Contrary to its typical presentation in scientific publications as a certain and linear process, in reality, the experimental method, not least the design aspect of it, requires a great deal of trial-and-error and ad hoc decision-making on the part of the researchers. This uncertain and contingent aspect of research, although little known outside of the circle of experts, has important implications for our understanding of the nature of science and scientific findings. This paper offers a backstage perspective to experiment design, where the uncertain and contingent nature of experimental research is at its starkest. It draws on insights from the sociological perspective of ethnomethodology through the auto-ethnographic first-hand experience of the first author's own social psychology experiments. Based on detailed lab notes and planning documents on how and why design changes were made, the analysis focuses on the evolution of the experiment design, particularly the researcher's "in situ" practical reasoning for how to make the experiment work. From this, we show how ethno-methods shape experiment results and highlight the inseparability of social science experimentation from "in situ" practical reasoning.