Post conference course in Rimini – a report

Summary of the course Statistical methods for Evaluating the Effectiveness of Advertising, held after the ENBIS Conference in Rimini, 25-27 September 2002.

By Vesna Zabkar, participant of the course from Ljubljana, Slovenia

About 15 conference participants attended this course, organized by ENBIS and the Universita' di Bologna at Rimini. The course was held at Facolta' di Scienze Statistiche.

Day One

The first session was presented by Elisa Montaguti from Marketing, Department of Statistical Sciences, University of Bologna. The lecture was concentrated on communication process, on numerous models about how advertising works (market response models, economic theory, persuasive hierarchy models and low involvement models) and on advertising effectiveness measurement. We learned about two different types of measures; intermediate measures (recall, recognition, persuasion, liking) and measures of behavior (price elasticity, advertising elasticity, sales, market share). The conclusion of the session was that there is a need for more research on long-term effects of advertising both from practitioners and researchers, these is also a need for more validation across countries and about applicability of measures in different media.

The second session was given by Vittorio Bonori, Research Director at Optimedia. In well structured and vivid presentation we learned about the market role of a media agency, the media/communication strategy (from client's objectives to segmentation, media/communication mix, threshold settings, planning/buying to effectiveness measurement).We saw several examples of measurement of response in the advertising markets and several tools that media agency uses to assess results.

Day Two

Sergio Brasini from Department of Statistical Sciences, University of Bologna, started his lecture with reasons for measuring advertising effectiveness and major decisions in advertising (deciding on reach, frequency, impact, choosing among media types, selecting specific media vehicles and deciding on media scheduling). He presented Zielske's model, Broadbent's model and then concentrated on market response models in a simplified form as sales response models. A correct specification of the sales response function should take into account the existence of scale economies, the existence of the threshold level of advertising expenditures, the existence of a saturation level of advertising expenditures and the interaction between advertising and other marketing mix variables.

Giorgio Tassinari from Department of Economics, University of Bologna, presented sales and market share response models, from the shape of the response function, functional forms, models of competition (multiplicative competitive interaction model — MCI, and multinomial logit model — MNL) to parameters estimation. The models were presented through examples from the published research. The topic on advertising dynamics concentrated on the cumulative advertising effects and on advertising wear out.

Conclusion

The course was quite useful, however due to the fact that the topic of advertising effectiveness contains such a wide variety of issues, two days of the course could not be enough. Anyhow, it would be beneficial if future courses were not lectures but workshops with more active participation from the participants.