Enzymes with high activity are readily produced through protein engineering, but intentionally and efficiently engineering enzymes for an expanded scope is a contemporary challenge. Measuring reaction outcomes on mixtures of substrates, called here SUbstrate Multiplexed Screening (SUMS), has long been used to rigorously quantitate enzyme specificity. Despite the potential utility of SUMS to guide engineering of promiscuous enzymes, this approach has not found widespread adoption in biocatalysis. Here, we develop principles of how to design robust SUMS methods that, rather than assess absolute specificity, use heuristic readouts of substrate promiscuity to identify hits for further investigation. This rich information enables engineering of activity for multiple substrates simultaneously and identifies enzyme variants with altered promiscuity, even when overall activity is lower. We demonstrate the effectiveness of SUMS by engineering two enzymes to produce pharmacologically active tryptamines from simple indole precursors in a biocatalytic cascade. These advances leverage common laboratory equipment and represent a highly accessible and customizable method for enzyme engineering.