Schema mappings are high-level specifications that describe the relationship between two database schemas. Schema mappings have been explored in depth during the past decade and have turned out be the essential building blocks in data exchange and data integration. Since in real-life applications schema mappings can be quite complex, it is important to develop methods and tools for deriving and explaining schema mappings. A promising approach to this effect is to use ``good'' data examples that illustrate the schema mapping at hand.
In this talk, we will present an overview of a body of work on characterizing and deriving schema mappings via a finite set of data examples, including characterizations of schema mappings via a finite set of universal examples and the learnability of schema mappings from data examples. Along the way, we will encounter tight connections between unique characterizability of schema mappings and homomorphism dualities in constraint satisfaction.