The estimation of Customer Lifetime Value (CLV) has gained importance in understanding the relationship between companies and their customers. However, its implementation often encounters difficulties, ranging from solutions being highly specific to the context for which they are developed to the attributes used violating customer privacy. This work aims to build a generic solution for CLV prediction, using only attributes derived from the date and value of customer transactions with the company. The developed model was evaluated on five different datasets, demonstrating its applicability, in addition to being compared with five reference solutions from the literature. The results showed that it was possible to achieve a reduction of up to 34% in CLV calculation error in the scenarios.