Feb 21, 12:00 AM UTC 


New Perspectives on Marketing Analytics


With abundant and increasing customer data, marketing professionals and scholars face some of the most exciting challenges in data analytics today. This conference, hosted in conjunction with the Wharton Customer Analytics Initiative (WCAI), will convene leading minds across multiple disciplines to share their perspectives on data analytics and marketing, drawing on their expertise in economics, information systems, statistics, computer science, and other related fields.

A limited number of seats are available for WCAI partners and friends. Please contact Hanna German at hgerman@msi.org to register.

Co-sponsored by

Wharton CAI

Agenda Description

8:00 – 9:00 a.m. Registration & Networking Breakfast
9:00 – 9:30 Welcome Remarks
Earl Taylor, Chief Marketing Officer Marketing Science Institute
Carl F. Mela, Duke University and Executive Director, Marketing Science Institute
Eric Bradlow, Chairperson, Wharton Marketing Department and Co-Founder and Academic Director, WCAI
9:30 – 10:00 The Problem with Attribution Models
Steven Tadelis, University of California, Berkeley
Attribution models are based on the premise that that pre-purchase engagement with ads can help predict the effectiveness of ads. However, spurious correlations between customer intent and customer behavior make this a challenging exercise that often yields biased estimates of ad effectiveness. This presentation explains these problems and shows how experimentation and A/B testing can alleviate some of these problems.
10:00 – 10:30 Incrementality & Attribution
Will Bullock, Director of R&D, Facebook
Marketers regularly make decisions regarding which ad campaigns are most effective, such that they can invest in strategies and tactics that grow their business. Almost all these decisions are made based on last-click attribution or standard reporting metrics. To gain deeper insights regarding their value, we develop new approaches to leverage a large corpus of experiments. These new approaches help demonstrate which metrics serve as the best guides for marketers focused on incrementality for both brand and conversion outcomes.
10:30 – 11:00 Break
11:00 – 11:30 Leveraging Multi-Attribution Modeling
Pankaj Kohli, Head of Applied Marketing Data Science, Adobe
Multi attribution modeling has emerged as an important approach to leverage customer level online touch points and behavioral data. These models will continue to evolve as our ability to obtain more diverse customer centric touch points through technology grows with time. In this presentation, we focus on how this approach is being leveraged at Adobe and provide an overarching perspective on this approach.
11:30 – 12:00 p.m. Optimal Targeting Policy Evaluation
Sanjog Misra, University of Chicago
The availability of ubiquitous data, sophisticated machine learning approaches, and easy avenues for experimentation affords firms and researchers a unique opportunity to estimate heterogenous effects of marketing treatments and use these estimates to implement improved targeting decisions. Even so, the evaluation of different targeting policies is limited by the costs of experimentation. In this presentation we will demonstrate how a single randomized experiment can be used to evaluate the profitability of an arbitrary number of targeting policies without incurring such implementation costs. Working with a large retailer we conduct use our proposed approach to implement and compare a large number of machine learning approaches to estimating individual level responses to catalog mailings. Our analysis reveals substantial differences in profitability across machine learning approaches as well as the mailing strategies obtained from them. Finally, we outline a new optimal mailing strategy that generates a substantial increase in profits compared to the status-quo currently deployed by the firm.
12:00 – 12:30 p.m. Post Purchase Search Engine Marketing
Shawndra Hill, Senior Researcher, Microsoft Research
Though consumer behavior in response to search engine marketing has been studied extensively, few efforts have been made to understand how consumers search and respond to ads post purchase. This is in part due to the fact that purchases are difficult to track and link to search queries. Thus, it is unsurprising that advertisers have been targeting consumers on search engines similarly regardless of the heterogeneity in their search intents and context. Advertising to current customers the same way as to prospective customers inevitably leads to wasteful and inefficient marketing. Employing a unique dataset that combines both search query and purchase data, we examine consumers’ searching behavior and response to search engine marketing after purchase. We study large advertising campaigns for two popular technology products. We find that over half of the branded keyword searches come from consumers who already purchased, and that advertising response varies based on whether searchers are pre or post purchase. In general, post-purchase searchers are less likely to click on focal brand ads (i.e., they are less responsive to ads for products they already own). However, post-purchase searchers are still responsive to advertising, and much more likely to click on ads for complementary products (i.e., they are more responsive to ads for relevant products other than the focal product). Our findings offer unique academic contributions regarding consumer behavior along with practical implications for how platform should market post-purchase targeting and how marketers should advertise to customers post purchase.
12:30 – 2:00 Lunch
2:00 – 2:30 New School Methods with Old School Models
Justin Rao, Head Economist and Vice President of Data Science, HomeAway, an Expedia Company
In this talk Justin will describe the collision between new ML and “data science” methods for advertising and other marketing optimizations and more traditional marketing/economic models. In his experience, a principled synthesis of both provides the best business outcomes, and he will discuss specific examples and case studies to demonstrate this point. Justin’s presentation will draw from his own academic work, the work of others and industry experience.
2:30 – 3:00 Targeting and Privacy in Mobile Advertising
Hema Yoganarasimhan, University of Washington
Mobile in-app advertising is growing in popularity. While these ads have excellent user-tracking properties through mobile device IDs, they have raised concerns among privacy advocates. There is an ongoing debate on the value of different types of mobile targeting, the incentives of ad-networks to engage in behavioral targeting, share user-data with advertisers, and the role of regulation. We use large-scale data (over 150 million impressions) from the leading in-app advertising platform of a large Asian country to answer these questions. Using a combination of machine learning and economic models, we quantify the relative value behavioral and contextual targeting, the value of user-identifiers, and the ad network’s incentives to preserve user privacy. Our findings are of relevance to both the mobile advertising industry and regulatory bodies.
3:00 – 3:30 Break
3:30 – 4:00 Customer-Based Corporate Valuation
Daniel McCarthy, Emory University
Dan will discuss new ways of valuing corporations from the “bottom up”—i.e., determining the forward-looking financial valuation of the customer base — as a complementary perspective to the standard “top down” methodologies that dominate current valuation practice. This notion, sometimes called “customer equity”, is gaining increasing interest among a variety of functional areas both inside corporations (e.g., corporate development, business intelligence, accounting and finance, marketing) and outside of them (e.g., hedge fund, mutual fund, private equity, venture capital). This session will introduce this new concept, show how it fits within traditional valuation approaches, then apply the methodology to several popular publicly traded companies operating across different business settings. These examples highlight customer-based corporate valuation’s growing ability to move markets, making it a crucially important methodology for the C-suite and everyone below it to understand.
4:00 – 4:30 Practitioner & Academic Reflections on the Day
Gerard Tellis, University of Southern California
Jim Oliver, Vice President Member and Business Intelligence, SamsClub.com
4:30 – 4:45 Wrap Up
Carl Mela, Duke Univeristy and Executive Director, Marketing Science Institute
Eric Bradlow, Chairperson, Wharton Marketing Department and Co-Founder and Academic Director, WCAI
4:45 – 6:15 Networking Reception
8:00 – 9:00 a.m. Registration & Networking Breakfast
9:00 – 9:15 Opening Remarks
Earl Taylor, Chief Marketing Officer Marketing Science Institute
9:15 – 9:45 A Perspective Across the Demand Side
Linda Vytlacil, Vice President, Retail Tech Data Science Walmart Labs
Companies have accelerated their use of marketing science to improve effectiveness and return on advertising and marketing communications spend. Among retailers, we are also benefiting from employing a customer-centric view to the application of big data analytics for other areas of marketing including price, product, and placement. In this practitioner’s overview, we discuss three uses cases for marketing analytics on solutions across the demand side of retail.
9:45 – 10:15 Online Markdowns
Kostas Bimpikis, Stanford University
Online retail reduces the costs of obtaining information about a product’s price and availability and enables customers to better time their purchases to potentially take advantage of lower prices. At the same time, firms can observe and exploit their customers’ monitoring behavior and use this information to better price and target their promotions. This talk explores this interplay and provides prescriptive recommendations that may help retailers in designing their online channels.
10:15 – 10:30 Break
10:30 – 11:00 Active Consideration: Connecting Discovery to Do
Gunnard Johnson, Head of Measurement Science and Insights, Pinterest
Pinterest’s mission is to help Pinners discover the things they love and inspire them to do them in real life. This presentation will review analyses and signals of Pinner behavior that mark when a consumer shifts from passive browsing to active consideration of a product or service. As such we’ll highlight the implications for brands looking to drive new customer growth as well as higher return on ad spend. This presentation will be relevant for anyone wanting to learn more about Pinterest as well as data science techniques used to quantify digital behavior.
11:00 – 11:30 Agile Analytics
Kathy Koontz, Practice Director Customer Journey, Teradata
Just like Shaun White’s double-corked 1440s in the Winter Olympics, you can do a lot of complex activity in Marketing Analytics. But if you don’t stick the landing, it doesn’t really count. As companies continue to solve challenges around leveraging big data and applying advanced analytics, that landing comes when those capabilities are automated across the entire enterprise. Optimal analytic value comes when insights and new models can be integrated into operational decision making in a matter of days from the time they’re identified. Companies struggle to rapidly iterate; develop, tune and deploy models; and optimize and automate decisions across hundreds of analytic models, often directing different business processes. Kathy Koontz, Practice Director for Teradata’s Customer Journey Practice will provide perspectives on how organizations can ensure that after all the amazing twists and turns of their analytic work, they can stick the landing.
11:30 – 12:00 p.m. Closing Remarks
Carl Mela, Duke Univeristy and Executive Director, Marketing Science Institute
Eric Bradlow, Chairperson, Wharton Marketing Department and Co-Founder and Academic Director, WCAI
Earl Taylor, Chief Marketing Officer Marketing Science Institute

Hotel Accomodations

MSI has secured a block of rooms at Le Meridien San Francisco at a discounted rate of $319 per night. In order to secure a room at the group rate, please contact the Le Meridien at (866) 837-4184 and reference the MSI Data Analytics Conference when booking.

MSI’s group rate has been extended through February 2, 2018. Please be sure to make your reservation as soon as possible to secure the discounted rate.

Le Meridien San Francisco
333 Battery St
San Francisco, CA 94111

Please note, Le Meridien requires notice 72 hours before check in for cancellations.

By using MSI.org you agree to our use of cookies as identifiers and for other features of the site as described in our Privacy Policy.