Customer lifetime value prediction and management in hotel organizations: An integrated approach with a web-based application

This thesis presents an innovative approach that combines customer lifetime value (CLV) and RFM model to analyze customer behavior in the hotel industry. The study uses machine learning techniques to predict CLV with behavioral and financial data and generate explainable solutions. Thus, by integrating reservation data and in-hotel extra spending, customer segmentation is realized and strategic recommendations are developed based on these segments. Comparative results are presented using XGBoost, LightGBM and CatBoost machine learning algorithms for CLV predictions. SHAP analysis is used for model explainability. Supervised and unsupervised (for RFM clusters) segmentation techniques are used. CatBoost (0.93) emerged as the best model in the study. The features that contributed the most to the model were variables such as Recency and Monetary as a result of SHAP analysis. For the visualization of the analyses, an interactive visualization infrastructure was created with Python Plotly and the application created was taken into reality in the web environment. The findings of the research provide important contributions to management decisions on identifying high-value customers, converting new customers into loyal customers and increasing in-hotel spending. In particular, the impact of factors such as Lead Time, Length of Stay and Season on customer behavior is examined and segment-specific strategies are proposed. In addition, the web-based application developed in the study contributes to sectoral innovations by enabling hotel managers to perform real-time analysis and make data-driven decisions. The results of this study have made significant contributions in both theoretical and applied fields and offer a new perspective on customer analysis and revenue management processes in the hospitality industry. Future research is recommended to integrate larger sample groups, different contexts and innovative technologies into tourism research.