Annual business reports – containing consolidated financial statements and management reports, as well as other information – are statutory instruments of financial accounting in Germany. They are an important source of information for business analysts. The information in management reports is mostly unstructured text, and is therefore complex for an algorithm to analyze. For some analysis questions related to economic research, we have identified techniques from speech technology that can effectively support the analysis. We have implemented these techniques in a prototype. It became clear that an approach based on semantic analysis and ontological information is useful for this purpose. Natural Language Processing (NLP) techniques are used to help building an ontology database.