Unformatted text preview:

MICHAEL L. RAY and ALAN G. SAWYER*Recent advertising media models demand estimates of effects of repeti-tive exposures an consumers in particular advertising situations. A laboratorytechnique for providing such estimates is suggested, and a study using this tech-nique indicates the need for significantly different repetition functions fardifferent kinds of products, brand positions, advertising formats, and ad-vertising goa Is. Further development of the technique is also indicated.Repetition in Media Models: A LaboratoryTechniqueThe problem of repetition in advertising is easy tostate but difficult to solve. Managers are never certainexactly how repetition operates in- any given advertisingsituation: for example, should an ad be run for a longtime without change, or will frequent changes beneeded? To slightly bend an old advertising statementoften attributed to John Wanamaker: “I know my adver-tising works with repetition. The trouble is I don’t knowwhen and how and why.” 1The problem, therefore, is that advertisers have hadto rely almost completely on judgment to decide howoften, how long, and in what media combinations to runan ad or campaign. Copy tests give some information onwhether one ad is “better” than another, but not on theeffectiveness of alternative advertising approaches re-peated over time. In fact, there is some suspicion that anad which does well in a single exposure C-opy test ‘doesnot do well with repetition [18, 31, 33, 35].Recent developments in advertising media modelsfurther highlight the need for better information on howparticular advertising will operate with repetition. Untilrecently, media mti-els have not had the capability toinput differential repetition functions for different ad-vertising situations which exist in a single run of the* Michael L. Ray is Assistant Professor of Marketing, Stan-ford University, and Alan G. Sawyer is Assistant Professor ofMarketing, State University of New York at Buffalo. The fieldwork and analysis were partially supported by a grant toStanford from the American Association of Advertising Agen-cies Educational Foundation. The mobile unit used in the projectwas provided by Foote, Cone & Belding. The authors wish toexpress their appreciation to William Massy, co-director of theAAAA project, and Gerald Eskin for their invaluable help indata analysis.model. Some early models [5, Chapter 5] did not eventake repetition into account; they assumed all exposureswere equivalent. Other early models used one simplefunction, such as Kotler’s [17], based on a general as-sumption, such as decreasing economies of scale.More recent models such as MEDIAC [21] orPOMSIS (1] have provided for different repetition func-tions for different market segments within the entiremedia schedule. With POMSIS, for instance, every mar-ket segment or every individual represented in the modelcan have a different function. Another model, FOCUS[34], gives nine different exponential curves to representthe estimated effect of each ad campaign over repeatedexposures within each market segment.Management science has clearly provided advertisingdecision makers with models potentially flexible enoughto take almost any estimated effect of repetition intoaccount. It is now time for behavioral science to providemeans of obtaining better information with which to es-timate repetition effect.This afiicle suggests an approach to provide this in-formation for advertising planners using media models.It discusses the kind of repetition information neededby advertising planners, the kind presently available, andresults and implications of some experimental researchon a laboratory technique designed to provide betterdata on repetition effect for specific advertising situations.PAST RESEARCH ON REPETITIONIdeally the advertising planner should have specificinformation about advertising repetition. Media models‘ The original Wanamaker statement was purported to be:“1 know half the money I spend on advertising is wasted, but Ican never find out which half” [26, p. 259].20472Journal of Marketing Research,Vol. VIII (February 1971), 20-9IIEconometric or Campaign Insertion StudiesThese studies are inadequate because they provide noreal information on the effects of repeated exposures toindividuals. Instead, these nonexperimental studies ex-amine the effects of increased advertising expenditures[36, 37, 40, 43] or increased numbers of insertions ofthe campaign in the media [24, 35] on advertising pay-out, usually in terms of sales or some kind of purchasingresponse. Both expenditure and insertion are poor in-dependent variables of repetition, because they are aslikely to increase reach as to increase frequency of ex-posure of a campaign; sometimes increased expendituredoes not even signal an increase in reach, as when adver-tising is moved to media requiring greater space andproduction costs. Since adequate data on gross reachREPETITION IN MEDIA MODELSallow specificity by advertising situation, and have begunto treat the problems of overlap and cumulative audi-ences [1, 2, 3, 7, 27]. Thus the planner now has a reason-able idea of the proportions of his audience that will beexposed to each media vehicle a given number of times.In other words, it is possible to think of the repetitionproblem in terms of the number of individual exposures,instead of the rather gross consideration of number ofdollars or insertions in various media.Keeping these developments in mind, it is easy to seethat the planner needs to know the level, shape, and slopeof the repetition function for various advertising situa-tions. The repetition function is here defined as the rela-tionship between exposures and effect, i.e., number ofadvertising exposures under natural conditions and effectin terms of advertising goals (awareness, comprehension,conviction, action, etc.). The advertising situation isdefined as all those variables that could influence repeti-tion effect: audience or segment; product or topic; mar-keting environment (including competitive effects); ad-vertising appeal, format, media, and schedule.Unfortunately, despite the great amount of researchon repetition in both the social sciences and marketing,research provides little more than general guidelines forthe advertising planner and virtually no data on the rep-etition function for specific advertising situations. Asseen in the table, each type of research fails to provideadequate repetition data in some important


View Full Document
Download Repetition in Media Models
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Repetition in Media Models and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Repetition in Media Models 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?