This paper discusses issues in relation to the creation of databases of figurative language expressions. More specifically, we describe the Hamburg Metaphor Database Project. It gives details on the procedure followed, problems encountered and insights gained while constructing this database. We also provide information on the theoretical framework and semantic–conceptual representation that give ground to the project. Corpus–based examples are included in the database. These and the annotating procedure is described and evaluated. Finally, we make an overview of other related projects that have attempted to compile databases on figurative language expressions.