Passenger Loyalty Analytics
Alsa, a leader in passenger transport, needed to go beyond the traditional measurement of its satisfaction KPIs, specifically the Customer Satisfaction Index (CSI/ISC) and Net Promoter Score (NPS). The main challenge was to scientifically identify which specific service factors (both linear and non-linear) truly influence customer loyalty in order to optimize them. The company required an understanding of how variables such as service type, demographics, or incidents affected user perception to make precise operational decisions.
WhiteBox implemented a comprehensive analytical solution that included the creation of a specific Datamart and the development of Machine Learning models (linear regression and decision trees) to capture complex relationships. Advanced model interpretation techniques (SHAP) and a deep Exploratory Data Analysis (EDA) were applied to isolate the exact weight of each variable. Additionally, users were segmented (by age, frequency, and travel reason) to personalize the analysis of satisfaction triggers.
The project provided Alsa with actionable and quantifiable insights, revealing that comfort (especially in third-party services) and delay management are the most critical factors for NPS. The solution equipped Alsa with robust models to predict future satisfaction and define marketing and operational strategies differentiated by segments (e.g., specific actions for young audiences or users of external services). This transformed complex data into a clear roadmap for continuous improvement.
Smart DUM Data Analytics
