Forecasting Customer Invoice Settlement with behavioural analytics

Empirical evidence across diverse sectors illustrates that organizations frequently encounter challenges with the collection of payments from customers. Research findings indicate that a significant proportion of invoices issued to small and medium-sized enterprises, as well as business-to- business entities, in the United States and United Kingdom are settled beyond their due dates. The primary objective of this study is to investigate customer payment behavior in relation to invoice payments and introduce an analytical framework for studying and projecting payment patterns. Our reasoning can subsequently be integrated into a decision support framework, enabling decision makers to formulate forecasts relating to future disbursements and to initiate appropriate measures aimed at recovering any outstanding liabilities, or to modify their financial strategies in accordance with projected cashflow figures. Our examination involves the utilization of a comprehensive dataset comprising over 1.6 million customers, encompassing their invoice details, payment history, and interactions, such as email communication, SMS, and phone calls, initiated by the company issuing the invoices in order to prompt payment. We employ both supervised and unsupervised learning methodologies to anticipate the likelihood of a customer settling their invoice or outstanding balance by the upcoming due date, drawing upon the interactions initiated by the company and the corresponding customer responses. We introduce an innovative behavioral scoring framework to be utilized as a predictive model input. The outcomes of logistic regression examined in the study demonstrate a maximum accuracy of 97%, irrespective of whether preclustering of customers was conducted, compared to the other two machine learning methodologies evaluated. This model exhibits considerable potential to assist decision-makers in formulating strategies that enhance the financial robustness of the organization through effective cash flow management and reduction of extraneous corporate credit lines.