Diagnosing Decision Quality: A Case Study in Hotel Revenue Management Decision Making

Human decision-making has often being criticized as boundedly rational, and subject to biases (e.g. Tversky and Kahneman 1974). Important information cues are often misweighted (Dawes 1979, Camerer & Johnson 1991) and feedback delays hamper learning (Sterman 1989). Providing useful decision support tools is also not a sure-win solution, as users often erroneously discount the value of the quantitative models they embody (Goodwin and Fildes 1999, Davern and Kauffman 1998, Davern 1998, Davis and Kotteman 1995, 1994). Experimentally information such as cue-criterion relationships has been shown to be of incremental value (Balzer et al. 1992, Balzer et al. 1989). In the field such information is only readily available in trivial settings. Generalizing from lab based approaches, we present a new field methodology for the “Diagnosis of Decision Quality” (DDQ) that aids decision makers in the discovery of such diagnostic information. We illustrate the application of our approach in the context of hotel revenue yield management using daily data obtained from an international upper-upscale hotel chain. We demonstrate the effectiveness of our methodology by identifying context specific systematic failures in decision making processes in a manner that facilitates the generation of adaptive changes (e.g., the introduction of specific decision aids) in the process to improve subsequent performance. More generally, we provide evidence of a decision making bias consistent with a failure to recognize “doing nothing” as a valid course of action. Significantly, this bias is consistent with the rhetoric of revenue management but is counter to the economic incentives of the revenue manager and hotel chain.