Journal of Marketing Analytics最新文献

Partial least squares structural equation modeling (PLS-SEM) is a highly popular multivariate data analysis method. The SmartPLS 3 software program helped many marketing researchers analyze the complex relationships between latent variables (i.e., mediation, moderation, etc.), which they measured by means of sets of observed variables. This program’s intuitive graphical user interface and various features, such as new metrics (e.g., HTMT, model fit indexes), advanced techniques (multigroup analysis, PLS_predict), and complementary techniques (e.g., confirmatory tetrad analysis, importance-performance map analysis), which impacted many business disciplines. SmartPLS 4 represents a significant leap forward in development with its completely revamped graphical user interface, faster processing speed for data estimation, and new model assessment features (i.e., cross-validated predictive ability test, endogeneity assessment, and a necessary condition analysis). This paper reviews SmartPLS 4 and discusses its various features, thereby providing researchers with concrete guidance that fits their analytical research goals.

Despite the extensive research on loyalty, academic discussions on simple and effective loyalty indices are surprisingly scarce. One of the few suggestions for loyalty indices is the net promoter score (NPS). Approximately two-thirds of Fortune 1000 companies currently use this index due to its overwhelming simplicity. However, scholars have criticized NPS for being inferior to other loyalty indices. Accordingly, scholars have mainly used Likert scales that cannot compare products/services using a single question and are subject to a response bias due to cultural differences. To fill this gap, we address the following four main criticisms of NPS: First, its advantages over the Likert scale of preferences and recommendation and purchase intentions are unproven; second, it cannot compare a competing product or service using a single question; third, its threshold setting is subjective; and fourth, it suffers from a cultural response bias tendency. This study asks the following research question: “Is brand selection superior to Likert scales in explaining sales performance?” To form a generalized conclusion, we examine 100 brands selected from 20 industries in the US market. The results of randomized controlled trials prove the superiority of selection over scale. Thus, brand selection should be emphasized to measure brand loyalty.

Customer churn can be defined as the phenomenon of customers who discontinue their relationship with a company. This problem is transversal to many industries, including the software industry. This study uses Machine Learning to build a predictive model to identify potential churners in a Portuguese software house. Six popular Machine Learning models: Random Forest, AdaBoost, Gradient Boosting Machine, Multilayer Perceptron Classifier, XGBoost, and Logistic Regression, were developed to assess which one would have a better performance. The experimental results show that boosting techniques such as XGBoost present the best predictive performance. The XGBoost model presents a Recall of 0.85 and a ROC AUC of 0.86. Additionally to the model performance, the study of the model's feature importance revealed that some factors, such as the time to solve a support ticket, the type of application, the license age, and the number of incidents, significantly influence customer churn. These insights can help the software industry key drivers of churn and prioritize retention efforts accordingly.

This research investigates consumer behavior when selecting snacks and beverages from vending machines in educational settings. The study aims to discern the impact of age on decision-making processes, perceptions, and memory factors related to product choices. Three experiments involved eye-tracking technology to monitor visual attention and a questionnaire to gather additional insights. The experiments analyzed the number of fixations, perceptions, overall product impressions, and memory influence. The sample comprised 90 adults aged 18 to 65 from the Pontificia Universidad Javeriana Cali. This sample includes a population distribution of 60% women and 40% men. Additionally, the socioeconomic distribution across economic levels was reported as follows: 40% falling within levels 1, 2, and 3, and 10% within levels 4, 5, and 6. The experiments revealed age-related differences in consumer behavior. Older individuals placed greater emphasis on healthier beverages, whereas young-adult individuals favored healthier snacks. Perception differences existed between age groups for beverages and snacks. Young-adult participants exhibited stronger positive perceptions of somewhat healthy options. Memory influenced choices, with young-adult participants relying more on memory cues. The findings provide valuable insights for tailoring marketing strategies and enhancing public policy initiatives to promote healthier vending machine choices.

The role of virtual influencers has risen in popularity. However, the revisions on this are scarce and incomplete. In this framework, two main objectives guide this work: (i) to carry on a performance analysis to measure the visibility/impact of the scientific production on virtual influencers’ marketing (most cited authors, journals, and themes) and (ii) to visually present the scientific structure by topics of research in virtual influencers marketing as well as its evolution along time. A final set of 1740 papers about virtual influencer marketing was retrieved from the ISIWeb of Science (from 1997 to 2021). Three different periods of time were identified (i) from 1997 to 2012 (649 papers), (ii) from 2013 to 2016 (540 papers), and (iii) from 2017 to 2021 (514 papers). Our results have identified seven promising future research lines on this topic, highlighting the role of tourism and sports celebrities.