Companies regularly plan and implement marketing campaigns in order to increase their sales by communicating with the customers. Whether these marketing campaigns are effective and successful in persuading customers to engage in certain behaviors remain a question unanswered. Predicting customer behavior is not enough to drive and manage marketing actions optimally because it does not show the influence of the marketing action on the customers' future behavior. Prescriptive analysis can offer better ways to guide marketing strategies. Uplift modeling promises a clear opportunity to diminish costs compared to traditional predictive analytics by simply maximizing the impact for any treatment decision where the objective is to apply an influence. The purpose of this paper is to suggest an uplift modeling framework in cross-sell marketing campaign management in telecommunications sector. Dataset used in this study includes demographic and behavioral characteristics of 21.439 customers comprising December 2020-January 2021 period. Using alternative machine learning techniques, we segmented the business customers of a leading telecommunication company into 4 groups as: persuadables, sure things, lost causes, do not disturbs.