|
Track 11: Business Marketing
Tuesday, May 1
9:10am-10:00am
Managing Customers for Profits
Tim Bohling, MBA
Director, IBM Global Business Services Market Intelligence
IBM Corporation
V. Kumar, PhD
ING Chair Professor of Marketing
Executive Director, ING Center for Financial Services
School of Business, University of Connecticut
In this session, Bohling and Kumar will discuss case studies that exemplify different strategies used by companies to maximize profit. The reality is that companies treat customers differentially. But how do they decide which customers should be given preferential treatment, which customers to interact with through inexpensive channels like the Internet or touch-tone phone, and which customer to let go? How do firms decide the timing of an offering to a customer? What kinds of sales and service resources should be allocated to generate word-of-mouth referrals? In order to implement a profitable strategy, companies develop a measure called Customer Lifetime Value (CLV)) and use that indicator on which to base their marketing decisions. Kumar and Bohling will describe the organizational and implementation challenges that surround the adoption of customer value management. Several case studies of profitable implementation of a CLV-based approach will be presented.
Tim Bohling has held several marketing leadership positions in IBM. Currently, as Director of IBM Global Business Services Market Intelligence, he is responsible for formulation and execution of a Market Intelligence plan that delivers thought leadership, deep insights and recommendations that drive IBM to take better action, improve performance and fuel growth. Bohling directs the Worldwide Market Intelligence program consisting of market analysis, primary research, secondary research, opportunity analysis, database marketing, customer analytics and competitive intelligence. His consulting experience includes many Fortune 50 companies including AT&T, Exxon, Texaco, PepsiCo, Bank of America, among others. Bohling frequently presents at industry conferences and has several areas of interest including marketing strategy, customer management, brand equity, marketing implementation and market intelligence. His research has been published in several journals including Marketing Science, Journal of Industrial Marketing Management, Journal of Interactive Marketing, among others. He received a BBA in Marketing and an MBA with Highest Honors from the University of Houston.
V. Kumar (VK) is the ING Chair Professor of Marketing and Executive Director, ING Center for Financial Services in the School of Business, University of Connecticut. He has been recognized with over 20 teaching and research excellence awards including the Paul H Root Award twice for the paper published in the Journal of Marketing that contributes to the best practice of marketing, and the Don Lehmann Award twice for the best paper published in the Journal of Marketing/Journal of Marketing Research over a two-year period. He has published over 125 articles in many scholarly journals in marketing including the Harvard Business Review, Journal of Marketing, Journal of Marketing Research, Marketing Science and Operations Research, book chapters and books. His books include Customer Relationship Management: A Databased Approach, Marketing Research, and International Marketing Research. He was recently listed as one of the top five ranked scholars in marketing worldwide. He received his PhD from the University of Texas at Austin.
10:30am-11:20am
How to Project Customer Retention
Peter S. Fader, PhD
Frances and Pei-Yuan Chia Professor of Marketing
The Wharton School of the University of Pennsylvania
At the heart of any contractual or subscription-oriented business model is the notion of the retention rate. An important managerial task is to take a series of past retention numbers for a given group of customers and project them into the future in order to make more accurate predictions about customer tenure, lifetime value and so on. In this session, as an alternative to common curve-fitting regression models, Fader will develop and demonstrate a probability model with a well-grounded story for the churn process. He will show that the basic model—which can be implemented in a simple Excel spreadsheet—provides remarkably accurate forecasts and other useful diagnostics about customer retention. A detailed appendix covering the implementation details and offering additional pointers to other related models will also be included.
Peter S. Fader is the Frances and Pei-Yuan Chia Professor of Marketing at the Wharton School of the University of Pennsylvania. His research, teaching and consulting interests focus on the analysis of behavioral data to understand and forecast customer shopping/purchasing activities. His projects are drawn from a wide range of industries, and much of his work highlights the common behavioral patterns that exist across them. Fader believes that marketing should not be seen as a “soft” discipline, and he works with many companies and industry associations to improve their “quantitative literacy” in this regard. His work has been published in (and he serves on the editorial boards of) a number of leading journals in marketing, statistics and the management sciences. He has won many awards for his teaching and research accomplishments.
11:30am-12:20pm
Data Fusion: The Good, the Bad and the Ugly
Gina Pingitore, PhD
Chief Research Officer
J.D. Power and Associates
Increasingly, companies need a broader, more in-depth understanding of what consumers think, feel and do. At the same time, however, acquiring this additional information from the same customer can be challenging. As an alternative, several researchers have tested the viability of using data integration mechanisms known as data fusion. Fusion leverages various analytic techniques to link information from one set of respondents from one dataset to others in another dataset. In this talk, Pingitore will:
- Provide an overview of fusion and the best analytical practices to fuse different data sets together.
- Provide guidelines on how to evaluate the success of fusion efforts. While some researchers will rely on spilt-sample assessment, few researchers conduct construct validation assessments. The findings clearly show that doing both is needed to best assess the successfulness of fusion efforts.
Co-authors: Minghuang Zhang, MS; Weihua Haung, PhD; Gene Cameron; Steve Witten; JD Power and Associates.
Gina Pingitore is Chief Research Officer at J.D. Power and Associates. She joined the firm in 2002 and has more than 15 years of experience designing, implementing and evaluating quantitative and qualitative research studies. Previously, Pingitore was Director of Research and Planning at Publicis Dialog, a marketing and advertising firm, where her responsibilities included identifying consumer insights to create compelling advertising messages; tracking ad campaign effectiveness through statistical analyses; design and implementation of online research to allow rapid deployment of consumer surveys for message testing; and development of ad pre-testing and measurement processes to predict the effectiveness of advertising executions. Earlier in her career, Pingitore worked in academia as a behavioral researcher, where she honed her ability to conduct advanced statistical analyses, including multivariate, repeated measures analyses and logistic regression analyses. Pingitore earned a doctorate in psychology from Loyola University of Chicago and a master’s degree in psychology from Edinboro University of Pennsylvania. She is also a licensed clinical psychologist and has authored over 30 articles published in a variety of professional journals.
3:10pm-4:00pm
Optimization of Product Variety Offerings for Automotive Retail Sales
Gint Puskorius, MS
Senior Technical Leader, Business Systems Research
Ford Motor Company
The world of automotive marketing and sales provides a rich and challenging set of interrelated problems that can be addressed by a broad range of advanced analytical methods. Puskorius will focus in this presentation on the long-standing problem of what level of product variety should be offered to the customer, with implications for mix rate forecasting, web-based marketing, vehicle distribution, inventory management and complexity of incentives. He will explore opportunities and limitations associated with the use of historical sales and web-based configuration data in addressing this question, will discuss mathematical formulations for the problem and methodological approaches, and will provide representative results.
Gint Puskorius is a Senior Technical Leader at Ford Motor Company's Research & Advanced Engineering Center, and currently leads its Business Systems Research activity. He joined Ford Motor Company's Physics Department in 1982, pursuing research in the areas of laser scanning devices, machine vision and robotics. He subsequently became a charter member of Ford's Artificial Neural Networks group, where he worked on the development of advanced training methods for recurrent neural networks applied to diagnostics and control; this work led to an industry-first neural network production implementation for misfire detection. In 1997, Puskorius was a founding member of Ford's Business Systems Research group, and has led a variety of efforts related to residual value analysis, sales volume estimation, incentive analysis and vehicle feature valuation. He has authored over 60 technical papers, has been awarded 10 U.S. patents, and is the recipient of five Ford technical awards and two external awards.
4:10pm-5:00pm
Marketing Decision Models Go to the Movies: Scheduling for Multiplex Theaters
Charles R. Weinberg, PhD
Presidents of SME Vancouver Professor of Marketing and Chair
Sauder School of Business, University of British Columbia
For almost a decade, Weinberg and his co-authors have been working with Holland’s leading movie exhibitor to help in choosing movies to play. Few studies report how modeling systems develop over time, but an evolutionary approach to model development is a key aspect of this presentation. Moreover, in this environment, creative elements are important and some managers distrust analytic approaches. Weinberg and his team originally developed a model for scheduling movies in a single theater. Management next asked for weekly recommendations for choosing and scheduling movies by screen, across all their Amsterdam theaters. The team created a model with two key components, forecasting (based on statistical modeling and managerial inputs) and (heuristic) optimization. More recent work concerns the micro issue of setting the precise times when movies should start. While the models’ technical aspects will be covered, the presentation’s focus is how to adapt and be sensitive to managerial needs and operational constraints. Co-authors: Jehoshua Eliashberg , Wharton Business School; Sanjeev Swami, Indian Institute of Technology; Berend Wierenga, RSM Erasmus University.
Charles B.Weinberg is the Presidents of SME Vancouver Professor of Marketing and Chair of the Marketing Division at the Sauder School of Business, University of British Columbia. For more than 30 years, he has studied the arts and entertainment industry. His early work focused on live entertainment and included the ARTS PLAN model for marketing and scheduling performing arts events. More recently, he has focused on the movie industry with such publications as “Competitive Dynamics and the Introduction of New Products: The Motion Picture Timing Game (Journal of Marketing Research ), “Sequential Distribution Channels” (Journal of Marketing ), and “Implementation and Evaluation of SilverScreener” (Interfaces). A former editor of Marketing Letters and area editor of Marketing Science, Weinberg will chair the 2008 Marketing Science conference. |