ABSTRACT The rapid expansion of e-commerce platforms has generated an enormous volume of user-written product reviews, creating a growing need for intelligent decision-support mechanisms that can interpret complex and unstructured feedback. Traditional recommendation approaches such as Support Vector Machines, Naïve Bayes classifiers, and Recurrent Neural Networks have been widely used for sentiment analysis and rating prediction, yet they continue to struggle with cold-start issues, limited interpretability, and an inability to effectively capture long-range dependencies in lengthy reviews. To address these limitations, this study introduces a hybrid Transformer–GRU (Gated Recurrent Unit) model designed to enhance the accuracy and robustness of product recommendations. The Transformer module leverages multi-head self-attention to extract deep contextual information from reviews, while the GRU component efficiently models sequential dependencies and evolving user behavior over time. Additionally, the integration of BERT embeddings, TF–IDF scores, and Word2Vec representations enriches semantic understanding and improves feature expressiveness. Experiments conducted on the Kaggle Women’s Clothing E-Commerce Reviews dataset demonstrate that the proposed model achieves superior performance, yielding a low MAE of 0.02 and RMSE of 0.07. These results highlight the effectiveness of the Transformer–GRU hybrid architecture in delivering accurate and personalized e-commerce recommendations.

ABSTRACT The rise of generative artificial intelligence (AI) represents a new marketing phenomenon, particularly in the context of phygital ecosystems, where customers experience value through the convergence of physical and digital worlds. Given the novelty of generative AI and phygital in marketing and the scarcity of academic literature, but the readily available thought leadership of practitioners, this article conducts a review of practice articles to explore the impact of generative AI on phygital customer experiences, thereby revealing both its bright and dark sides. To do so, the review adopts the experiential research methodology using experiential screening of practice articles, guided by the phygital research paradigm and the sensemaking approach of scanning, sensing, and substantiating. In doing so, the review identifies that the bright sides of deploying generative AI for phygital customer experiences include creativity and innovation through analysis of market trends and customer feedback, generation of insights that guide new solutions, product and service innovation, and enhanced creative outputs in phygital marketing; data security and ethical considerations supported by predictive analytics that enable proactive issue resolution and real-time insights in phygital settings; efficiency and productivity through automation of routine tasks, focus on complex phygital interactions, quicker response times, improved customer satisfaction, and higher productivity; and personalization and engagement through tailored marketing messages and product recommendations, individualized offers, and customized experiences across phygital touchpoints that enhance engagement and satisfaction. Whereas, the dark sides emerge when outputs are inaccurate or biased and lead to misinformed phygital marketing decisions that suppress creativity and innovation and expose brands to potential brand-reputation damage; when data security and ethical concerns around privacy, transparency, and fairness heighten vulnerability to breaches and cyberattacks; when integration with phygital infrastructures and interoperability across platforms are complex and consistent quality in AI-generated responses is hard to maintain, thereby eroding efficiency and productivity; and when personalization and engagement are perceived as impersonal and raise expectations, particularly for tasks that require a human touch, thereby reducing satisfaction. These bright and dark sides of generative AI for phygital customer experiences are also discussed using relevant theories, thereby providing a theoretical foundation to spur and support future research in this nascent yet promising area of marketing.

ABSTRACT This study examines the impact of cultural dimensions and consumer ethnocentrism on purchase intentions toward domestic brands in Latin America – a region characterized by high levels of migration and cultural hybridity. Grounded in Hofstede’s cultural framework and the Theory of Planned Behavior, data from 535 respondents across six countries were analyzed using structural equation modeling. Results show that consumer ethnocentrism is a strong and consistent predictor of purchase intention, while Hofstede’s dimensions did not have significant direct effects. These findings highlight the importance of contextualizing cultural values within dynamic social environments and underscore the role of attitudinal factors, such as perceived behavioral control, in shaping consumer behavior. The study offers a more comprehensive understanding of how macro-level culture and micro-level psychology intersect in shaping consumer behavior. Practical implications include designing segmentation strategies and marketing messages that resonate with local identities and generational traits across diverse Latin American markets.

ABSTRACT Customer segmentation is a vital for data-driven marketing, yet traditional RFM (Recency – Frequency – Monetary) analysis overlooks evolving behaviours. This study introduce a dynamic FPM (Frequency – P(alive) – Monetary) model that integrates RFM (Recency – Frequency – Monetary), BG/NBD (Beta Geometric/Negative Binomial Distribution), and Gamma-Gamma for predicting customer behaviour. Frequency represents cumulative transactions, P(alive) indicates the probability of a remaining active, and Monetary reflects average order value at t = t 0 and t = t 0 + Δt . Using online supermarket data, clustering methods, including K-Means, K-Medoids, and K-Means++, are employed for customer segmentation based on the dynamic FPM values. Results show the model enhances clustering quality, with K-Means performing best (Silhouette Index = 0.42; Davies-Bouldin Index = 0.8). Furthermore, supervised machine learning models are employed to predict the changes in purchase behaviour in each segmentation at t = t 0 + Δt with significant precision (~0.99). The findings provide actionable insights for churn detection, high-value customers identification, and more effective personalised marketing and retention strategies.

ABSTRACT This study investigates the influence of augmented reality (AR) on purchase intention and on the formation of consumer perception profiles within contemporary e-commerce environments, with a specific focus on Generation Z. Using a between-group experimental design, the research compares two shopping conditions: a traditional online experience without AR features and an immersive alternative enriched with AR technology. A total sample of 260 university students assessed six essential attributes of online shopping through a forced-ranking procedure and subsequently reported their purchase intentions. The findings reveal that exposure to AR significantly enhances consumers’ willingness to buy, while also fostering more differentiated and emotionally driven perception profiles. Conversely, the non-AR condition produces more homogeneous and predominantly rational profiles centred on functionality and practical evaluation. Cluster analysis allowed the identification of nine distinct consumer profiles: five emerging from the non-AR environment and four associated with the AR experience. Overall, the results highlight the strategic relevance of AR not only as a tool for boosting conversion rates but also as a mechanism for refining audience segmentation and tailoring personalised digital marketing initiatives directed at digital-native consumers.

ABSTRACT Previous research suggests that the limited mainstream adoption of sustainable consumption is mainly due to consumers’ struggle to perceive compelling economic value in sustainable products, which are often associated with fewer benefits and higher prices. While eco-labels are frequently present in sustainable offerings as a signal of environmental practices, their potential to influence consumers’ perceptions of economic value remains overlooked. Supported by Signaling Theory, the present research investigates whether eco-labels can generate positive perceptions regarding the functional and symbolic benefits of the product itself and, subsequently, impact purchase behavior. Two quantitative studies (Study 1: n = 195; Study 2: n = 392) were implemented. The findings reveal that the impact of eco-labels on purchase is solely influenced by the mediating effects of perceived symbolic benefits, rather than functional benefits. Consumers’ moral convictions regarding sustainable consumption, along with their familiarity and understanding of eco-labels, play a significant role in shaping their behavior. Further implications are drawn.

ABSTRACT Existing research often frames firms with a prospector orientation as integrating corporate social responsibility (CSR) to enhance reputation and competitiveness. However, this perspective typically relies on the resource-based view (RBV) and overlooks the institutional context that influences CSR. Addressing this gap, we examine how national philanthropic environment (NPE), a component of the normative institutional environment, moderates the relationship between prospector orientation and CSR. Drawing on institutional theory and RBV, we analyze cross-country, firm-level data from 262 companies across 10 countries. Using hierarchical linear modeling, we find that in countries with high NPE, prospector orientation is positively associated with CSR, whereas in low-NPE countries, the relationship is negative. These findings contribute to strategic marketing and CSR literature by demonstrating that the normative institutional context shapes when prospectors align strategies with CSR. The study also underscores the value of integrating institutional and resource-based perspectives when examining firm-level strategic decisions across national environments.

ABSTRACT This study aims to examine the relationship between store innovativeness and organic food purchase intention, with the mediating role of shopping experience and the moderating role of customer service. A survey method was employed to collect data from 1,051 consumers in a key emerging market in Asia. Results from structural equation modeling show that store innovativeness positively influences consumer intention to purchase organic food. Furthermore, in-store shopping experience mediates this positive association, and customer service moderates the relationship between store innovativeness and shopping experience. This study adds to the limited research into how retailers can contribute to promoting sustainable food consumption by investigating the impact of key in-store factors on organic food purchase intention. This study’s findings encourage retailers to focus on store innovativeness to enhance shopper experiences and organic food purchases.

ABSTRACT This study aims to evaluate the relationship between consumer innovativeness and the adoption of new luxury fashion products through the lens of personal values. We maily employs a quantitative dataset of 775 consumers recruited from the six largest cities in Vietnam, an emerging market in Asia. The conceptual and structural models were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings reveal a positive relationship between three personal values (self-direction, stimulation, ecocentrism) and consumer innate innovativeness. Furthermore, self-direction, achievement, and power are values that demonstrate a strong correlation with luxury fashion innovativeness. Self-direction, stimulation, achievement, and ecocentrism values were found to have a direct impact on new luxury fashion adoption. We also confirm that consumer innate innovativeness positively impacts luxury fashion innovativeness. Luxury fashion innovativeness directly influences new luxury fashion adoption, while consumer innate innovativeness indirectly influences new luxury fashion adoption through luxury fashion innovativeness.

ABSTRACT Customer segmentation underpins Customer Relationship Management (CRM) and growth, yet purchase patterns are often sparse and irregular. This study addresses this gap by integrating an extended RFMI framework (Recency, Frequency, Monetary, and Interpurchase) with the density-based HDBSCAN algorithm and applying it to an H&M transactions dataset. The approach detects non-spherical structure and permits a ‘noise’ label for irregular shoppers. This study derives RFMI features, standardises inputs, and estimates segments with HDBSCAN. The solution yields five segments: Low-Value Inactive, Low-Value Dormant, Mid-Value Occasional, Loyal Mid-tier, and Premium Champions, plus a small noise group (n = 21,011). Each group displayed unique recency, frequency, monetary value, and interpurchase profiles. Product analysis shows shared preferences for upper- and lower-body garments, with high-value customers engaging more broadly. Managerial implications include sharper retention allocation, targeted reactivation, and assortment/promotion design aligned to segmentation and price sensitivity. The RFMI+HDBSCAN pipeline offers a scalable alternative that improves segment fidelity in real-world retail data.