Mobile applications are expected to be fast and responsive to user interaction, despite challenges mobile platforms and devices face in terms of limited computational power, battery, memory, etc. Ensuring that applications are performant is however not trivial, as performance bugs are difficult to detect, fix, and verify. In order for mobile applications and devices to appear perfectly responsive to the user, they need to meet a 60 frames per second frame rate, and keep load times preferably between 0-300 ms. Meeting these expectations means that there is no room for performance bugs, so there is a need for improving and developing better testing tools and strategies in order to help mobile developers improve performance in their applications. This thesis investigates strategies for testing and improving performance in Android applications by conducting a literary study, and a case study with the Spotify Android application. Some of the key findings of this thesis include promising results from tools that visualise sources of performance bugs in the user interface of applications, as well as proposed strategies and tools aimed to help developers profile and improve performance in their Android applications.