A social network represents ties between people. Modeling social interactions to understand the underlying mechanism of herd behavior has been a topic of interest for the past several decades. Our focus in this work is to investigate the phenomenon of extreme, unexpected growth and gain in a social network, and how different diffusion models can be used to identify the "key players" in such a process. In social network analysis, it is a tedious task to find the most influential entities, and how ideas or news spread through the network. A popular method to solve this task is to represent the social network as a graph. In a dynamic or an ever-growing social graph, discovering when an idea starts spreading, and not when it started first, reflects the real nature of how viral ideas catch-on in society. Also, instead of locating the root of an idea, it is crucial to see who propelled the idea to the surface, and triggered the spread. In this paper, we develop a method to identify the set of nodes who are responsible for extraordinary change in graph growth. We propose a methodology that allows us to identify the phenomenon called "tipping point", both in terms of the the role of a node or a community in graph growth, but also with respect to the time frame it occurred. Our framework is employing different influence propagation approaches, to locate critical nodes or communities and indicate their role in the early stages of the sudden graph growth. We demonstrate how the methodology is applied using a real-life social network. We aspire that the proposed framework will help researchers better understand this phenomenon and be able to identify agents responsible for spreading trends, beliefs, opinions, etc. in society.