• 18 Apr 2000
  • Research & Ideas

Learning in Action

Most managers today understand the value of building a learning organization. But in moving from theory into practice, managers must realize there's no one-size-fits-all strategy applicable to every company and every situation. In this excerpt from his book Learning in Action: A Guide to Putting the Learning Organization to Work (HBS Press), HBS Professor David A. Garvin shows how different organizations put different learning strategies to work.
by David A. Garvin

"The most effective learning strategy depends on the situation," writes David A. Garvin. "There is no stock answer, nor is there a single best approach." In Learning in Action, he illustrated the diversity of learning organization strategies with examples from several organizations, including L.L. Bean, the U.S. Army's Center for Army Lessons Learned (CALL), AT&T's Bell Laboratories, the Timken Companies and General Electric's Change Acceleration Process (CAP).

L.L. Bean

L.L. Bean has long relied on inquiry as a source of learning to deepen its understanding of customer needs. Rather than collect impressions second- or third-hand, L.L. Bean goes to those in the know — experienced users who have lived with the product, often under demanding conditions, who then act as field testers. Testers are sent several samples, one from L.L. Bean and at least one from a competitor, to use for three months. During the test period, Bean encourages extensive feedback at all times, and formally solicits feedback at three points: when the product is first received, at the midpoint, and at the end of the test. The midpoint evaluations are especially revealing because of the creative approach the company employs: Bean typically structures these around an outdoor activity such as hiking or cross-country skiing. For example, when sales of the once best-selling Crest Hiker began to decline, L.L. Bean assembled groups of testers, product designers, marketers, and suppliers for a two-day hiking trip in Pinkham Notch, NH to test a new version of the boot. Participants were put in groups based on foot size, given two to three pairs of boots — the new Cresta Hiker, as well as the very best competitive offerings — and were instructed to hike, switching pairs every one and a half hours as they made their way up the mountain. After the two days testing, and constant interactive feedback, the design team returned to headquarters and immediately debriefed and extracted key learnings. Prototypes were quickly developed and sent to testers for their reactions. They approved, and the new Cresta Hiker was launched. The results were stunning. Sales rose more than 85 percent over the previous year and the initial shipment sold out within weeks.

Center for Army Lessons Learned

The U.S. Army's Center for Army Lessons Learned (CALL), based at Fort Leavenworth, Kansas, is a leader in participant observation. Founded in 1985, its initial role was to capture lessons from the National Training Centers where troops engage in long, simulated battles to test their readiness and skills. As the Army's mission broadened to include 'operations other than war' such as fire fighting, flood control, and other forms of disaster relief, CALL was charged with learning from these experiences as well. Today, CALL observation teams are among the first troops on the ground in any Army operation. They collect on-the-spot information about new practices and techniques, identify problems and trouble spots, distinguish approaches that work from those that do not, and share their findings throughout the organization. The learning that CALL collects forms a vast repository of knowledge — the Army's institutional memory — which can be drawn on as needed. This collection and dissemination of lessons learned ensures that the same mistake is never made twice.

Bell Labs

AT&T's Bell Laboratories conducted reviews of its own software engineers to determine why some were more productive than others with the goal of distilling their techniques into best practices that their less successful counterparts could adopt. After the stars were identified, they were interviewed in-depth about how they went about their work and, specifically, what they did to be more productive. Using these insights, the Bell Labs team designed an innovative hands-on training program, which the star performers then delivered to a subset of their peers. The results were immediate and impressive. Participants reported a quick 10 percent productivity improvement, rising to 25 percent a year later. There were equally striking gains in managers' evaluations of their ability to spot problems, conduct high quality work, keep their bosses informed, work across organizational boundaries, and attend to customer and competitive needs.

Timken Company

In 1988, the Timken Company, America's leading manufacturer of tapered roller bearings, embarked on a bold project to revolutionize the production process for manufacturing customized bearings that would reduce the turnaround time of their small, tailor-made orders from eight to ten months to weeks or days. Because they were breaking new ground, not only in the bearing industry, but also in manufacturing, the challenge clearly demanded exploratory learning. A design team was assigned responsibility for finding the equipment layout that would optimize product flow and plant capacity — critical to reducing process time — and made extensive use of simulations to generate insights. They created Cardboard City, a three dimensional model of a manufacturing cell built out of cardboard and two-by-fours and included full-scale mock-ups of each piece of equipment placed on casters so that they could be moved easily. The learning process included elements of both competition and cooperation to ensure that Cardboard City produced a wide range of possible layouts and simulated a wide variety of product flows. Three teams were formed and each was given approximately one hour to arrange the "equipment." Their challenge was to create the best material flows and best interaction among operators while using the minimum amount of space. Once built, operators then simulated the production process in each layout. None of the layouts emerged as obviously superior, so a fourth team was formed to draw from the best elements of the three original simulations, rearranging Cardboard City a final time. It resulted in a compact, tightly focused factory and well-positioned machines — saving the company millions of dollars — at a total cost of only $2,000.

General Electric

GE's Change Acceleration Process (CAP) represents the very best of experiential learning. The successor to the Work-Out Program, CAP grew from CEO Jack Welch's realization that the future was inherently uncertain — and was likely to stay that way. Surprises were inevitable, and, it was impossible to anticipate upcoming events. But it was possible to manage the change process more effectively. What was needed was a set of concepts, tools, and techniques for making rapid adjustments and adaptations. Welch assigned the task of developing a state-of-the-art model to four well-known consultants and CAP was born. Participants come to CAP in teams of eight to twelve people with a real problem to solve. At Welch's insistence, the problems are "need to do; not nice to do." Each team works with a coach, a process expert knowledgeable about the problem at hand who guides them through the problem. The coaches are constantly present, working with the teams at the home site before training begins, during training, and after CAP classes. The training alternates between learning the CAP framework— straightforward and easy to apply guidelines that divide the change process into seven steps: leading change, creating a shared need, shaping a vision, mobilizing commitment, making change last, monitoring progress, and changing systems and structures — and applying it to the problem they've brought with them. Does CAP impact the bottom line? Yes. At GE Plastics Japan there was nothing but red ink from 1989 to 1993. Managers signed up for CAP training as a last resort in 1994. By the end of the year the business was breaking even. In 1995 net income was $18 million and the company was on solid footing for the first time.

About the Author

David A. Garvin is a professor at Harvard Business School.