Nature is providing a bountiful pool of valuable secondary metabolites, many of which possess therapeutic properties. However, the discovery of new bioactive secondary metabolites is slowing down, at a time when the rise of multidrug resistant pathogens and the realization of acute and long-term side effects of widely used drugs lead to an urgent need for new therapeutic agents. Approaches, such as synthetic biology, are promising to deliver a much-needed boost to secondary metabolite drug development through plug-and-play optimized hosts and refactoring novel or cryptic bacterial gene clusters. Here, we discuss this prospect focusing on one comprehensively studied class of clinically relevant bioactive molecules, the polyketides. Extensive efforts towards optimization and derivatisation of compounds via combinatorial biosynthesis and classical engineering have elucidated the modularity, flexibility and promiscuity of polyketide biosynthetic enzymes. Hence, a synthetic biology approach can build upon a solid basis of guidelines and principles, while providing a new perspective towards the discovery and generation of novel and new-to-nature compounds. We discuss the lessons learned from the classical engineering of polyketide synthases and indicate their importance when attempting to engineer biosynthetic pathways using synthetic biology approaches for the introduction of novelty and overexpression of products in a controllable manner. This article is protected by copyright. All rights reserved.