Veranstaltungsprogramm der VHB Jahrestagung

Many phenomena in consumer psychology have recently been explained by processing fluency theory. The key tenet of the theory is that (marketing) stimuli that are perceptually or conceptually easy to process are liked better with important downstream consequences on relevant marketing outcomes. Recent research successfully identified stimulus characteristics that systematically increase/decrease the fluency of stimuli. Moreover, different marketing-relevant outcomes of experienced fluency have been reported in the literature such as liking, purchase behavior, perceived truth, and perceived trustworthiness. A key epistemic benefit of processing fluency theory is that it explains a wide variety of phenomena based on a parsimonious set of theoretical assumptions. This talk aims to (1) introduce recent advancements in processing fluency theory; (2) summarize exemplary applications of processing fluency theory to important marketing phenomena; (3) present the results of a comprehensive meta-analysis of processing fluency studies in major marketing journals; and (4) provide guidance and best practices how to conduct processing fluency studies in marketing research.

Visual product design plays a crucial role in driving consumer demand. Consequently, companies allocate a great deal of resources to generating and testing potential new designs. But while visual similarity to the current market has proven useful to understand market potential, might similarity to prior offerings also play a role? To begin to address this question, we develop a novel way to measure design similarity, employing a deep learning computer vision architecture trained on publicly available images. Using the car industry as a test case, we gather thousands of automotive design images spanning 120 years. We train a model to generate image embeddings representing prior design periods and use that to compute visual similarity scores. Above and beyond similarity to the current market, results suggest that similarity to past periods may also shape car sales. Rather than always being beneficial, however, results reveal that the relationship between similarity to past periods and sales may depend on the specific period examined. Overall, this work provides a novel approach to measure similarity, deepens the understanding around drivers of design’s impact, and provides managers with a useful tool to generate, select, and implement more successful designs.

Almost ironically, consumer behavior research often does not measure real behavior. In fact, many scholars rely on scale-based measures and run laboratory studies in artificial settings. These settings may cause biased results and limit the research questions that can be studied. Against this backdrop, scholars have developed online shop simulations where they observe hypothetical consumer behavior in a realistically designed online shopping environment. However, these simulations lack systematic validation. To address this lack of empirical evidence, using a self-programmed online shop simulation, the authors systematically validate the use of such simulations as an approach to collect in-depth consumer behavior data. They do so in three experimental studies (nStudy 1 = 336; nStudy 2 = 109; nStudy 3 = 444), one involving real purchases of customers of a fashion retailer. Additionally, they compare their online shop simulation to traditional settings with scale-based measures to show that this hypothetical simulation mimicking real online shops predicts incentive-compatible behavior better than other methods of eliciting consumer reactions. By validating their online shop simulation as an experimental research tool, results have implications for research on online shop simulations as well as experimental realism, and for consumer behavior research in general.