Stakeholder Perceptions of Data and Analytics based Auditing Techniques

Public accounting firms have invested significant resources to develop reliable substantive tests using large data sets and sophisticated algorithms ("data and analytics based procedures"). Developing reliable procedures, however, is just one hurdle firms must clear before leveraging such data sets and algorithms. In particular, firms will also need to convince audit stakeholders that relying on data and analytics based procedures will not only enhance audit efficiency, but also improve, or at least maintain, audit effectiveness. This study provides exploratory, experimental evidence to indicate how three key audit stakeholder groups-non-professional investors, peer reviewers, and jurors-perceive two prominent data and analytics based audit procedures (population testing and predictive modeling) relative to a more traditional substantive procedure (sample testing). Results suggest that key audit stakeholders are generally open to, and, in some cases, favorably disposed to the use of data and analytics based audit procedures. Nevertheless, participants also expressed some concerns about the appropriateness of relying on data and analytics based procedures, particularly predictive modeling, as primary sources of substantive evidence. This paper reports these and other key findings to develop an agenda for future research to help firms better understand and ultimately address stakeholder concerns.