The Web of Data is fed mainly by “triplifiers (or RDFizers)”, tools able to transform content (usually from databases) to linked data. Current triplifiers implement diverse methods, and are usually based on bulk recipes, which make fixed assumptions on the domain semantics. They focus more on syntactic than on semantic transformation, and al- low for limited (sometimes no) customization of the process. We present Semion, a method and a tool for triplifying content sources that over- comes such limitations. It focuses on applying good practices of design, provides high customizability of the transformation process, and exploits OWL expressivity for describing the domain of interest.