Educational games have remained more of a promise than a reality in changing the way we learn. Occasionally, a few educational games have gained prominence. However, the "code" to designing educational games that consistently meet the desired goal of learning through fun is elusive. A key to unlocking the potential of educational games is the creation of designs that merge the act of learning and playing. It is best if the player achieves the goal of learning by merely playing the game. Games with an "endogenous" design have the potential to meet this objective. In endogenous design, game elements are derived from the educational content. However, creating endogenous design is challenging. The commonplace technique of superimposing known gameplay over educational content will not lead to endogenous design. Superficial integration of content and gameplay, often referred to as' chocolate-coated broccoli’, results in games that are neither educational nor fun. Designers can achieve better integration with appropriate guidance. Unfortunately, research in educational games has focused less on the synthesis of educational games, especially so on endogenous design. The need for endogenous design has been identified in the current literature, but research has not progressed toward building specific guidance on creating such designs. The aim of our research is to bridge this gap. We identify specific strategies that designers can use to design endogenous games. Our research is guided by a pragmatic approach that the outcomes should be of use in practice while also adding new knowledge to the domain of educational games. We chose Design-Based Research (DBR) as an overarching research approach because of its focus on practical utility. The DBR hints at mining knowledge from existing practises and improving upon them. Hence, the strategies for design are discovered through the study of design practice. By inviting participants to design endogenous games, we explicate the heuristics they may use in doing so. Using ‘think aloud’ protocol analysis as a technique for data collection and analysis, several raw strategies are identified. We evaluate, aggregate, compare, categorize, and extend these discovered strategies. The strategies are organised into a process framework using the "Function Behaviour Structure" approach. This framework, named "Endogen, is the core contribution of our research. Exploration and translation are the two primary groups of strategies in the framework that enable endogenous design. The exploration strategies help in identifying the gameable elements in the content, such as actors, objects, movements, events, situations, and more. The translation strategies help in translating the extracted elements into game elements such as mechanics, resources, and gameworld. These two are supported by other strategies for core design, elaboration, and verification, to make the design process complete. With a focus on practical use, our framework is validated for its utility using a multi-method approach. The validation exercise confirms the utility of the framework in enabling endogenous design. The studies also bring up a few shortcomings, some of which are addressed in this thesis. In the end, we enlist issues regarding the adoption of new frameworks and address them in the context of the Endogen framework. The Endogen framework will benefit educational game designers and academicians, as well as students, in designing effective and enjoyable learning games. Researchers can extend our work in the future by adding newer strategies, applying them across contexts and in more ways than we can yet imagine.