When executives develop strategy or make decisions, they rarely reflect on the methods by which they zero in on their conclusions. Who has time to think about thinking when the pressure is on to get the thinking done fast and thought translated into action? The middle of a crisis is no time for pondering the subtleties of cognitive method. But when emergencies are not emerging and exigencies have exited (for the moment), executives can improve their decision making skills by understanding more about their own reasoning processes. Giovanni Gavetti and Jan W. Rivkin make this very point in the introduction to “How Strategists Really Think: Tapping the Power of Analogy,” in the April 2005 issue of Harvard Business Review.
Although both deductive reasoning and trial and error contribute to business decisions, most of the reasoning done by leaders makes use of analogy, according to Gavetti and Rivkin. (These are the only three options they acknowledge.) When confronted with an unfamiliar problem or opportunity, executives find it very natural to use analogical reasoning, most likely without being aware of doing so. They will simply think back to a specific prior context that they regard as essentially similar, then extract a lesson or principle from that situation which they can translate from its original setting and apply to the new one.
Based on research in psychology, cognitive science, and political science, Gavetti and Rivkin contend that analogical reasoning is a powerful method for generating breakthrough ideas. At the same time, it is extremely easy for reasoning to go seriously astray when informed by poorly considered analogies. Gavetti and Rivkin rightly contend that executives can do themselves a favor by understanding how analogical thinking works and why it is appropriate for a wide range of business situations, and how analogies typically misfire.
Click the View button above to download a three-page summary of the model (PDF format), complete with a ManyWorlds commentary that discusses the effectiveness of the model.