How effective is advertising in contemporary markets? What does advertising effectiveness mean? By the term “advertising effectiveness,” we mean what change advertising achieves in markets. Advertising also creates changes in awareness, attitudes, beliefs, and intentions. These are all valid effects of advertising. However, in the interest of focus and parsimony, this article focuses on the effects of advertising on only market behavior.
This topic has been the subject of research from the time firms began to advertise. Scientific research has begun to accumulate especially in the last 50 years (Tellis 2004; Tellis & Ambler 2007). This research falls within one of two paradigms: behavioral research and econometric research. Behavioral research uses theater or lab experiments to address the effects of advertising on awareness, attitudes, beliefs, and intentions. On the other hand, field research uses field experiments and econometric models to assess the effects of advertising on market behavior. This article will attempt to classify and summarize what has been learnt from the latter paradigm of research.
We use six terms quite generically here: product, firm, brand, consumer, market, and sales. The term “product” refers to any good, service, idea, or person being advertised. The term “firm” refers to any organization doing the advertising, whether it is a for-profit company, governmental agency, or nonprofit institution. The term “brand” refers to the names under which the firm advertises the product. The term “consumer” refers to the target of the advertising. The term “market” refers to the aggregate of consumers. The term “sales” refers to the aggregate behavioral response of the market of consumers, sought by an advertiser, such as unit sales, $ revenues, registrations, votes, subscriptions, etc.
We can classify field research into three groups: contemporaneous effects of ad intensity, dynamic effects of ad intensity, and effects of ad content. These subjects are the focus of the next three sections.
Contemporaneous Effects of Ad Intensity
We can further classify research in this area by which one of three aspects of advertising intensity the research addresses: weight, elasticity, or frequency.
Research on Advertising Weight
The term “weight” refers to the level or intensity of the advertising budget. Weight studies examine the effect of differences in ad budget across time periods or regions. The main focus of such studies is to determine whether an increase in budget translates into a proportional or profitable increase in sales of the advertised product. The same criterion holds for a decrease in advertising. Researchers have carried out over 450 market experiments to assess the effectiveness of advertising. Details of these studies are in Tellis (2004) or Tellis and Ambler (2007).
Research from weight studies leads to the following four important and surprising findings. First, changes in weight alone do not cause dramatic or substantive changes in sales. That is, a big increase in weight does not boost sales while a big cut in advertising does not decrease sales. Second, prolonged cessation of advertising sometimes leads to deleterious effects on sales. Third, if advertising is effective, its effects are visible early on in the life of a campaign. Conversely, if advertising is ineffective early on, then repetition will not create or enhance its effectiveness. Fourth, changes in media used, content of the ad, product advertised, target segments, or scheduling of ads are more likely to cause changes in sales than are changes in weight alone.
These results suggest three implications. First, firms could be over-advertising; as a result, cutbacks in advertising do not lead to a loss in sales. Second, advertising may have delayed or even permanent effects, so that continued advertising at the same level is not always necessary. Third, a firm’s budget increase or original budget itself may be more fruitfully employed in changes in media, content, target segments, product, or schedule rather than in weight alone.
Research on Advertising Elasticity
Advertising elasticity is the percentage change in sales for a 1 percent change in the level of advertising. It is free of any units. Studies in advertising elasticity focus on the rate and shape of the mathematical function by which sales respond to advertising. Researchers estimate this function via a branch of science called econometrics (Dekimpe & Hanssens
2007). Thus, research on advertising elasticity allows a more fine-grained analysis of advertising effectiveness than does research on advertising weight. There are over 400 estimates of advertising elasticity carried out in numerous studies across countries, product categories, and time periods. Meta-analyses of this literature summarize the mean advertising elasticity and explain its differences across various characteristics of firm, product, and market (Assmus et al. 1984; Sethuraman & Tellis 1991).
Research on advertising elasticity leads to the following important findings. First, across all studies, the mean estimated advertising elasticity is about 0.1, i.e., about one twentieth the corresponding price elasticity. Second, advertising elasticity is higher in early than late stages of the product life cycle, for food than other products, in Europe than the US, and for durables than non-durables, especially in comparison to the price elasticity. Third, advertising elasticities are lower in models that incorporate advertising carryover and exogenous variables than in those that do not, and in models that are multiplicative than in those that are additive. Fourth, however, advertising elasticities are invariant over the measure of the dependent variable, the measure of the product, or the method of estimation.
These results suggest the following four implications. First, price discounting may lead to greater increase in sales than does an advertising increase; however, whether that increase is profitable would depend on the level of price cut, which consumers use it, and how much of that gets to the ultimate consumer rather than being pocketed by distributors (Sethuraman & Tellis 1991). Second, advertising may be more profitable for new products, while price discounting may be more profitable for mature products. Third, advertising may be more profitable for durables, while price discounting for may be more profitable for non-durables. Fourth, advertisers in the US may be over-advertising or those in Europe may be under-advertising.
Research on Ad Frequency
A firm’s advertising budget normally impacts consumers through the exposure of consumers to ads through the media. The term “frequency” refers to the number of ad exposures each consumer receives in a particular time period. The advertising budget in a time period ultimately translates into a sequence of individual exposures targeted to one or more consumers. Similarly, sales may be considered an aggregate of consumers’ choices of brands, called brand choice. Research on ad frequency normally examines the effect of ad frequency on consumer choice. Such research provides a more fine-grained and insightful analysis of advertising response than does that on advertising elasticity or advertising weight. In particular, such research can indicate the optimal level of exposures the advertiser should target at a particular market segment or even a particular consumer. However, research on ad frequency is not without limitations. Knowing the optimal frequency does not immediately translate into ascertaining the optimal budget. The advertiser would still need models that relate advertising frequency to budget and consumer choices to sales. So the tradeoff here is between detail and insight versus managerial usefulness.
Research on advertising frequency leads to the following five findings. First, the effects of advertising exposure are less prominent and immediate and more fragile than those of price or promotion on brand choice. Second, in general, increasing frequency of exposures increases probability of brand choice at a decreasing rate. Third, for mature, frequently purchased products, the optimum level of exposure may be relatively small, ranging from one to three exposures a week. Fourth, brand loyalty may moderate response to ad exposures, in that established brands have an earlier and lower peak response to ad exposures than newer brands. Fifth, brand choice may be more responsive to the number of consumers the ad reaches than to the frequency with which it is repeated.
These findings suggest some implications. First, advertisers need to target loyal and non-buyers of their products with differing levels of exposures. Second, consistent with findings from prior sections, heavier exposures need to be reserved for new consumers and brands.
Dynamic Effects of Ad Intensity
Research on advertising effectiveness has focused on three dynamic aspects of advertising: carryover effects, wear-in, and wear-out.
Research on Advertising Carryover
The carryover effect of advertising is that effect it has on sales beyond the moment or time of exposure. Econometric studies have typically estimated the size and duration of the carryover effect of advertising. Here again there are a large number of studies, which have been analyzed in a couple of meta-analyses. This section summarizes findings from these meta-analyses as well as the primary studies themselves.
Research on advertising carryover leads to the following main findings. First, advertising typically has some carryover, so that all its effectiveness does not occur in the contemporaneous time period to its exposure to the consumer. Second, the estimated effect of advertising depends on the level of data aggregation. Estimated carryover effects tend to be longer with the use of more aggregate data. In general, the more disaggregate the time period of data, the less biased is the estimated effects of advertising carryover. In general, the effect of advertising may last for fairly short periods – hours, days, or weeks – rather than for long periods such as months or years. Third, advertising’s effects vary by region, city, and time of the day.
These findings suggest the following three implications. First, advertisers should neither assume their advertising has only contemporaneous effects nor assume that is has very long-term carryover. Second, advertisers need to evaluate the carryover effect of their advertising and do so with as temporally disaggregate data as they can find or collect. Third, advertisers need to analyze effects by region and time period rather than rely on simple generalizations.
Research on Wear-In and Wear-Out
A series of exposures of an ad is called an advertising campaign. Wear-in and wear-out are effects of an ad campaign. Wear-in is the effect of an ad that keeps increasing with repetition of the ad within the campaign. On the other hand, wear-out is the effect of an ad that keeps decreasing with repetition of the ad within a campaign. If they occur, wear-in normally does so early in the life of a campaign while wear-out normally does so late in the life of a campaign.
Research on wear-in and wear-out leads to the following important findings: wear-in either does not exist or occurs quite rapidly. It occurs more slowly when exposures are spread apart, when consumers are not forced to attend to the ad, for ads that contain emotional appeals rather than arguments, for consumers who are not highly motivated to attend to the ad or actively processing the ad content, and in markets relative to theater or lab settings. Wear-in also might be stronger with ads that have higher persuasion scores.
Ad campaigns wear out if run long enough. Wear-out occurs more slowly for ad content that is complex, emotional, or ambiguous, for ads that are less rather than more effective, for infrequently rather than frequently purchased products, for exposures spread apart rather than clustered together, for light rather than heavy viewers of TV, and for campaigns with increasing variety of ads or ad content. A break in a campaign may lead to an increase in effectiveness of the ad; if that happens, the ad wears out even faster than it did the first time around. In rare cases, possibly for new products, advertising seems to have permanent effects. That is, the effect of advertising persists even after the advertising is withdrawn.
These findings have several implications for advertisers. Most importantly, an ad which is ineffective early on should be discontinued. It is futile to assume that an initially ineffective ad still needs to wear in, because wear-in is rapid or nonexistent. Second, the findings suggest typical types of wear-in and wear-out that advertisers may expect for their campaign, depending on the characteristic of that campaign. However, whenever resources and time permit, advertisers should test their ads for wear-in and wear-out and accordingly decide on the duration of the ad campaign.
Effects of Ad Content
The term “content” refers to what is in the ad as opposed to its external characteristics, such as weight or frequency. Aspects of content include the appeal (argument, emotion, and endorsement), the duration or length, the use of color, sound, or video, the amount and type of text, etc. While a vast number of theater and lab studies have examined the effectiveness of various aspects of ad content, only a few market studies have done so. Thus, generalizations of findings in this area need to be made cautiously.
Research on ad content seems to suggest the following preliminary findings. First, changes in the creative, medium, target segment, or product itself sometimes lead to changes in sales, even though increases in the level of advertising by itself does not. Second, informative appeals may be more important early rather than late in the product’s life cycle. Conversely, emotional appeals may be more effective late rather than early in a product’s life cycle.
These findings have two important implications for advertisers. First, to increase effectiveness, advertisers should modify content more than increase weight or frequency. Second, advertisers need to test and typically vary the content of their advertising with the life stage of the product.
- Assmus, G., Farley, J. U., & Lehmann, D. R. (1984). How advertising affects sales: Meta-analysis of econometric results. Journal of Marketing Research, 21, 65 –74.
- Dekimpe, M., & Hanssens, D. M. (2007). Advertising response models. In G. J. Tellis & T. Ambler (eds.), The Sage handbook of advertising. London: Sage, pp. 247–263.
- Sethuraman, R., & Tellis, G. J. (1991). An analysis of the tradeoff between advertising and pricing. Journal of Marketing Research, 31(2), 160 –174.
- Tellis, G. J. (2004). Effective advertising: How, when, and why advertising works. Thousand Oaks, CA: Sage.
- Tellis, G. J., & Ambler, T. (eds.) (2007). The Sage handbook of advertising. London: Sage.