Denied by an (unexplainable) algorithm:teleological explanations for algorithmic decisions enhance customer satisfaction

Automated algorithmic decision-making has become commonplace, with firms implementing either rulebased or statistical models to determine whether to provide services to customers based on their past behaviors and characteristics. In response, policymakers are pressing firms to explain these algorithmic decisions. However, many of these algorithms are “unexplainable” because they are too complex for humans to understand. Moreover, legal or commercial considerations often preclude explaining algorithmic decision rules. We study consumer responses to goal-oriented, or “teleological,” explanations, which present the purpose or objective of the algorithm without revealing mechanism information that might help customers reverse (or prevent future) service denials. In a field experiment with a technology firm and in several online lab experiments, we demonstrate the effectiveness of teleological explanations and identify conditions when teleological and mechanistic explanations can be equally satisfying. Whereas the epistemic value of explanations is well established, we study how explanations mitigate the negative impact of service denials on customer satisfaction. Yet in situations where companies do not want to, or cannot, reveal the mechanism, we find that teleological explanations create equivalent value through the justifications they may offer. Our results thus show that firms may benefit by offering teleological explanations for unexplainable algorithm behavior.