Studying information diffusion through social networks has become an active research topic with important implications in viral marketing applications. One of the fundamental algorithmic problems related to viral marketing is the Influence Maximization (IM) problem: given an social network, which set of nodes should be considered by the viral marketer as the initial targets, in order to maximize the influence of the advertising message. In this work, we study the IM problem in an information-overloaded online social network. Information overload occurs when individuals receive more information than they can process, which can cause negative impacts on the overall marketing effectiveness. Many practical countermeasures have been proposed for alleviating the load of information on recipients. However, how these approaches can benefit viral marketers is not well understood. In our work, we have adapted the classic Information Cascade Model to incorporate information overload and study its countermeasures. Our results suggest that effective control of information overload has the potential to improve marketing effectiveness, but the targeting strategy should be re-designed in response to these countermeasures.