The world keeps changing more rapidly. Induced by context change disruptions such as individual life-course changes and macro socio-economical events, the way people carry out their everyday life doings is also undergoing a dynamic transition process, which may open up windows of opportunity for design to transit people's behavior in a more sustainable direction. A successful behavior transition entails not only changing people's wrongdoings but also retaining the existing desired doings. However, over the last decade, the field of Design for Sustainable Everyday Life seems to have grown accustomed to the concept of change. The potential role that design may play in retaining people's existing sustainable doings has been ill-addressed. This dissertation aims to develop an activity-based theoretical approach to help design researchers and practitioners better understand how people transit behavior when they undergo context change disruptions, and further explore design implications informed by the sustainable behavior retention perspective. The study comprises two parts. In the first part, six explorative case studies were used to investigate the applicability of adopting activity theory (AT) as a theoretical lens for understanding context change-induced behavior transition phenomena. As a result, an ATbased framework was iterated, developed and validated. In the second part, by incorporating the proposed framework with the theoretical understanding generated from a prescriptive meta-synthesis study, an AT-informed toolkit prototype was developed and evaluated. Three key findings can be identified. First, at a conceptual level, the study reveals that the design for sustainable behavior retention perspective may complement the design for behavior change perspective by facilitating a bottom-up and context-focused relative approach to achieve sustainability. Second, at a design analytical level, three dimensions of AT: i). hierarchical structure, ii). long-term development and iii). reality-based contextual scales of analysis are especially useful for systematically analyzing the impacts of context change disruptions on people's everyday life doings. Third, at a design synthesis level, the ATinformed design toolkit prototype and the extracted design implications can provide a systemic view that helps designers take both sustainable behavior change and retention perspectives into early-stage design ideation. The contribution of the dissertation is two folds. First, it introduces the perspective of sustainable behavior retention into the field of Design for Sustainable Everyday Life. Second, it provides an activity-based theoretical framework as a potential lens for designers to better cope with context change disruptions.