An increase in transcriptional noise generally assumed to characterize aged cells and tissues is shown to derive instead from technical and biological issues that underlie single-cell RNA sequencing experiments.