Bayesian Causal Maps as Decision Aids in Venture Capital Decision Making: Methods and Applications † Short title: Bayesian Causal Maps for VCs

Improving venture capitalists’ decision processes is key to reducing failure rates for venture capital backed companies and to improving portfolio returns. In this paper we describe the use of a novel technique—Bayesian causal maps—to support and improve venture capital decision making. We combine causal mapping and Bayesian network techniques to construct a Bayesian causal map. The resulting probabilistic model represents salient features of decision makers’ mental models and inference processes. Heeding the call of prior research, we focus not on generating incremental descriptive insights into the actual decision processes of venture capitalists, but concentrate on creating a model designed to serve as a practical decision aid. The process of constructing Bayesian causal maps is presented using the real case of an experienced venture capitalist specializing in early stage, high technology investments. We then turn our attention from the construction of Bayesian causal maps to their application. Bayesian causal maps can support venture capital decision making through bias reduction, reduction of unsystematic error, assumption surfacing, what-if analyses, and by facilitating systematic learning from experience, both individual and collaborative. We discuss the advantages of Bayesian causal maps as well as the limitations and challenges inherent in their construction and use. Suggestions for future research are also offered.