Protein kinases have become the second most exploited group of drug targets after G-protein-coupled receptors (GPCRs), accounting for 30% of drug discovery projects at many pharmaceutical companies with dozens of compounds in clinical development. Most early kinase inhibitors exhibited poor selectivity between kinases, and the trend in recent years has been toward ever more selective inhibitors in an attempt to minimize the risk of side effects. The risk with highly selective inhibitors is that their efficacy for treating complex diseases like cancer might be compromised by the redundancies in signaling pathways. The increasing interest in multitarget drug discovery (MTDD) stems from a belief that modulating more than one target can provide superior efficacy and safety profiles compared to single target drugs. Currently, there are two contrasting MTDD philosophies: the first involves combining agents that are selective for a single target to achieve an additive or synergistic effect, and the second involves discovering agents that are simultaneously capable of addressing two or more targets. Although this perspective focuses primarily on the latter, the advantages and disadvantages of both approaches will be highlighted. Very few drugs are truly selective for a single target, and in reality most biologically active small molecules have a degree of promiscuity by their very nature. Many clinically useful drugs are now known to have multiple activities, but most of these multitarget drugs (MTDs) were discovered serendipitously and their mechanisms of action were only established retrospectively. The deliberate and prospective design of ligands that act in a "selectively nonselective" manner on multiple targets of therapeutic interest is an emerging trend in drug discovery. Increasing numbers of these so-called designed multiple ligands (DMLs) are being reported in the medicinal chemistry literature, particularly for identifying multikinase inhibitors (MKIs) with specific multiple activity profiles for cancer treatment. Five years ago, there were few examples of DMLs for kinase targets, but interest in this area has grown explosively since then. Marketed MKI drugs vary in the number of kinases they inhibit, with some being highly promiscuous and others inhibiting only a few. These apparent differences in selectivity are partly influenced by the extent of selectivity screening, as some inhibitors appear more promiscuous simply because they have been more rigorously profiled. As the title indicates, the goal for medicinal chemists in the MKI field is to strike the right balance between nonselectivity (promiscuity) required for efficacy and selectivity required for safety. Currently, it is difficult to intentionally design a MKI with activity only at the desired kinases, but increasingly rational and elegant medicinal chemistry approaches are being used to solve this problem. This Perspective aims to capture the current state of the art and explore future challenges and strategies in this area, with the terminology used herein summarized in Figure 1 using known inhibitors.