No abstract

This study investigates the effectiveness of customized promotions at three levels of granularity (mass market, segment specific, and individual specific) in online and offline stores. The authors conduct an empirical examination of the profit potential of these customized promotion programs with a joint model of purchase incidence, choice, and quantity and through optimization procedures for approximately 300 conditions. They find that (1) optimization procedures lead to substantial profit improvements over the current practice for all types of promotions (customized and undifferentiated); (2) loyalty promotions are more profitable in online stores than in offline stores, while the opposite holds for competitive promotions; (3) the incremental payoff of individual-level over segment- and mass market–level customized promotions is small in general, especially in offline stores; (4) for categories that are promotion sensitive, individual-level customized promotions can lead to a meaningful profit increase over segment- and mass market–level customized promotions in online stores; and (5) low redemption rates are a major impediment to the success of customized promotions in offline stores. Optimal undifferentiated promotions should be the primary promotion program in this channel, and firms can benefit from offering a combination of optimal undifferentiated and customized promotions for suitable categories in offline stores.

This study investigates consumers' attention to retail feature ads and proposes a method to optimize the design of the ads. Utilizing a large dataset of consumers' attention to over 1,100 individual feature ads collected with eye-tracking technology, we analyze the effects of the five key design elements of feature ads--brand, text, pictorial, price, and promotion--on consumers' attention to them. Attention is measured in terms of selection and gaze duration. We focus on the effects of the surface sizes of the design elements. A key feature of our model is that it takes into account the impact of visual clutter in the ad display page. To capture the clutter effects, we propose two new entropy-based measures that characterize the salience of feature ads in their competitive environment based on Attention Engagement Theory. In a Bayesian framework, we simultaneously estimate the parameters of the model and optimize the design of feature ads in terms of surface sizes of the five design elements. Our optimization results and comparisons with alternative design approaches indicate that significant improvements in attention to feature advertising can be achieved without increase in costs, and that the resultant optimal feature ad designs create win-win opportunities for manufacturers and retailers.retailing, promotions, visual attention, hierarchical Bayes, eye-tracking

Despite the growing literature on loyalty program (LP) research, many questions remain underexplored. Driven by advancements in information technology, marketing analytics, and consumer interface platforms (e.g., mobile devices), there have been many recent developments in LP practices around the world. They impose new challenges and create exciting opportunities for future LP research. The main objective of this paper is to identify missing links in the literature and to craft a future research agenda to advance LP research and practice. Our discussion focuses on three key areas: (1) LP designs, (2) Assessment of LP performance, and

Deep neural networks (DNNs) often suffer from "catastrophic forgetting" during incremental learning (IL) -an abrupt degradation of performance on the original set of classes when the training objective is adapted to a newly added set of classes. Existing IL approaches tend to produce a model that is biased towards either the old classes or new classes, unless with the help of exemplars of the old data. To address this issue, we propose a classincremental learning paradigm called Deep Model Consolidation (DMC), which works well even when the original training data is not available. The idea is to first train a separate model only for the new classes, and then combine the two individual models trained on data of two distinct set of classes (old classes and new classes) via a novel double distillation training objective. The two existing models are consolidated by exploiting publicly available unlabeled auxiliary data. This overcomes the potential difficulties due to unavailability of original training data. Compared to the state-of-the-art techniques, DMC demonstrates significantly better performance in image classification (CIFAR-100 and CUB-200) and object detection (PASCAL VOC 2007) in the single-headed IL setting.