Fast and efficient communication is crucial for workers that are required to collaborate. Instant messaging has been found to be more efficient than email and other asynchronous messaging systems when used for asking quick questions. Group chats have also been shown to stimulate collaboration between multiple users. In this thesis, we explore how mobile instant messaging can facilitate and stimulate collaboration between health data managers working in different facilities in Rwanda. An instant messaging application for Android has been developed and tested during this project. The application connects to the health management information system in Rwanda that is used by the data managers to collect and aggregate health data. The application also supports collaborative and social data analysis, by providing the users with a platform to share and discuss health data visualizations in groups. We have applied Social Presence Theory, Media Richness Theory and Social Data Analysis for the development of the Android application, as well as to analyze empirical data concerning the design and implementation of an instant messaging application for health data managers. The research method used in this project is called Action Design Research (ADR). Three cycles of testing have been conducted, both at the University of Oslo and in Rwanda. The empirical data collected during a month-long field test in Rwanda and subsequent data analysis has resulted in nine design principles for how to design an instant messaging application. To facilitate collaboration, we found low data consumption, loose coupling of the system and self-explanatory graphical user interface (GUI) to be important design principles. As well as having predefined HMIS users and private and self-monitored servers to run the systems on. Presence awareness, statistical integrations, flexible group compositions and attachment integration is found as the design principles that stimulate collaboration.