The rapid progress of single-cell RNA-sequencing (scRNA-seq) at large scales has led to what seemed impossible until recently: the generation of comprehensive transcriptional maps of nearly all cells in multicellular tissues. We pinpoint three key elements as being critical to the production of these maps: scalability, spatial information, and accuracy of the transcriptome of the individual cells. Here, we discuss the ramifications of traditional cell-isolation protocols when capturing the transcriptional signature of cells as they exist in their native tissue context, the methods that have been developed to avoid these distortions, and the biological processes that have unraveled on account of these upgraded methodological approaches.