Impact of AI-driven Personalization on Customer Loyalty: A Case Study of Starbucks

Thesis Abstract

This study examines how AI-driven personalization strategies influence customer loyalty within the coffeehouse sector, focusing on Starbucks as a case study to illuminate the mechanisms by which algorithmic tailoring of offers, recommendations, and experiential cues impacts attitudinal and behavioural loyalty constructs. The problem addressed is the mixed evidence on the effectiveness of personalized AI interventions in building enduring customer relationships in high-frequency, service-led retail contexts, where loyalty is a composite of repeat purchase, brand advocacy, and hedonic satisfaction. The aim is to determine the extent to which AI-driven personalization enhances customer loyalty and to identify the mediating and moderating factors that shape this relationship. Specific objectives are (1) to assess the direct impact of personalization intensity on repeat purchase behavior; (2) to evaluate how perceived value, trust, and customer satisfaction mediate the personalization–loyalty link; (3) to examine the moderating effects of customer demographics, techno-literacy, and privacy concerns; and (4) to develop a parsimonious model illustrating the pathways from AI personalization to different dimensions of loyalty. The study adopts a mixed-methods design underpinned by the Expectancy-Value Theory and the Technology Acceptance Model to capture both perceptual and behavioural dimensions of loyalty. The population comprises Starbucks customers in the United States who have engaged with the mobile app or loyalty program within the last six months. A stratified random sample of 600 customers will be recruited to complete a structured survey, with 520 valid responses anticipated after screening for completeness. In-depth interviews will be conducted with 20 customers and 6 Starbucks digital product managers to triangulate survey findings and provide explanatory depth. Data collection instruments include a 7-point Likert-scale questionnaire measuring personalization perceivedness, perceived value, trust, satisfaction, and loyalty constructs (emotional attachment, repeat purchase intension, and advocacy). The interview guides will probe experiences with personalized recommendations, rewards, and communications, as well as privacy perceptions and app usability. Quantitative data will be analyzed using structural equation modeling (SEM) to test the hypothesized direct and indirect effects, with maximum likelihood estimation and bootstrapped confidence intervals for mediation analysis. Multi-group SEM will explore moderation by age, income, and tech-savviness. Complementary multiple regression analyses will examine robustness of the mediation paths. Qualitative data will be analyzed using thematic analysis, with coding conducted via NVivo to identify recurring patterns related to perceived personalization quality, trust formation, and loyalty dynamics. Integration of findings will follow a convergent parallel design, with results compared and synthesized to develop a coherent interpretation. Expected findings include (i) a positive and significant association between AI-driven personalization intensity and loyalty outcomes, mediated by perceived value, trust, and satisfaction; (ii) stronger effects among younger, more tech-savvy segments with higher privacy tolerance; (iii) evidence that personalization quality (relevance, timeliness, and non-intrusiveness) mediates the extent to which loyalty is strengthened, beyond mere frequency of interactions; and (iv) nuanced insights into how privacy concerns may dampen the personalization–loyalty relationship for certain customer cohorts. The study anticipates identifying thresholds of personalization beyond which perceived intrusiveness negates loyalty gains, and delineating optimal configurations of personalized offers and communications for Starbucks. Contribution to knowledge includes (a) advancing understanding of AI personalization in service marketing by integrating loyalty theory with digital personalization processes; (b) proposing a validated model linking personalization, perceived value, trust, satisfaction, and loyalty within a premium coffeehouse context; (c) providing managerial implications for tailoring personalization strategies that balance value creation with privacy considerations; and (d) offering generalizable insights for other high-frequency, experience-based retailers implementing AI-driven customization. The main conclusion is that well-designed AI-driven personalization can meaningfully strengthen customer loyalty at Starbucks when it enhances perceived value and trust without eroding privacy comfort, with moderation by customer characteristics. Recommendations include calibrating personalization intensity to segment-specific preferences, investing in transparent data governance and consent mechanisms, and iterating personalization algorithms to emphasize relevance and timeliness while reducing intrusiveness. Suggestions for future research encompass cross-market replication, longitudinal designs to capture loyalty evolution, and exploration of personalization’s impact on non-purchase behaviors such as brand advocacy and community engagement.