The objective of an audit of financial statements is to obtain reasonable assurance that the financial statements as a whole are free from material misstatements. Over decades, this audit conclusion has been achieved predominantly based on samples by applying the audit risk model. Over the last couple of years, the audit profession started to adopt audit procedures summarized under the concept “data analytics”. The novel aspect of data analytics is that in contrast to the traditional sample-based audit, the concept presumes an approximate full audit, or at least achieving the audit opinion based on the total population of transactions. However, the application of data analytics is limited to specific audit areas only. To a large extent, the audit of internal controls is performed manually by inquiries, observations and tests of samples. In the recent past, methods and techniques subsumed under the term “process mining” receive increasing attention by audit theory and audit practice. With process mining it should be possible to transfer the idea of data analytics to the audit of internal processes and controls. The technology uses data stored in the auditees’ IT systems to identify and analyze the entities’ accounting relevant processes and controls. However, the methodical integration and conformance with national and international auditing standards has been unclear as of now. Acceptance problems are also foreseeable with regard to the practical application of process mining in the audit, as the audit of internal controls is performed almost unchanged from a methodical perspective since decades. Regardless of that, process mining is continuously propagandized as an audit instrument especially by the Big Four audit firms. These aspects give rise to holistically explore the use of process mining as an audit instrument. The central research questions of this thesis are: (1) Which requirements have to be addressed by a data-based process analysis in an audit of financial statements? (2) Which findings and challenges can be identified based on the empirical evaluation of the implementation of process mining in the audit practice? (3) Provided the methodical requirements on the one hand and the practical challenges on the other hand, which modifications of process mining are necessary to address the key points of criticism?