Last weekend I had the privilege of attending a research symposium sponsored byStubHub and hosted by the Wharton Business School at the University of Pennsylvania. Wharton has a research program called the Wharton Consumer Analytics Initiative which is “the preeminent academic research center worldwide focusing on the application of customer analytics to business problems.” The main activity of the WCAI are the “Research Opportunities” that pair corporate sponsors and academic researchers in order to address data-centered challenges.
Last year I submitted a proposal to a WCAI Research Opportunity sponsored by StubHub entitled “Customer Purchasing Behavior as a Function of Sport Team Performance” that sought to study how ticket purchasing behavior by customers varies as a function of the performance of their preferred sports team. SRC’s proposal was one of the six accepted, which is a significant achievement by itself, but I didn’t know how significant until I presented last Friday. It was the first time that a private company had submitted — and gotten accepted! — a research proposal to a WCAI Business Opportunity.
I started SRC with the goal of conducting world-class research which would be incorporated into products and services for customers in the soccer industry. To submit a research program that stood out over scores of other submissions, some of which from elite business schools, is something of which I am very proud and represents the type of work that SRC aspires to deliver to its clients.
The other projects presented at the symposium considered the effects of various types of customer engagement, whether by online banner or email promotion, on customer behavior. My project was different in that it sought to understand the characteristics of the customer. There is a lot of arguing among sports fans over which fans are most loyal: which set of fans are more likely to pack stadiums to watch their team even when they aren’t playing well? The objective of my research was to examine how sensitive customer purchasing decisions were to the current performance of their preferred sports team, and if so develop a statistical model to predict those behaviors. I focused on customers who were buying tickets for the Big 4 sports leagues in North America (NFL, NHL, NBA, and MLB). I would have liked to have examined purchasing behavior in MLS, but there weren’t enough data available in the dataset to make any results meaningful.
The presentation was very well received by the StubHub staff and the Wharton faculty who attended the symposium. The major finding was that there is some mild elasticity of customer purchasing behavior to team records, especially for teams at the bottom of their respective conferences. I found it very interesting that it’s possible to develop a predictive model with fairly strong predictive power (say, 70% accuracy) even if the model captured just 10% of the variance in the output variable.
I’m preparing a journal paper on the research so I’m not able to present all results at this time, but if you’re interested in an overview of the work you can download a research summary from SRC’s anonymous FTP site.
My sincere thanks to Professors Elea Feit, Peter Fader, and Eric Bradlow for their hospitality during my visit to Wharton, and thanks to the StubHub personnel and other attendees for stimulating discussions about this work. I set out to establish a foothold in sport marketing analytics and develop analytics tools that can be applied for Soccermetrics. This project has allowed me to achieve both objectives.