Recognition and recall tests, also known as readership studies, are most often associated with Starch INRA/Starch Readership Report and have long been viewed as a fundamental tool for assessing ad effectiveness in print media. They generally involve asking a representative sample of respondents whether they saw a given advertisement, read it and how much of it they remember.
The day-after test of TV commercials came to be associated with Burke Marketing Research Co., and by association came to be known as the DAR (day-after recall) test or the Burke DAR test. In the DAR test, respondents exposed to the commercial under scrutiny are interviewed the day after that spot airs. Both aided and unaided recall are used to measure advertising effectiveness.
Recall vs. recognition
There has been much debate over the use of recall vs. recognition to test ad effectiveness. Beginning in the early 1990s, researchers began to note that consumer recall of TV spots was declining and attributed that trend to the decreasing attention viewers give TV—not to smaller audiences, clutter or poor-quality programming. That does not mean, however, that researchers should forgo the use of recall measures, but rather they need to be aware that recall and recognition measures have different implications.
The advertising response model shows that in order to gain attention, an ad must break through the clutter that surrounds it. ARM uses multiple copy testing measures in order to identify both message- and execution-related variables, which are important in influencing persuasion. This method can help identify how an ad is processed by viewers and whether the variables were appropriate for the communication objectives set by the company.
The Advertising Research Foundation's Copy Research Validity Project assesses various copy-testing measures, including persuasion, salience, recall, communication, communication reaction (liking) and communication reaction (diagnostics). The ARF's study suggests that an ad's likeability is a valid measure of advertising effectiveness and that "liked" ads are apt to be informative as well as entertaining.
Likeability and humor
Liking an ad has important implications for memory. It has been suggested, in fact, that liking an ad may have less to do with entertainment and more to do with communication and persuasiveness. A 1992 study found that liking was highly correlated with persuasiveness. An earlier study found that people who liked a commercial "a lot" were twice as likely to be persuaded by it as people who simply felt neutral.
Humor has often been used to increase an ad's likeability, but that effect is not inevitable. Many consumers expect information from ads, and humor alone does not necessarily cause them to form favorable attitudes toward a product or service. On the other hand, people who want to think only as much as is necessary may look to likeability cues such as humor. The use of humor in advertising should, therefore, be regarded as a tool for segmenting potential purchasers.
Because of their social influence, celebrities are often used to aid recall and recognition of ads. Nevertheless, there is always risk involved when using celebrities to promote products and services. In the 1990s, for example, the negative publicity surrounding such figures as Michael Jackson and O.J. Simpson made marketers realize that celebrity endorsers could in fact have a deleterious impact on the public's attitudes toward the brand.
Developments in the 1990s
Toward the end of the 20th century there was an increase in the use of persuasion measures such as consumer purchase intentions and pre- and postpurchase attitudes as sources of both creative and diagnostic information.
The single-source study is another new type of research that emerged in the 1990s. Here, tracking studies conducted at regular intervals measure changes over time in ad awareness, recall, interest in and attitudes toward the ads and purchase intentions. One drawback of this kind of study, however, is that it fails to establish audience exposure to the advertisement.