Nonsense-mediated decay (NMD) in eukaryotic cells largely functions as a quality control mechanism by degrading faulty mRNAs that terminate translation prematurely. In recent years it has become evident that NMD also eliminates a subset of naturally occurring mRNA during proper gene expression. The mechanism of NMD in mammalian cells can be distinguished from the mechanism in, for example, Saccharomyces cerevisiae or Caenorhabditis elegans, by its apparent restriction to newly synthesized mRNA during a pioneer round of translation. This dependence can be explained by the need for at least one exon-exon junction complex (EJC) that is deposited on newly synthesized mRNA during the process of pre-mRNA splicing. Additionally, mammalian-cell NMD is promoted by the cap-binding protein heterodimer CBP80/20 that also typifies newly synthesized mRNA. When translation terminates sufficiently upstream of an EJC, the NMD factor Up-frameshift (Upf)1 is thought to join the stable EJC constituent NMD factors Upf2 and Upf3 or Upf3X (also called Upf3a or Upf3b, respectively), and undergo phosphorylation. Phosphorylation appears to trigger translational repression and mRNA decay. Although there are established rules for what generally defines an NMD target in mammalian cells, as with any rule there are exceptions and, thus, the need to experimentally verify individual mRNAs as bona fide targets of NMD. This chapter provides guidelines and protocols for how to define NMD targets using cultured mammalian cells.