Internal auditing is increasingly challenged by the vast amounts of data generated by digital transformation. To address this, new techniques like outlier detection are being explored, which can identify irregularities without needing extensive domain knowledge. While many studies recognize outlier detection as a preliminary step, they emphasize the difficulty of converting detected outliers into actionable audit findings. This work investigates how outlier explanation and visualization can assist auditors in transforming potential findings into actual insights. It begins with an overview of current outlier explanation methods, gathers requirements from the internal auditing perspective, and develops new approaches to fill two significant gaps: support for mixed-type data and effective visualization. Following a quantitative evaluation, one approach is integrated into a prototype for qualitative assessment within the internal audit function of an international automotive manufacturer. Both evaluations indicate that the developed method enhances the application of outlier detection in internal auditing by providing clear explanations and visualizations, thereby helping auditors manage the increasing data volume and mitigate risks by revealing previously unnoticed issues.
