Target Oriented Network Intelligence Collection (TONIC) is a problem which deals with acquiring maximum number of profiles in the online social network so as to maximize the information about a given target through these profiles. The acquired profiles, also known as leads in this paper, are expected to contain information which is relevant to the target profile.In the past, TONIC problem has been solved by modelling it as a Volatile Multi-arm bandit problem with stationary reward distribution. The problem with this approach is that the underlying reward distribution in case of TONIC changes with each exploration which needs to be incorporated for making future acquisitions. The paper shows a successful solution to the TONIC problem by modelling it as Volatile Bandit problem with non-stationary reward distributions. It illustrates a new approach and compares it’s performance with other algorithms. © 2020, Springer Nature Singapore Pte Ltd.