I'm working on a project in which we have to predict the Customer Lifetime Value (CLV) for a group of customers. In order to calculate CLV in a non-contractual setting, we use probabilistic BTYD models in combination with the gamma-gamma submodel. With only a transaction log we can estimate the future cash-flow of an individual. However, most often we need a transactionlog with at the very least one year of data.

However, CLV calculation might be even more useful if we can assign a CLV value to a customer before an individual has bought something. A company can then focus on acquiring the most valuable customers. Obviously, we don't have transaction records for these individuals. I was curious to know if someone knows how to calculate the CLV for such individuals? What type of data do we need? And what type of models can we use to predict "early" CLV for these kind of individuals?

Any tips, suggestions or resources are much appreciated!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.