Small-scale farmers play a vital role in providing food for a growing urbanized population and improving food security in Southern Africa. The smallholder farms are highly heterogeneous in terms of types of farming, levels of productivity and commercialization. These heterogeneous groups of smallholder faming systems require different forms of government interventions, depending on the objective and characteristics of each group. The aim of this paper is to analyze the typologies of small-scale farmers in South Africa based on a wide range of objective variables regarding their personal, farm and context characteristics, which support an effective, target-group-specific design and communication of policies. For this, a cluster analysis is conducted on the basis of a comprehensive survey among 212 small-scale farmers in the Limpopo region in 2019. An unsupervised machine learning approach with Partitioning Around Medoids (PAM) for the subsequent clustering is used. According to the results, the small-scale farmers can be grouped into four clusters. The largest cluster with 37.7% of the farmers represents the group of subsistence oriented farmers, while the smallest cluster with 14% of respondents indicates the emerging (commercial-oriented) farmers. The other two clusters are the semi-subsistence livestock farmers as well as the and crop oriented farmers that predominantly producing for own consumption and selling their surplus at their farm. According to the results, implications for target-group-specific policies are exemplary derived with regards to the topics of extension services, the adaptation of irrigation technologies and credit access.