This study leverages data and machine learning to optimize client prospecting in a large HR company in Norway, fictitiously named HR-Nor. In the increasingly competitive business landscape, data analytics and machine learning can enhance efficiency and decision-making in the prospecting process. The research aimed to (a) identify common attributes among high-demand prospective clients; (b) determine suitable machine learning algorithms for optimizing prospect lists; (c) assess these models and identify the most suitable model for HR-Nor; (d) propose recommendations for implementations of the study’s findings. 

All predictor variables were analyzed to learn about how each influence the chances for a successful client engagement. These variables included, among others, existing frame agreement, whether other units in the client hierarchy had purchased from another branch in HR-Nor, the industry in which the client operates, and the client’s size. 

Three machine learning models were evaluated for their ability to optimize HR-Nor’s prospect lists, including logistic regression, random forest and XGBoost. Each model estimated the probability of client visits being converted to sales.