Ideally, questions about comparative effectiveness or safety of interventions would be answered using an appropriately designed and conducted randomized experiment. When we cannot conduct a randomized trial, we analyze observational (big) data to obtain, hopefully, the same answers as the “ideal randomized trial”. The concept of emulating the ideal randomized trial – the target trial – aims and helps to avoid many biases occurring when naively using and analyzing observational data at hand when researchers deviate from simple principles that guide the design and analysis of randomized trials. Avoiding such biases is crucial. Examples will be presented how naïve approaches lead to wrong results and how the target trial approach helps to avoid these biases – again, in the absence of well designed and conducted randomized trials.