Musings on Entrepreneurship and Innovation
Wednesday, September 28, 2005Failing to Learn from Experience
Why do apparently smart entrepreneurs who are good at making timely decisions in the face of ambiguity falter? I suspect that many stumble because they fail to learn from the feedback that results from their actions. But, why should they fail to learn? Well, consider the following:
When individuals have bounded powers to perceive or recall fine gradations, they tend to divide up actions into discrete choices, even when those actions have a continuous character...We think of...statements as "true" or "false"...Discreteness or finiteness can be viewed as a way of adding noise or distortion to past signals...when individuals see past signals only through a discrete filter--for example, whether an action was adopted or rejected--then learning is surprisingly imperfect and can quickly become completely blocked. [1]
This penchant to divide the world into two opposing forces--"right" versus "wrong," "good" versus "evil"--and to ignore or deny the existence of any middle ground, may be termed the two-valued orientation...Action resulting from two-valued orientation notoriously fail to achieve its objectives. [2]
In an entrepreneurial context, making a decision means investing resources--time, money, and expertise--in a course of action. There is inherent discreteness. To do something typically requires some minimum amount of energy. So, even if the policies that guide decisions allow for a multi-valued orientation, subtlety can get lost as the original assumptions get internalized. The result is ironic. That is, as early actions yield results and, consequently, better data, decision-making policies can become more clumsy. Just as the opportunity to learn increases, it's possible that the capability to learn diminishes.
If this really can occur, it's clearly not inevitable: the best entrepreneurial companies learn and adapt effectively. I suspect, however, that such learning is predicated upon a disciplined re-examination of assumptions in order to prevent "best guesses" from becoming anti-learning dogma.
[1] Sushil Bikhchandani, David Hirshleifer, and Ivo Welch, Learning from the Behavior of Others: Conformity, Fads, and Informational Cascades.
[2] S.I. Hayakawa and Alan R. Hayakawa, Language in Thought and Action.
More: Why Smart People Defend Bad Ideas
