A Link Prediction-Based Method Towards Lead Management

Abstract

Lead management is an essential part of the customer acquisition and retention stages. However, as the number of leads increases, data-driven management automation is critical for better customer acquisition and retention. In this context, the present work proposes a method that supports lead management to identify and recommend to the sales team, future interests of leads that already exist in an organizations database in order to acquire or retain customers. To fulfill this objective, the network representation learning and link prediction models are explored. A case study is presented to demonstrate the effectiveness of the proposed method. All generated models reached a value between 0.873 and 0.998 of ROC-AUC. However, the prediction models showed low coefficient values, far from 1, the ideal value. Nevertheless, the method shows promise to be investigated in practice. For future work, a deep understanding of technical capabilities of network learning is suggested to obtain better results from link prediction models.