Clandestine exploratory analyses are all too common, producing nominally confirmatory results through hidden analytical steps instead of harnessing exploration’s potential to modify and inform hypotheses, models, and theories. For such valuable exploration to evolve, however, methodological guidance, elaboration and implementation in the publication system is required. We present some basic conceptions to stimulate further development. In this first of two papers, we describe the current blending of confirmatory and exploratory research and how to separate the two via severe testing and the use of and compliance with an evidential norm. We consider bias and deal with the related Bayesian approach to severity. We discuss the role of pre-registration, outlining how conducting transparent exploration counteracts the pressure to produce seemingly confirmative findings and arguing that collaboration in confirmatory-exploratory research chains ("chained explorations") may already be beneficial in the current incentive system. The second paper will provide suggestions on how to do valuable exploration and how to implement exploratory research.