The ability to accurately predict and understand customer behavior is central to the success of modern Customer Data Platforms (CDPs), which rely on unified customer profiles to drive personalization, marketing strategies, and customer retention. Traditional segmentation models, such as Recency, Frequency, and Monetary (RFM), struggle to capture customer interactions’ dynamic, sequential, and interconnected nature in travel reservation environments. This paper introduces the Behavioral Pattern Analysis Framework (BPAF), a hybrid approach integrating Markov Chains and Graph Neural Networks (GNNs) for scalable, real-time customer behavior prediction. BPAF models customer interactions as a series of states (e.g., browsing destinations, searching for flights or accommodations, booking reservations) to track transitions through various customer journey stages. Markov Chains capture the sequential nature of customer actions, providing insights into the likelihood of moving from one state to another. Meanwhile, Graph Neural Networks model the complex relationships and dependencies between customer actions, allowing for a more comprehensive understanding of the interplay between behaviors across touchpoints. This combination enhances the ability to uncover advanced behavior patterns, such as delayed bookings, cross-category travel interests, and the influence of external factors (e.g., seasonality or promotions). Integrating Markov Chains and GNNs within BPAF ensures scalability by allowing the system to handle large datasets and model complex relationships efficiently. This hybrid framework can process vast amounts of data in real-time. It is particularly well-suited for dynamic environments like travel reservations, where customer behaviour evolves rapidly, and interactions span multiple channels. Experiments using real-world travel reservation datasets demonstrate that BPAF significantly outperforms traditional models in predicting customer actions, improving conversion rate predictions and enabling more accurate personalized recommendations. The paper concludes by discussing the potential of BPAF to drive more effective customer engagement, optimize booking strategies, and support growth through scalable, behaviour-driven insights.