Resource title

Mastering unforeseeable uncertainty in startup companies: an empirical study

Resource image

image for OpenScout resource :: Mastering unforeseeable uncertainty in startup companies: an empirical study

Resource description

Novel startup companies, as well as new business development in established companies, are not only risky but also pose unforeseeable uncertainty. This refers to the inability to recognize and articulate relevant variables and their functional relationships. Established management methods exist to manage risk, or foreseeable uncertainty caused by known factors whose values cannot be predicted with certainty. However, these methods are insufficient for managing unforeseeable uncertainty. Two fundamental approaches have been identified that management teams can use to respond to major unforeseen influences. Trial and error learning means that the team starts moving toward one outcome (the best it can identify), but is prepared to repeatedly and fundamentally change both the outcome and the course of action as it proceeds, and as new information becomes available. Selectionism means conducting several parallel trials to see, ex post, what works best: one out of many trials is selected when more information has become available and the unforeseeable influences have played out. In this paper, we test predictions from theory under which circumstances selectionism or trial and error learning leads to higher performance. Based on a sample of 62 startup companies in Shanghai, we find evidence that traditional planning is most efficient when unforeseeable uncertainty threatening the venture is low. When uncertainty is high while complexity remains low, trial and error learning is preferred. When high uncertainty and complexity combine, selectionism offers the best returns provided that trial selection can be postponed until after the uncertainty is resolved.

Resource author

Resource publisher

Resource publish date

Resource language


Resource content type


Resource resource URL

Resource license

Copyright INSEAD. All rights reserved