/or boosting sales. The idea is that using a sexy model alone without reinforcing a unique element of the brand is unlikely to be effective.

6) We don't know a whole lot about the impact of pre-releasing ads on ultimate effectiveness of Super Bowl ads

For about the past five years, conventional wisdom for most firms seems to be that because of the huge dollar investment associated with running an ad, advertisers should pre-release the ad, or at least a teaser, to maximize exposure via social media. The idea is that pre-release may generate views, shares and tweets before the game in addition to the huge audience during the game. In addition, the ad may go viral after the game. While there is little doubt that some advertisers have capitalized on this idea, including Volkswagen's “The Force” ad featuring a boy in a Darth Vader costume, which has become the most shared ad of all time. Some earlier research documented advantages of early release, including Nail (2007), who found that it led to higher levels of postgame social media discussion.

While there are still some advantages of pre-release in terms of enabling exposures before, during, and after the game, with a majority of advertisers now allowing some level of pre-release, its effectiveness is coming into question. One aspect of this is that the element of surprise is lost and it is may be harder for a pre-released ad to stand out (Steinberg 2016). Moreover, with so many ads attempting to maximize exposure, some of them are bound to get “lost in the shuffle.” Clearly, the impact of pre-release is an area in need of more research.

7) Ads for iconic American brands using local consumer culture appeals rend to be effective

From ads depicting Super Bowl parties, to those showing parades, American athletic stars, cowboys, or Budweiser's Clydesdales, one strategy that seems to work more often than not in Super Bowl advertising is for large U.S. brands such as Coca-Cola, McDonald's Heinz, Pepsi, Chevrolet, and Budweiser invoking local “American culture” images in their advertising. Our research is showing that ads that invoke iconic American images with a focus on heritage tend to perform above the average.

One issue that is interesting here is whether global consumer culture appeals (see Taylor and Okazaki 2015) could be used effectively by large global advertisers. Moreover, it would be interesting whether such appeals to local consumer culture can work for smaller companies or so-called underdog brands (Jun et al., 2015). To date, such strategies have not been tried very often so it is difficult to know. That said, the Super Bowl at its core is a large “pop culture” American event featuring not only the penultimate game in the country's most popular sport, but also ads that the public watches and big event halftime shows with acts such as the Rolling Stones, Katy Perry, Beyoncé, and Coldplay.

Concluding Thoughts

So what does all of the above tell us? Aside from the possible exception of U.S. based brands with long histories in the country appealing to Americana, there does not seem to be a tried and true strategy for boosting the likelihood of running a successful Super Bowl ad. Clearly, more research is needed on the Super Bowl and other “big events” worldwide as they offer unique opportunities for high reach in today's very fragmented media environments.