Exploring Repeat Exposure Effects of Internet Advertising


Janghyuk Lee and Donnel A. Briley (2005) ,"Exploring Repeat Exposure Effects of Internet Advertising", in AP - Asia Pacific Advances in Consumer Research Volume 6, eds. Yong-Uon Ha and Youjae Yi, Duluth, MN : Association for Consumer Research, Pages: 259-260.

Asia Pacific Advances in Consumer Research Volume 6, 2005      Pages 259-260


Janghyuk Lee, HEC School of Management, France

Donnel A. Briley, Hong Kong University of Science and Technology, Hong Kong


Repeated exposure is one of the key issues in advertising as most consumers have a chance to be exposed to an advertisement more than once but the advertisement rate is usually proportional to the exposure frequency. Therefore the magnitude and conditions of repeat exposure have been studied by numerous researchers in the field. In measuring the effectiveness of advertising, three variables are considered in principle: awareness, attitude, and selection along with the consumer choice process (Nedungadi, Mitchell, and Berger 1992). The wear out effect, decreasing effectiveness of repeated exposure, was reported by Grass and Wallace (1969) with data from field experiments in which they measured the consumer’s level of attention to the commercial based on CONPAAD (Conjugately Programmed Analysis of Advertising) procedure developed by Lindsley (1962). Apart from inverted 'U’ shaped negative repeated exposure impacts on brand attitude (Cacioppo and Petty 1979; Calder and Sternthal 1980), repeated advertising exposure has been shown to provide positive effects on the consumer choice process. It is reported to increase accessibility of the brand (Ray and Sawyer 1971; Sawyer and Ward 1979), information about the brand and brand attitudes (Berger and Mitchell 1989). Especially the positive effect of repeat exposure on brand awareness is reported by Nedungadi, Mitchell, and Berger (1992) in mixed choice condition where subjects are given brands to choose. Our repeated exposure probability function is based on above previous findings that explain positive but marginally decreasing effects of repeat exposure on the consumer’s level of brand awareness.


Data are from advertising post-test survey results conducted by a New York based marketing research company. This company measured the performance of advertising by surveying individuals who had not been exposed to the advertisement (control group) in addition to those who had been exposed. Visitors of a site where the target advertisement is posted are tracked by cookies. In this way, the company systematically measures various variables of interest including the number of exposures, then surveys these individuals by intercepting on the Web. Survey responses are collected from exposure and control groups to examine the performance of the concerned advertisement. In our data set, we focus on the brand awareness that is answered in Boolean ("Have you heard about brand X?" "Yes/No") and the number of exposures to the advertisement. Data for three advertising campaigns were selected, with 2,720 respondents (1,809 in the exposure condition). These three campaigns are considered successful as there was a significant difference of brand awareness level between control and exposure groups (difference for brand A is 18.4%, for brand B is 22.5%, and for brand C is 16.3%).


Due to the limited numbers of observations for repeat exposure and its variation across campaigns, we proceed to assess the impact of repeat exposure on advertising performance in a 2-level hierarchical model (Raudenbush and Bryk 2002). At first we analyze the impact at the individual level (level 1) and aggregate it at the campaign level (level 2).

In this model, the exposure frequency (EXPOFREQ) is designed at the key independent variable to explain its impact on "Brand awareness" (coded binary as P). As the awareness is a binary observation ("have heard of" or "not have heard of"), a Logit link is applied to estimate the coefficients of the model instead of the identity link.

The model is composed of 2 levels (See Table 1).

As the random effects are not significant enough to be integrated into the model, this brand awareness function can be composed of the intercept (-1.367662) and the positive effect slope of repeat exposure (.040338). The probability function of aware Brand X will be

Prob (Y=1/B) =                            1                               

                                              1+exp(1.367662 - 0.040338 x exprfreg

Graph 1 illustrates its functional form.


Our analysis holds some limitations. First, it may not be appropriate to generalize this functional form to all types of Internet brand performance measures. As this analysis is based only on a field post-test performance survey, additional results of similar performance survey measures will be necessary to confirm these findings. Secondly, our assumption of the Logit functional form needs to be tested. We adopt this Logit form as previous research results show only limited findings of repeat exposure of positive but marginally decreasing effect (Grass and Wallace 1969). In further research, this Logit form needs to be tested with more data points. Finally, the possible effects of other factors should be considered in future work. As we focus only on the repeat exposure effect on brand awareness, we do not incorporate possible factors that could influence this repeat exposure awareness function. To get a more precise functional form, additional covariates of both the campaign and individual levels needs to be integrated.





Grass, Robert C. and Wallace H. Wallace (1969), "Satiation Effects of TV Commercials," Journal Advertising Research, 9(3), 3-8.

Cacioppo, James T. and Richard E. Petty (1979), "Effects of Message Repetition and Position on Cognitive Response, Recall and Persuasion," Journal of Personality and Social Psychology, 37 (January), 97-109.

Ray M.L. and Alan G. Sawyer (1971), "Repetition in Media Models: A Laboratory Technique," Journal of Marketing Research, 8, 20-30.

Sawyer Alan G. and S. Ward (1979), "Carry-over Effects in Advertising," In J.N. Sneth (Ed.), Research in Marketing (Vol. 2, pp.259-314). Greenwich, CT: JAI Press.

Lindsley, Ogden (1962), "A Behavioral Measure of Television Viewing," Journal of Advertising Research, 2 (September), 2-12.

Calder Bobby J. and Brian Sternthal (1980), "Television Commercial Wearout: An Information Processing View," Journal of Marketing Research, 17(May), 173-86.

Nedungadi, Prakash, Andrew A. Mitchell, and Ida E. Berger (1992), "A Framework for Understanding the Effects of Advertising Exposure on Choice," In A.A. Mitchell (Ed.), Advertising Exposure, Memory and Choice (pp.89-116). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Berger, Ida E. and Alan A. Mitchell (1989), "The Effect of Advertising on Attitude Accessibility, Attitude Confidence, and the Attitude-Behavior Relationship," Journal of Consumer Research, 16, 269-279.



Janghyuk Lee, HEC School of Management, France
Donnel A. Briley, Hong Kong University of Science and Technology, Hong Kong


AP - Asia Pacific Advances in Consumer Research Volume 6 | 2005

Share Proceeding

Featured papers

See More


Losing Fast or Slow? Preferences for Uncertainty Resolution

Kurt P. Munz, New York University, USA
Alixandra Barasch, New York University, USA

Read More


The Subjective Experience of Goal Failure: How Choosing the Lesser Evil Eradicates the Negative Consequences of Goal Failure

Kamila Sobol, Concordia University, Canada

Read More


L7. The Joy of Shopping: Reconciling Mixed Effects of Positive Emotions on Shopping Behavior

Kelley Gullo, Duke University, USA
Duncan Simester, Massachusetts Institute of Technology, USA
Gavan Fitzsimons, Duke University, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.