An impact evaluation of a pro-poor market system development programme, ‘Making Markets Work for the Poor’(M4P), poses several methodological issues for evaluators. M4P interventions intend to change the contextual conditions in which stakeholders take business decisions so that it triggers changeprocesses in the wider social system and ultimately benefits poorpeople. An impact evaluation design for such a programme thusneeds to be robust enough to adequately capture these systemicoutcomes, acknowledging dynamic changes in interventiondelivery as well as in market conditions over time. Theory-basedevaluation can provide learning and accountability when itincorporates methods that allow a critical reflection on the keycausal steps in an intervention’s theory of change. We presentour learnings about indicators and methods to reflect on theimportance of the contributions to market system change of a large M4P programme in Ethiopia.