Directed evolution has emerged as a general way to engineer essentially any catalytic property of enzymes, but due to the bottleneck imposed by the necessity to screen large libraries of mutants, it is often time-consuming. In order to make this type of protein engineering faster and more efficient than in the past, improved methods for probing protein sequence space need to be developed. This review focuses on recent advances which help to solve the traditional numbers problem in laboratory evolution, as in the directed evolution of enantioselective enzymes. Our contribution in this endeavour is iterative saturation mutagenesis (ISM), which can be used to enhance the enantioselectivity and/or the thermostability of enzymes. The option to use reduced amino acid alphabets as defined by the appropriate codon degeneracies supplements in a crucial way the toolbox in this knowledge-guided approach to laboratory evolution.