Balancing Causal and Predictive Modeling at Wayfair
Balancing Causal and Predictive Modeling at Wayfair


With the rapid advancement in technology within online advertising, marketers are required to turn to black-box algorithms to implement effective ad buying and targeting strategies. In this talk, I focus on how Data Science at Wayfair balances causal and predictive modeling techniques to build our display retargeting bidding AI. Using evidence from a large scale field experiment, we utilize a two-stage least squares regression to model the marginal impact of a single impression. Comparing the causal model to a simple predictive model, we demonstrate that merely maximizing the predictive power of an algorithm can potentially lead to undesirable results for the business.


Bradley is a Data Scientist at Wayfair, the largest online-only furniture retailer in the U.S., where he builds machine learning models to better understand how marketing can best solve consumer problems. Specifically, he’s working on the bidding and targeting algorithms which power Waystack, Wayfair’s internal ad-serving platform. He earned an MBA and is finishing his Ph.D. in Quantitative Marketing Strategy at Arizona State University.

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google