- Problem definition: A company’s project portfolio is an important success factor. Employing strategic buckets to segment the overall budget into budgets for different project types is a commonly used approach for managing project selection. Strategic buckets typically refer to sets of projects of a certain type, such as safe and risky projects. A strategic bucket specification defines the number of buckets and thresholds between them. This paper addresses the question of how different strategic buckets specifications affect decision makers’ project selection behavior. Methodology/results: We develop a behavioral model of the effect of strategic buckets on project selection and use laboratory experiments to analyze how bucket specifications affect project selection decisions. For various strategic bucket specifications where a rational decision maker would allocate the budget to projects of the project type matching their risk preference only, we find that actual decision makers haveProblem definition: A company’s project portfolio is an important success factor. Employing strategic buckets to segment the overall budget into budgets for different project types is a commonly used approach for managing project selection. Strategic buckets typically refer to sets of projects of a certain type, such as safe and risky projects. A strategic bucket specification defines the number of buckets and thresholds between them. This paper addresses the question of how different strategic buckets specifications affect decision makers’ project selection behavior. Methodology/results: We develop a behavioral model of the effect of strategic buckets on project selection and use laboratory experiments to analyze how bucket specifications affect project selection decisions. For various strategic bucket specifications where a rational decision maker would allocate the budget to projects of the project type matching their risk preference only, we find that actual decision makers have the tendency to allocate the budget evenly among buckets and among project types within buckets. This observation can be explained by the naïve diversification bias, and we observe this effect in experimental settings with different selection processes, project definitions, and subject pools. Managerial implications: Our findings allow companies to better understand the effect of buckets guidelines on actual project selection behavior and to manage their project portfolio selection by choosing the right bucket specification.…

