The jeopardy in Double Jeopardy

This post was written in 2009, but I believe that the issues discussed are still valid. The post predated the update of BrandDynamics and so refers to the old Pyramid model.

I have recently been asked to contribute to a point of view on the paper "Brand Advertising As Creative Publicity" by Andrew Ehrenberg, Neil Barnard, Rachel Kennedy and Helen Bloom (the article was originally published in 2002). The article considers the implications of Double Jeopardy for advertising. It reminds me of a debate held in the Journal of Advertising Research in 1996/7 and is another salutary reminder that some debates are never put to rest. For the record, here is my take on what we can learn from Double Jeopardy.

Double Jeopardy is an empirical generalization of a phenomenon that has been observed across a multitude of product and service categories: the greater a brand's penetration, the higher its repeat purchase rates. This means that most of any brand's market share is explained by its penetration, and relatively little is explained by its repeat rate. Double Jeopardy appears to have most sway in categories where the brands are readily substitutable, and differentiation is low.

The Double Jeopardy relationship also applies to attitudinal data. For instance, in Millward Brown's tracking and brand equity data, there is almost always a strong relationship between brand awareness and claimed trial or purchase intent, with better-known brands getting more than their fair share of recent usage and consideration. In order to predict actual market shares (as we do with our Consumer Value scores in BrandDynamics), stated purchase intentions must be adjusted to take account of the Double Jeopardy relationship. People who claim they will buy a small brand are less likely to fulfill their intentions than those that claim they will buy a big brand. When looking at image data, it is apparent that better-known brands attract higher levels of endorsement on positive attributes.

All well and good—Double Jeopardy is a fact of life. Big brands have a significant advantage over small ones. The key question is, what do you do with that information? According to Ehrenberg, Barnard, Kennedy and Bloom, the answer is to publicize your brand as widely and loudly as possible (irrespective of size) and not worry too much about what you say. By calling the brand to mind and refreshing the associated memory structures, "salience" will be created, and that will ensure that the brand is considered for purchase. The specific reasons for purchase, the authors suggest, are so specific to an individual consumer that they are not worth worrying about. Simply highlight the brand in a memorable way and people will buy it.

(Note: the definition of "salience" as used in the article combines awareness and memory traces, plus familiarity, plus assurance. This sounds very similar to Millward Brown's Bonding metric, except that we would insist that awareness, familiarity and assurance are created by memory traces, and therefore memory traces have a different status than those other attributes.)

The "advertising as publicity" model may seem bizarre to people brought up on a diet of "advertising as persuasion," but in fact there is a great deal of truth to this argument. For established brands, particularly in low-involvement and habitual-purchase categories, it is true that much of what advertising does is to reinforce existing brand memories and impressions. However, the argument breaks down when we consider the case of new brands, and brands that actually have a new and compelling message to convey. There is good evidence from a variety of sources that when people encounter information that is new and personally relevant, their perceptions of brands (and, by implication, their perceptions of competitive brands) can be changed. Therefore, advertising that conveys new news can have a substantial effect on sales. However, in most categories, the advertising that generates the strongest response combines these attributes: it is well-branded, highly memorable AND it evokes a conscious, positive response to the impression delivered by the ad.

Now, of course, that combination does not occur that often. As noted in the paper, "new news" is only newsworthy for a limited period of time. Thereafter, upon repeat viewing, an ad will remind people of why their opinion changed, but it won't create further change. The challenge for marketers, then, is to continue to present new innovations, or to make old news seem new again. In the absence of new and compelling news, advertising does become a battle for salience. Our research finds that over the long term, brands that have an effective share of voice greater than their market shares tend to grow.

So to my mind, "Brand Advertising As Creative Publicity" presents an important but one-sided view of how advertising works to generate sales. The analysis places too much emphasis on lessons learned from static analyses of steady-state markets and not enough on how brands grow by disrupting the existing status quo. As a result, the impact of delivering new, relevant, and believable impressions is undervalued. Just because brands rarely have compelling and relevant news to deliver does not mean companies should quit trying to find it. Quitting commits the brand to a long, slow war of attrition rather than a quick step change in market share.

So what are your thoughts on Double Jeopardy? Why does this relationship exist and what should marketers do about it?