Speed, search and the failure of simple contingency

It is widely accepted that an organizations internal design should be contingent on the nature of its external environment. Yet attempts to construct simple contingency relationship i.e., one-to-one mappings from environmental conditions to appropriate design elements have met with limited success. We shed light on this lack of success by means of an agent-based simulation in which modeled firms of different designs face various environmental conditions. We find robustly that turbulent environments call for organizational features that generate high speed of improvement, and complex environments call for features that engender diverse search. The precise design features that produce speedy improvement and diverse search, however, vary dramatically from one decision-making archetype to another. A feature that accelerates improvement in a decentralized firm, for instance, may slow it down in a hierarchical firm. It is this subtlety that undermines simple contingency relationships. We argue that the intermediate constructs speed of improvement and diversity of search clarify the mapping between environment and appropriate design and may point the way to more nuanced contingency hypotheses.