Unquestionably, advertising that successfully launches a new product may well produce results that last for decades. However, our research shows that, beginning with the first advertising execution that is run for a brand, each successive execution has less of a long-term impact than the one that came before it. Although situations vary, we have observed that advertising during the first one or two years of a brand's life creates its user base and establishes a level of usage among these individuals. Subsequent advertising helps maintain or marginally enhance this equilibrium, but it primarily produces short-term incremental sales results.
If an expenditure of advertising dollars is considered to be an investment in the future, one should at least have reason to believe there will be a noticeable difference in sales 12 months hence, depending on whether or not the expenditure is made today. In other words, holding everything else constant, the question is, "Will my sales be noticeably different beginning 12 months from now, depending on whether or not I expend a reasonable number of GRPs this month?"
The answer is that sales will very likely not be much different one year later. Advertising does have long-term consequences for a brand, but these consequences accumulate over the years as advertising is being run. However, each successive execution adds less and less to the cumulative total. One's current advertising by itself has a small impact on the future; it just adds another brick to the brand's foundation.
What is the "short term" for a brand, in terms of its current advertising? It's the category purchase cycle. Once a brand is established, its advertising succeeds primarily by temporarily altering how consumers make choices the next time they shop the category. As GRPs are expended behind a persuasive execution and consumers are effectively reached, some people who would have purchased a competitor instead purchase the brand with the persuasive execution. The short-term incremental impact of the advertising is completely exhausted at the point where every consumer who has been effectively reached subsequently makes a purchase in the category.
It is worth noting that in a competitive environment most of these results, although real, may be hidden from view as brands advertise and promote one another to a standstill. Most advertising dollars produce incremental sales by saving business that would have been lost to a competitor. The actual effect of virtually all advertising for in-line brands is defensive.
The important point from the perspective of this discussion is that the incremental impact an advertising execution has takes place within just one purchase cycle. Not the average purchase cycle, by the way, but the individual purchase cycles for each consumer who has been effectively reached.
How do we know this? First, our exhaustive analysis of scanner panel data proves it. Second, we have built a computer model that accurately forecasts the incremental results of advertising. The model employs ARS persuasion scores, which measure the brand-switching that results from exposure to a commercial in a lab setting, to forecast incremental sales as GRPs are expended in the real world. Both this model and another, which additionally accounts for the entire advertising/promotion environment, implicitly assume that the advertising impacts just one purchase cycle.
This second model accurately forecasts total sales, not just incremental results, based on the entire marketing mix. We have experimented with our models by changing their programming to assume that additional purchase cycles or fractions of a purchase cycle are impacted by current advertising. When we do this, the forecasts are consistently too high.
Our analysis clearly shows that there is no economic rationale for treating an established brand's advertising as a long-term investment for tax purposes. Very little of the impact of an in-line brand's current expenditure is felt long term and, to the extent it is, the impact varies from brand to brand. There is a rationale for treating new-product advertising as an investment (this includes advertising used to launch line extensions and product improvements), but doing this would mainly serve to punish product innovation.
In reality, the government will tax whatever it can get away with politically-and most people view advertising as a penalty they must pay to be entertained. A tax on advertising, be it direct or indirect, may have a political rationale, but it does not have an economic one.M
Mr. Stephan is president and Mr. Tannenholz is chairman of Princeton Brand Econometrics, Princeton, N.J.