Using constructed value of information to identify key uncertainties for a decision tree analysis in iterative structured decision making

Abstract Conservation decisions are sometimes delayed due to pervasive uncertainty and the perception that more information is needed before acting. However, postponing decisions can be costly, both financially due to extended research efforts and ecologically due to missed opportunities for timely intervention. In structured decision‐making (SDM), these are known as information problems, where the challenge is deciding whether to act amid uncertainty or wait to gather more data. Constructed value of information (CVOI) is an emerging decision‐analytic tool that helps quickly identify which uncertainties are most critical to resolve. We propose that applying CVOI to an information problem in an initial SDM rapid prototype round can help identify key uncertainties that can subsequently be used in a decision tree, simplifying the decision context and enabling tractable analysis. We illustrate this approach through a two‐round SDM workshop focused on conservation planning for the endangered Curiously Isolated Hairstreak butterfly. In the first round, the decision was diagnosed as a multi‐objective information problem and we used CVOI to identify key uncertainties to potentially prioritize for research. In the second round, those uncertainties informed a recharacterization of the problem as a risk‐based decision, which we evaluated using decision trees. This case study demonstrates how CVOI can reduce analysis paralysis and support timely, effective conservation action.