Paul Green spent his childhood convinced he was going to be a famous chemist, commandeering the musty attic of his mother’s suburban Philadelphia boarding house to lose himself in his microscopes and Bunsen burners.
But two happy accidents redirected the burgeoning scientist: During an attic chemistry experiment, he brushed a high-voltage Tesla coil against a water pipe, narrowly escaping blowing up his house. Years later, in 1946, he returned home from his military service to Penn armed with a university scholarship to discover that all the chemistry courses were reserved for pre-med students. Green chose economics and mathematics instead.
For almost all of the nation’s Fortune 500 companies, law firms, hospitals, and government agencies, it’s a good thing he did. Today, Green is often called “the father of conjoint analysis,” the powerful predictive statistical technique and backbone of market research.
The framework has been used internationally by thousands of companies, including the hotel conglomerate Marriott Corporation in creating its Courtyard by Marriott chain, and by regional transportation agencies in the New York/New Jersey metropolitan area to investigate the potential of the now-successful EZPass electronic toll collection device, as well as in medical research, public policy and industrial engineering. Conjoint analysis allows marketing managers make accurate decisions about what products and services to sell—and helped make Green marketing’s most cited author. “Wharton has been at the forefront because Green always published in the best publications and his breakthrough methodologies always had a practical touch to them,” says Jerry Wind, Wharton marketing professor, director of the SEI Center for the Advanced Studies in Management and one of Green’s most significant research co-authors and collaborators. “Green inspired through example, his involvement in recruitment, and his value system, his scholarship, his hard work and his sense of the importance of research for the field as a whole.”
But when Green, who retired this year, came on the scene, the world of marketing was in a statistical infancy and market research was a much less mathematical discipline. “Market research done in the late 50s was a lofty enterprise that was more descriptive than proscriptive,” he says.
The Early Days: Sun Oil and Lukens
Green went on to earn his bachelor’s degree in mathematics from Penn, supplementing his income playing piano in a jazz band in what he calls “bars of dubious reputation” a few nights a week in and around Philadelphia. After graduating, Green got a job in the market research department at Sun Oil’s home office in Philadelphia, then went to work for Lukens Steel in the fledgling market research department while working on his Ph.D. at Penn. At Lukens, he began asking questions based on operations research, a kind of methodology developed in World War II in which experts from various disciplines came together to solve wartime problems such as cryptographic puzzles and building better guidance systems. “Operations research provided a new way to look at marketing,” Green says. “Before marketing was a survey. Now we had an accurate way to provide models and procedures.” This blending of disciplines and the research and theories that came out of it were a revelation to Green.
While working in industry, Green carried on with his studies, completing his master’s degree under the tutelage of luminary Simon Kuznets, who would go on to win the Nobel Prize in 1971. “Everyone viewed Kuznets with awe, as did I, although perhaps a bit less so, as I was working in industry,” says Green, adding that Kuznets was “kind but all business.” He went on to become Wharton professor Morris Hamburg’s first doctoral student in statistics. “My tutelage was without peer,” Green says.
In 1961, Green was recruited by DuPont to work in market planning at their Wilmington headquarters. There Green was able to apply the subject of his PhD dissertation on Bayesian inference in his study of the cost versus the value of marketing research information. When Wharton professor Wroe Alderson offered him an academic appointment in 1962, Green left DuPont to work full-time in Wharton’s marketing department. Alderson went on to become a major influence in Green’s academic career, co-authoring Planning and Problem Solving in Marketing with him and persuading companies to partner with Wharton in myriad market research studies that made use of Green’s techniques.
Green’s 12-year stint in industry provided the real- world direction his research would become famous for. “Sometimes these two motivations—the theoretical and the pragmatic—will merge and lead to a high-impact result, that is, an idea that is both intellectually exciting and appealing to the practitioner,” he observed in a University of Pennsylvania Fellow’s Award speech in 1992.
Those high-impact results are now commonplace market research techniques. Perceptual mapping, segmentation, product mapping, cluster analysis, and most importantly, conjoint analysis, are used every day in university market research departments around the world. Many of these market research methods were key in deciding the future of products today’s consumers now consider indispensable.
In 1963, for instance, Green and others in the marketing department worked with Bell Laboratories on the first cell phone. At the time, AT&T had developed a picture phone that had not been successful. Green’s study involved a kind of car phone that was not yet on the market. A thousand people were recruited, loaned cars and given gasoline allowances to test out various intra-car telephones. Green and his team surveyed subjects on their driving and talking patterns and preferences such as what kind of headset they liked, even including a subset of people who enjoyed listening to certain tones and quality of speech.
Changing the Field of Marketing
Green came up with the idea and the name for conjoint analysis while reading a research article from a mathematical psychology journal. The paper, “Simultaneous conjoint measurement: A new type of fundamental measurement,” by R.D. Luce and J.W. Tukey, was published in 1964 in the Journal of Mathematical Psychology and provided a new system to measure rank order data.
“It occurred to me after reading the article that this could be applied to marketing as opposed to just a measurement,” Green said. “We could give people bundles of things that they might want and measure how they react.” The idea that his models could be useful beyond finding out what characteristics already appealed to people was a revelation. Green began to wonder if he could predict what people would do in the future based on how they answered questions about likes and dislikes.
Indeed, he found that he could. Green’s first commercial application of conjoint analysis was with Bissell, a vacuum manufacturer then interested in creating a new kind of product container. Today, Green’s statistical modeling technique has been applied to an enormous list of products and services. And all kinds of companies, from those selling bar soaps and gasoline to those selling luxury automobiles and pharmaceuticals, have partnered with Green and Wharton to learn more about consumer preference and market segmentation.
For his contributions, Green has been recognized with dozens of the marketing field’s highest honors. In 1996, for example, he won the Lifetime Achievement Award from the American Marketing Association, while last year, he won the INFORMS Impact Prize for lifetime achievement and was named the first recipient of the MIT Sloan School of Management Buck Weaver award.
Despite his “retired” status, Green, 78, comes in to his office at least one morning a week, often as early as 6 a.m., to see and work with long-time office mate and friend Abba Krieger, a Wharton statistics professor with whom he continues to conduct research. At home, he often spends hours each day indulging his life-long love of playing jazz on his piano.
And unlike his statistical models, Green makes few predictions about the field of market research, though he does note causal modeling, economic modeling and game theories as likely areas of present and future interest. Charmingly self-deprecating, Green dismisses his bringing of Bayesian analysis and statistical methodology to marketing as “picking the low hanging fruit.”
“The market researcher of today knows much more and has drawn from rigorous disciplines. They’ve learned new sets of techniques,” he says, “and borrow heavily from statistics, operations research, quantitative psychology and applied economics.”
Now, thanks to high-speed computers, market researchers can “look at everything,” says Green. “We can do segment analysis. We can analyze the choices of one group of people within a larger group and we can optimize. For instance, we can ascertain which groups are different from other groups and which ones are alike. All of this matters, of course, because the market is not homogenous.”
A five-year plan for Wharton’s marketing department is in the works with the tentative tagline Wharton Marketing Creates Value. The idea is that Wharton creates methodologies that solve problems to help managers make better decisions. And that, says Wind, is the direct legacy of the impact of Paul Green.
Lea Jacobson is a Philadelphia-based writer. This is her first story for the Wharton Alumni Magazine.