Evaluating education innovations rapidly with build-measure-learn: Applying lean startup to health professions education

Purpose: The Lean Startup approach allows innovators (including innovative educators) to rapidly identify and refine promising ideas into models that actually work. Our aim is to outline key principles of Lean Startup, apply these to health professions education, and illustrate these using personal experience. Methods and Results: All innovations are grounded in numerous assumptions; these assumptions should be explicitly identified, prioritized, and empirically tested (‘validated learning’). To identify and test assumptions, innovators need to get out of the office and interact with customers (learners, teachers, administrators, etc). Assumptions are tested using multiple quick cycles of Build (a ‘minimal viable product’ [MVP]), Measure (using metrics that relate meaningfully to the assumption), and Learn (interpret data and decide to persevere with further refinements, or pivot to a new direction). The MVP is a product version that allows testing of one or more key assumptions with the least effort. We describe a novel ‘Lean Education Canvas’ that synopsizes an innovation and its business model on one page, to help identify assumptions and monitor progress. We illustrate these principles using three cases from health professions education. Conclusions: Lean Startup has tremendous potential for rapid, robust innovation and evaluation in education. Practice points The Lean Startup approach allows innovators (including innovative educators) to rapidly identify and refine promising ideas into models that actually work. All innovations are grounded in numerous assumptions; these assumptions should be explicitly identified, prioritized, and empirically tested (‘validated learning’). To identify and test assumptions, innovators need to get out of the office and interact with customers (learners, teachers, administrators). Assumptions are tested using multiple fast, iterative cycles of Build (a minimal viable product), Measure (using metrics that relate meaningfully to the assumption), and Learn (interpret data and decide to persevere with further refinements, or pivot to a new direction). The ‘minimal viable product’ (MVP) is a version of the product that allows innovators to test one or more key assumptions with the least effort.