Indigenous knowledge (IK) is a key component of disaster risk management (DRM) and development planning, yet it is often overlooked, with practitioners preferring to use scientific knowledge. Critics of IK have termed it archaic, primitive, a constraint to development and inferior to scientific knowledge, which has contributed to its widespread marginalisation. However, smallholder farmers in rural Zimbabwe have utilised IK for generations, especially in predicting rainfall patterns and managing drought conditions, showing that IK can be a useful tool in DRM. This article presents findings from research on drought vulnerability and coping conducted in Zimbabwe’s Buhera and Chikomba districts in 2009, particularly relating to utilisation of IK in smallholder farming communities, and argues that unless IK is documented and preserved, its marginalisation will persist. The research followed a mixed-methods approach whereby both quantitative and qualitative data were collected and analysed. Whilst smallholder respondents were randomly selected for household surveys, snowball sampling was employed for key informant interviews. Respondents indicated that they utilised some indigenous rainfall pattern predictions gained from observing and interpreting plant and animal behaviour. Some cultural practices that were critical to development and utilisation of certain IK were also threatened with extinction. The article argues for ’marrying’ IK and scientific knowledge, in the hope that the two will offset each other’s weaknesses, resulting in some kind of hybrid knowledge that will be critical for promoting sustainable agricultural production in Zimbabwe. However, this is not for disregard the challenges associated with knowledge hybridisation, as these two types of knowledge are grounded on differing foundations.