The accurate quantification of cellular and tissue mRNA and microRNA content is reliant upon the selection of stable endogenous control transcripts for normalizing quantitative real-time-PCR (qRT-PCR) data. Using the combination of unbiased and informed approaches and a wide range of human adipose tissues and cells, we sought to identify invariant control transcripts for mRNA and microRNA. A total of 26 mRNA transcript candidates were selected from the literature. MicroRNA candidates were selected from a microRNA-microarray (Agilent, n = 22 tissues), and together with candidates from the literature resulted in 14 different microRNAs. The variability of these mRNA and microRNA transcripts were then tested in a large (n = 180) collection of a variety of human adipose tissues and cell samples. Phosphoglycerate kinase-1 (PGK1) and peptidylprolyl isomerase A (PPIA) were identified as the most stable mRNAs across all tissues and panels. MiR-103 was overall the most stable microRNA transcript across all biological backgrounds. Several proposed and commonly used normalization transcripts were found to be highly variable. We then tested the effect on expression of two established adipocyte-related transcripts (fatty acid binding protein 4 (FABP4) and microRNA-145 (miR-145)), either normalized to the optimal or a commonly used controls transcript. This test clearly indicated that spurious results could arise from using less stable control transcripts for mRNA and microRNA qRT-PCR.