- 1 - Anatomy of a Newsvendor Decision: Observations from a Verbal Protocol Analysis Srinagesh Gavirneni Johnson Graduate School of Management Cornell University Ithaca, NY 14853 sg337@cornell.edu Alice M. Isen Johnson Graduate School of Management Cornell University Ithaca, NY 14853 ami4@cornell.edu Abstract We report on the information gathering and decision making efforts of subjects involved in newsvendor decision making. Previous research in this area identified significant biases (e.g. anchoring and insufficient adjustment) in inventory decisions, but was unable to identify specific thought processes that led to these biases. We overcome this by using a “think-aloud” approach and recording the thought processes underlying the subjects’ eventual decisions. A protocol analysis of the transcripts revealed the following interesting insights. Most of the decision makers sought the very basic information, but failed to seek the additional, but non-trivial, information that could have significantly influenced their decision. A majority of the subjects struggled to deal with the abstractness of the business setting and were very keen to know information on the product type, industry setting, decisions taken in the past, competitor’s situation, and vendor environment that they felt would have put them on a firmer footing. A large portion of the participants correctly identified the overage and underage costs that needed to be considered, but failed to convert that information into the optimal order quantity. This suggests that the mathematics involved is not as intuitive as perceived by the research community. Finally, the bias of the order quantity was significantly influenced by the specific type of risk (overage or underage) that was identified closer in time to the decision making point which indicates the presence of a recency effect. That is, if a subject first mentioned the risk of excess inventory followed by the risk of unsatisfied demand, then his/her order quantity was high. On the other hand, if a subject first mentioned the risk of unsatisfied demand followed by the risk of excess inventory, then his/her order quantity was low.