Zero-defect manufacturing and flexibility in production lines is driven from accurate Digital Twins (DT) which monitor, understand, and predict the behavior of a manufacturing process under different conditions while also adapting to them by deciding the right course of action in time intervals relevant to the captured phenomenon. During the exploration of the alternative approaches for the development of process twins, significant efforts should be made for the selection of acquisition devices and signal-processing techniques to extract meaningful information from the studied process. As such, in Industry 4.0 era, machine tools are equipped with embedded sensors that give feedback related to the process efficiency and machine health, while additional sensors are installed to capture process-related phenomena, feeding simulation tools and decision-making algorithms. Although the maturity level of some process mechanisms facilitates the representation of the physical world with the aid of physics-based models, data-driven models are proposed for complex phenomena and non-mature processes. This paper introduces the components of Digital Twin and gives emphasis on the steps that are required to transform obtained data into meaningful information that will be used in a Digital Twin. The introduced steps are identified in a case study from the milling process.