The success of a company is often influenced by the service and by a product they offer. If the supplied service or the offered product is a software system, a good performance will be essential to achieve desired goals such as high product sales. Slow applications and server responses due to performance issues, may cause a negative chain reaction. The amount of actual and potential users will probably decrease, and so does the users’ satisfaction and the number of product sales. Application Performance Management (APM) is necessary to avoid these cases. The usage of APM could help detecting eventual software problems and to remediate performance issues afterwards. Meanwhile, the usage of mobile devices, e.g., smartphones and tablets, for accessing enterprise systems is increasing in every application category. This expands the space where a potential software problem might be located in. Performance of mobile applications is more influenced by external circumstances, e.g., user location and access from bandwidth limited networks. APM tools not supporting mobile monitoring, are not able to recognize the mentioned performance issues. There are different implementation strategies for application monitoring agents such as call stack sampling and full source code instrumentation. The goal of this thesis is to research agent strategies for mobile devices, to develop an own version of each agent type, to analyze and evaluate the different agent approaches in combination of various mobile application types. The evaluation will be done with a series of experiments, by measuring the outcoming overhead of the developed agents, integrated into previously selected representative iOS open-source applications.