Customer Lifetime Value

Customer Lifetime Value (CLTV) prediction aims to estimate the total revenue a customer will generate over their relationship with a business, informing crucial marketing and resource allocation decisions. Recent research focuses on improving CLTV prediction accuracy by addressing the complexities of real-world data distributions, often employing advanced machine learning techniques such as hierarchical Bayesian models, deep neural networks (including transformer architectures), and ensemble methods like stacked regression. These advancements, along with novel approaches like multi-view frameworks and optimal distribution selection, are enhancing the precision and applicability of CLTV models across diverse industries, leading to more effective customer relationship management and improved business outcomes.

Papers