As Customer Lifetime Value directly ties customer level data to business revenue, this metric has been of increasing importance in comparison to other traditional customer level data assessing loyalty, satisfaction, and more intangible valuations. Whereas invoice and sales data traditionally has been used for evaluating overall business performance – Sales Growth, Gross Profit Margin, Cash Flow Margins – utilizing this data for customer level analysis can provide significant insight into a tangible evaluation of customer contribution to these overall metrics. This study took this idea and expounded on existing research to create 18 customer level quantitative and qualitative attributes based on invoice data to predict the future CLV, as measured by Revenue, individual customers contribute to a business. By utilizing readily understandable multiple linear regression techniques and verifying that the model did not violate the theoretical assumptions for its appropriate utilization, a predictive CLV model was developed by incorporating these predictive variables to assist business leaders in predicting future CLV based on characteristics of its existing customer base. Then were performed three variable selection processes to create reduced models with fewer predictive variables to both avoid potentially inflated model performance measures inherent in complex models as well as help business leaders prioritize which customer level characteristics should be focused on for increasing a customer’s CLV.
An exmple image of work involves:
