Mentoring Is Overrated. Try Tutoring InsteadMichael Schrage
Posted on Conversation Starter: September 21, 2009 5:32 PM
The idea that the best way to learn a subject is to teach it may be the bane of undergraduates left to the mercies of graduate teaching assistants, but it's remarkably true. In medical school, the cliché "See one; do one; teach one" has become a dominant pedagogical principle. In fact, George Bernard Shaw's notorious anti-educational quip gets flipped—instead of "Those who can, do; those who can't, teach," it's "Those who teach effectively learn how to do."
The power of this practice was recently reinforced at a statistical software customer conference I attended. A participant complained that one of the training sessions was really more of a "technical demo" than a class. The session leader was less a teacher or facilitator that a presenter. The collective frustration was palpable. This seminar's attendees could "see" what the presenter was doing and observe the outcomes but they simply couldn't "get" the underlying principles. You really couldn't divorce getting business value out of the software from understanding the core statistical techniques.
So what happened? Three participants—each from different companies—got together during the break to teach themselves (and each other) how to marry the software to the statistics. Intriguingly, this ad hoc group had synergistic skills: One knew the software but had a shaky understanding of the statistics; another understood stats but had only a casual acquaintance with the software; and the third had a problem he thought the software could solve. Fifteen minutes of explanatory give-and-take around the keyboard later, everyone had clearly "learned" more about their own skill and competence by attempting to "teach" their colleagues. The software jockey gained greater fluency with the package as he demo-ed how to integrate the problem with the statistics. The stats geek got a better sense of the math in the course of helping translate the problem into the software. The guy with the problem better understood its underlying challenges in the course of defining it for the statistician and the software.
Of course, they each came away with a better understanding of their colleagues' expertise too—a win/win/win. My opinion: None of these individuals could have succeeded on their own. Just as significantly, the challenge of "teaching" their particular expertise to their two other partners had really pushed their own understanding of their particular skill. I was impressed. I wasn't surprised.
Nobel laureate physicists such as Enrico Fermi and Leon Lederman took pride in teaching bright undergraduates because it forced them to keep in touch with the fundamentals of their field and express themselves simply and clearly. Teaching wasn't merely imparting knowledge—it was a learning experience. I see this all the time in software and finance: The "power user" isn't the individual who has spent the most time digging through and learning the intricacies of the code; it's the person who is teaching others how to use that software to solve unusual problems. Similarly, the "quant" designing a novel financial instrument typically discovers details, nuances and substantive insights in the course of "educating" colleagues about what makes that innovation special.
When I observe how communities of practice and expertise evolve in entrepreneurial firms or global enterprises, I'm struck by how often the designated "teachers" get so much more value from the experiences than the fledgling "learners." Indeed, what really creates critical mass and momentum is a surge in those small three-or-four person "study groups" where it's delightfully unclear whether the individual participants learn more by teaching or by collaborative learning. That's one reason I believe "mentoring" is overrated as a human capital investment. I suspect that there are CMOs and CFOs who would become far more expert—and effective—in their roles if they took the time to explicitly teach people core skills and competencies in their specialty. Better yet, the scalable impact would come when those "students," in turn, sought to reinforce their learning by teaching others. See one; do one; teach one.
It would be a wonderful—if appropriate—irony if the new paradigm for "executive education" emphasized that the best way for executives to learn well is to insist they teach well. When you look at what Jack Welch did with Crotonville, you can't help but wonder if the best way to have a "learning culture" is to invest in a "teaching culture."