Marketing Analytics


James Lemieux

Senior Applied Operations Research Professional, Operations Research, Research & Development
General Motors

Use At Your Own Risk: The Perils Of Implementing Advanced Marketing Analytics

The business community has witness tremendous growth in the use of advanced marketing analytics to address real-world problems in such areas as consumer choice, revenue management and price optimization, advertising and new product development. In recent years, these developments have accelerated in response to both the democratization of analytics via open-source software, as well as the so-called “big data” revolution fostered by industry pundits. These phenomena have put increasing pressure to more widely implement advanced marketing analytics in the software used to assist managerial decision making. Unfortunately, several of the models and methodologies that were developed to solve well-defined problems using hand-picked samples often fail to generalize when applied to a broader class of business objectives that representative all customers, products and industries. This presentation identifies several common hazards the author has faced when attempting to implement advanced marketing analytics as part of robust software solutions. Apart from offering a panacea, the author provides some advice and strategies for addressing such issues and warning: use at your own risk.


James Lemieux is a seasoned marketing analytics professional with over 14 years of hands-on experience applying advanced statistics and econometrics models for clients in industries including: automotive, hospitality, direct marketing, credit risk, logistics and high technology. James received his Ph.D. in Marketing, and M.S. in Mathematics from the University of Texas. He currently resides in Royal Oak, Michigan where he is a Senior Operations Research Professional in the Operations Research department at General Motors R&D.

[themify_hr color=”light-gray” width=”100%” border_width=”1px” ]


Daymond Ling

Professor, School of Marketing
Seneca College

Analytics Cannot Imbue You With A Soul

There is no question analytics is one of the hottest technologies of this era. And rightly so as analytics has solved problems hitherto unconquerable, delivered deep insights to foster innovation, created competitive advantage, what’s not to love about analytics?

While success stories can lead people to believe all they need to do is turn to analytics and all problems will be solved magically, this is simply not so, analytics in and of itself won’t tell you how to run a business. For every success story we see, how many failures are there that we don’t hear about? Like any other project, analytics projects have its fair share of failures. A quick search on the internet will reveal that the analytics path to the pot of gold at the end of the rainbow is littered with skeletons at the side of the road.

The challenge before us then is what we can do to ensure we deliver genuine value in our analytics projects. While there are many factors that influence success, this presentation focuses on aligning our analytics methodologies to business strategy. Two examples, Incremental Response Modelling and Direct Marketing Optimization, will be used to illustrate the need to thoughtfully adapt our analytics approach to business strategy to ensure long term success.


Daymond Ling is an advocate for applying Advanced Analytics to real world problems to create value. He is passionate in the Pursuit of Analytics Excellence, continuously drives for business and analytics innovation, loves to learn from people as well as share his knowledge.

Mr. Ling is able to bridge between analytics and business and has created, led and sustained world class analytics departments. Stewarding the analytics capability gave him in-depth knowledge of how to lead and motivate analytics professionals. He also had the privilege of mentoring, coaching and developing many analytics talents in the Art and Science of Advanced Analytics problem solving and learned tremendously from these experiences.

Mr. Ling has many years of industry experience in analytically transforming businesses. When done right, analytics is tremendously powerful and transformative as we bring insight, clarity and truth to help organizations run better today and innovate for a better tomorrow; he has also seen many cases where analytics are conducted inappropriately, results abused and projects failed miserably. There are a lot of powerful lessons to be learned from both successes and failures.   He is able to see far because he stood on the shoulders of giants. Many generations of brilliant people created this field that we now know as Advanced Analytics. Like those before him, he wants to share his experiences with future generations of analytics professionals and contribute to the development and success of future Analytics Unicorns.

[themify_hr color=”light-gray” width=”100%” border_width=”1px” ]

speaker-pieter-van bouwel












Pieter Van Bouwel

Senior Analyst
Python Predictions

Process Mining – The Road To A Superior Customer Experience

ING Belgium strives to create a superior customer experience. In this case, we present how Process Mining has enabled ING Belgium to understand and improve the customer experience in a crucial customer-facing process. Using internal process logs, we have started this journey by understanding and mapping the current performance of the end-to-end process. Next we applied both traditional (six sigma-style) approaches as well as more modern (Process Mining and Predictive Analytics) approaches to help ING Belgium serve its clients better. We illustrate the benefits, milestones, requirements and potential pitfalls we encountered. We conclude by summarizing ING Belgium’s future plans and ambitions in this exciting area.


Pieter is a senior analyst at Python Predictions. He’s a master in applied economics and an advanced master in marketing analysis. Pieter has built extensive domain experience in a wide range of industries, working on diverse analytical projects in predictive analytics and process mining. At Python Predictions, Pieter has the role of analytical expert and is mainly involved in project management.

Wouter Buckinx

Managing Partner
Python Predictions


Wouter Buckinx is Managing Partner at Python Predictions, a Belgian niche player with expertise in the domain of Predictive Analytics. He currently has over 10 years of hands-on experience in different industries such as retail, mail-order, telco, banking, insurances, utilities, subscription services and fundraising. Wouter has experience in a large variety of applications of Predictive Analytics, ranging from the development of predictive models for marketing (acquisition, cross-sales, retention), risk (credit risk, fraud risk) and operations (forecasting, process mining) – to the industrialisation and monitoring of large-scale advanced analytics, and the creation of recurrent analytical processes. During his PhD at Ghent University, Wouter has performed research for several European and worldwide companies. He has written several scientific publications, and has presented on different academic and business conferences. 

[themify_hr color=”light-gray” width=”100%” border_width=”1px” ]


























Alex Vayner

Vice President, Data Science Innovation

Applying Optimal Alignment Techniques To Identify Canonical Product Adoption Patterns

Across industries, there is sustained interest in understanding how individual consumers transition from one product to another over time. To address this need, Equifax has begun an initiative to model how a consumer’s overall pattern of product adoption and other life events – their product journey – proceeds across various Equifax data repositories. The analysis leverages an encoding scheme wherein each sequence of distinct event types is first rendered as a string. Associated with each event is the time at which it first occurred (such as originating an auto loan, or filing for bankruptcy) and, where relevant, the time at which the event ceased (such as paying off a mortgage). The events in each string are canonically ordered by first occurrence date. Based on this encoding, one defines a natural distance metric between two events of the same type (e.g., two auto loans). Using an optimal alignment paradigm, a distance metric is then defined between any two strings. With this string metric in hand, a customized k-median clustering analysis is then performed to identify clusters with patterns whose event types, relative ordering, and time ranges are similar. While this clustering framework exposes distinct product journey patterns, it is also useful for predictive analytic purposes, allowing, in particular, the prediction of future events in a subsequent outcome period.


Alex Vayner is a Vice President of Data Science Innovation @ Equifax, where he leads the team responsible for designing solutions and products that disrupt the traditional credit market place. Throughout his career Alex has focused on leveraging data and quantitative methods to establish business intelligence platforms and advanced analytics products that can generate sustainable data-driven actionable insights. Alex came to Equifax from PwC, where he was a Director in the Advisory Analytics practice.  Before PwC Alex was a VP and Analytic Practice Leader for Brand Velocity, a boutique management consulting firm.  Alex joined Brand Velocity from Altisource, a public technology firm where he headed strategy, analytics and innovation for the e-commerce portal Hubzu.

Earlier in his career Alex was an engagement manager at Booz Allen Hamilton, where he led analytics for a new commercial financial services practice. At his first consulting engagement, Alex worked at a boutique strategy consultancy, Mars & Co, where he worked on international projects in market research, financial modeling and valuation work for multiple Fortune 50 clients. Alex earned his bachelor’s degree in mathematics from University of Florida and his master’s degree in applied mathematics from Georgia Tech.


Trevis Litherland

Senior Data Analyst


Trevis Litherland is a Senior Data Scientist on the Data Science Innovation team at Equifax. A probability theorist by training, he received his Ph.D. in Mathematics from Georgia Tech. His interest in probability theory sprang from many years of practical experience as a credit scoring analyst at the consulting firm of Scoring Solutions, Inc. in Atlanta, where he worked both before and after his graduate studies. Throughout his career, Trevis has been interested in automation and optimization of all aspects of the credit risk modeling process, including auto-binning and reject inference. Since joining the Data Science Innovation team at Equifax in 2014, his attention has turned to bringing Equifax’s many data sources together in new ways that speak to emerging business challenges that clients are facing, with a recent focus on trended data solutions.

[themify_hr color=”light-gray” width=”100%” border_width=”1px” ]


Michel Wedel

Pepsico Professor of Consumer Science
Robert H. Smith School of Business, University of Maryland

Eye Tracking:  A Review of Concepts, Metrics and Findings

Motivated by its increasing popularity in applied and academic marketing research, I review the basics of eye movement research, including the physiological background, the workings of eye tracking equipment, and the computation of eye tracking measures and metrics. Then, I provide an overview of applications, focusing on key findings in advertising research. I conclude with an outlook on exciting new developments, including the use of eye tracking to explain and predict human search and choice


Michel Wedel is the Pepsico Chaired Professor of Consumer Science at the Robert H. Smith School of Business, and a Distinguished University Professor, at the University of Maryland. He has a Ph.D. in Marketing from Wageningen University, and MS.C.’s in Biomathematics and Statistics.

His main research interests are in the application of statistical and econometric methods to problems in marketing. Much of his recent work addresses issues in visual marketing, using eye-tracking technology. He has written books on Market Segmentation and Visual Marketing. He is area editor of Marketing Science, the Journal of Marketing Research and the Journal of Marketing.

Michel has published more than 170 peer reviewed articles which have been cited over 15,000 times. His work has received several awards. He was ranked the most productive economist in the Netherlands and the most productive market researcher in the world. He received the Muller award for outstanding contributions to the social sciences from the Royal Dutch Academy of the Sciences, the Churchill award for lifetime contributions to the study of marketing research from the American Marketing Academy, and is a fellow of the American Statistical Association and the Section of Marketing Science of the Institute for Operations research and Management Science. The University of Maryland has awarded him with the Distinguished Scholar teacher award.