Estimating marketing's power via Improved meta-analysis: A deep dive into advertising effectiveness
Marketing effectiveness is at the heart of all business success. Firms spend more on advertising than almost any other marketing activity. Research findings and insights regarding marketing effectiveness, including advertising effectiveness, are varied and conflicting. Meta-analysis has become an in...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universidad de Navarra |
| Repositorio: | Dadun. Depósito Académico Digital de la Universidad de Navarra |
| Idioma: | inglés |
| OAI Identifier: | oai:dadun.unav.edu:10171/70173 |
| Acceso en línea: | https://hdl.handle.net/10171/70173 |
| Access Level: | acceso abierto |
| Palabra clave: | Marketing Effectiveness Meta-Analysis Advertising Elasticities |
| Sumario: | Marketing effectiveness is at the heart of all business success. Firms spend more on advertising than almost any other marketing activity. Research findings and insights regarding marketing effectiveness, including advertising effectiveness, are varied and conflicting. Meta-analysis has become an increasingly popular and indispensable tool in marketing research aimed at discerning robust generalizations across studies in heavily researched topics. However, current meta-analytic methods recommended by and utilized in marketing lag behind those of other fields, leading to inaccurate effect estimates and incorrect inferences related to heterogeneity in effect: What meta-analytic methods are currently utilized in marketing? What methods should be used? What is the effect of superior meta-analytic methods on important topics related to marketing effectiveness such as advertising? What are typical advertising effect estimates and the relevant heterogeneity for both brand (primary demand) and category (secondary demand) advertising elasticities? What findings are suggested by prior meta-analytic and non-meta-analytic research in marketing effectiveness instruments and indicators—such as elasticities—and how can these indicators be used by practitioners? In this dissertation, I focus on understanding, distilling, and expanding collective knowledge in the area of marketing effectiveness through research on marketing effectiveness instruments and indicators as well as understanding and improving marketing meta-analytic estimation, in both cases predominantly concentrating on advertising as an outset. In the inaugural chapter, I meticulously dissect and outline the prevalent meta-analytic biases researchers should be cognizant of, compare recent historical meta-analytic methods and their application in marketing, and provide a comprehensive guide to robustly conducting meta-analyses in the field. In this research I (a) show that current meta-analytic methods in marketing are lagging behind other fields and that marketing still primarily relies on inferior methods and (b) provide a framework for future meta-analyses in marketing to follow. In the second chapter I pivot to a pragmatic application of the advocated methods detailed in the first chapter, concentrating on brand advertising effectiveness. This involves a thorough meta-analysis and review of antecedent studies, identifying significant reductions in effect due primarily to previously unaccounted for publication-selection bias. Through this research I find and correct for significant misspecification, aggregation, publication, and selection biases present in the advertising effectiveness literature and provide important updates regarding the size of brand advertising elasticities, resulting in estimates that are approximately 5-6 times smaller than those indicated by prior meta-analytic studies. In the third chapter, I carry out a more comprehensive advertising effectiveness meta-analysis by amalgamating brand and category elasticities in a more expansive, broader database. Through this research, I provide the first generalizable estimates of category advertising elasticities; show the usefulness of pooling primary and secondary advertising elasticities to expand heterogeneity analyses and allow for increased sophistication in terms of the number model conditions that can be incorporated; and estimate condition-specific effects for various brand and category advertising scenarios. Through supplementary chapters 4 through 5, I fortify the foundation by discussing price, advertising, and related elasticities as well as their uses and implications for practice from both a meta-analytic viewpoint and a general perspective. Taken together, the findings of my dissertation advance our understanding of marketing effectiveness and provide a path for improving meta-analytic estimation in marketing. In Chapter 1, I show that methods in marketing and advertising research lag behind other fields while drawing a clear path forward. In Chapter 2, I demonstrate that path in the context of a central topic in marketing effectiveness, brand advertising elasticities, while providing evidence of the presence of significant biases and overestimated brand effects in this literature stream, leading to updated brand advertising elasticity estimates. In Chapter 3, I estimate generalizable effects for primary demand advertising elasticities and showcase the benefits of and heterogeneity in a broad advertising meta-regression model that can be used to estimate both primary demand and secondary demand advertising effects under a variety of contexts and conditions. Finally, in Chapters 4 and 5, I distill prior findings in regard to the uses of advertising and other marketing instrument and indicator elasticities and detail how practitioners can utilize these indicators to improve marketing strategy. |
|---|