When studying living organisms, scientists are faced with inherent biological variation which is distinct from random noise or measurement error and which is fundamental to the correct interpretation of experimental results. As a consequence, rigorous standardization of both the animals and the environment can lead to a low external validity of research outcomes. In this paper, Bernhard Voelkl and Hanno Würbel argue that acknowledging this issue by adopting a reaction norm approach to phenotypic variation can help seeing reproducibility of animal studies in a new light and can help gaining new insights into adaptive responses and gene-by-environment interactions. The authors show how dominating environmental effects can affect inference and effect size estimates of studies and how elimination of dominant factors through standardization affects the nature of the expected phenotype variation. Introducing ‘the reaction norm of small effects’ concept, the authors discuss the consequences for statistical analysis, including the random lab model and its implications for the reproducibility debate and, more generally, what it means for the interpretation of experimental results in biomedical research.