Abstract. Requirements engineering is still an area of software engineering in which theory and practice greatly differ. This work presents the results of an empirical study of artifacts created and used in the requirements engineering process. We discover that meeting notes and lists of requirements are most commonly used, that they usually play the role of information sources, and that specification documents are information sinks. Furthermore we show that most projects create several different artifacts. Finally we find out that despite the quality risks, inconsistencies between artifacts are often accepted.