Tim Kastrup: Accounting, Analytics and Action: Insights from Financial Due Diligence

Abstract

This thesis explores the gap between the expectations toward and the actualities of data analytics (DA) adoption and use in accounting for decision making practice. Based on that, it outlines a practical pathway for effective DA adoption and use in this context.

To these ends, the thesis draws on empirical insights from two qualitative case studies of DA adoption and use in the financial due diligence (FDD) practices of two Big Four accounting firms in Sweden and invokes different theoretical lenses that assume and foreground different logics of action. This translates into four papers that shed light on different aspects of the expectations-actualities gap.

Paper I examines how accounting and commercial logic shape the use of new DA tools in FDD, from an institutional logics perspective. Paper II explores how practical and theoretical judgments are invoked in data-driven FDD, from a Deweyan inquiry perspective. Paper III investigates why accounting professionals continue to rely on spreadsheet software for much of their work, from a systems perspective. Finally, Paper IV analyzes how DA adoption is implicated in the re-formation of accounting, from a Millerian margins perspective.

Collectively, these papers show how DA-related action and inaction in FDD are driven by logics of appropriateness (LoAs) and logics of consequences (LoCs). Connected to that, Paper III furthermore shows how both logics can be considered concurrently in a systems approach.

The thesis’ primary theoretical contribution lies in developing LoC- and LoA-based explanations for DA-related action and inaction in accounting for decision making that – jointly – contribute to a better understanding of the expectations-actualities gap regarding DA adoption and use in accounting practice. These incorporate key links between accounting and commercial logic and DA conceptions and use, practical judgments regarding the feasibility and desirability of DA-related means and ends, suprasystem demands and supplies as well as trade-offs between hard benefits and soft costs, among other influences.

Based on these insights, the thesis outlines an “accounting-spirited” approach to DA adoption and use that is geared toward generating business value from DA without diluting one of accounting’s foremost selling points: actionability.