Stephen Powell and Bob Batt T’06 love a challenge, and ill-structured business problems give it to them. In their new book, Modeling for Insight, they present an original approach to modeling those kinds of problems, which cannot be solved in the conventional sense but can be explored for useful insights.
Imagine that the Red Cross would like to analyze a new policy under which it would pay volunteers for blood donations. This is an ill-structured problem because the Red Cross has no relevant data, it doesn’t know what the objectives are, and it doesn’t know how its decision would influence the outcomes. You can’t hope to solve this problem, but Powell and Batt show how to uncover deep insights that will help the Red Cross think through this question.
Powell, professor of business administration at Tuck, and Batt, a Tuck Fellow, are experts on business modeling and simulation. They point out that millions of people use modeling regularly, like the spreadsheets used to test the financial impact of investments or to analyze the effect of price on profitability. “Many business models,” they say, “focus on well-structured problems that are known to be solvable. But many business analysts, especially consultants, face important, ill-structured problems and struggle to use the same modeling tools effectively.” Powell and Batt wrote Modeling for Insight for skilled end-user modelers, regardless of the specific functional area. “We want to teach how to approach these problems,” they say, “to inculcate a specific problem solving process and teach specific tools for modeling.” Powell developed his approach to modeling through teaching The Art of Modeling, a course popular among Tuck students planning careers in consulting and investment banking, and through research into how novice modelers work. “When I began teaching modeling,” says Powell, “I told my students I couldn’t tell them how to model, but I could coach them through the process. Eventually, though, we worked out some principles and a structured approach to modeling, especially for the hardest problems.”
Powell S G and Batt R J, Modeling for Insight, Wiley, 2008