This paper initially discusses some of the consequences which the technological development has for lexicography, especially in terms of the different types of empirical basis which can be used in dictionary projects. The most important advantages and disadvantages of using the Internet as a corpus are then listed and compared to the usefulness of “traditional” corpora. As an example, the paper shows how the Internet is used as the main empirical source in order to select lemmata and meaning items in the Online Dictionaries of Spanish Valladolid-Uva. The methods and tools employed in the project are discussed together with the requirements to the lexicographers’ competences, knowledge and skills. Finally, the paper provides some general conclusions as well as recommendations and hypotheses for future lexicographical work and research.