Services like telecommunication, messaging, radio/television broadcast, and Web browsing are increasingly using the same Internet Protocol (IP)-based transport technology. Converged IP next-generation networks (NGNs) will enable easier blending and personalization of those services. For popular content, technologies like caching, broadcasting, and multicasting aim to increase network efficiency but also introduce extra complexity and bandwidth overhead. Additionally, the sheer amount of available content and the increasing flexibility for the end user to request content at any time from any place make it increasingly difficult to determine content popularity in order to predict the efficiency of using content distribution techniques. This paper describes a model for wireless networks to use network characteristics and a prediction of the popularity distribution to calculate the optimal combination of unicast and broadcast techniques when offering a number of media channels to the end user. Furthermore, an approach is described to optimize personalized content distribution efficiency in converged IP NGN networks utilizing efficiency calculations with this model. © 2008 Alcatel-Lucent.