You are previewing premium content. 
How Macys.com visualizes the future
- TAGS:analytics, BI, business intelligence, e-commerce, ecommerce, Insider Content, Macy's, Macys.com, predictive analytics, retail
- IT TOPICS:Development, Enterprise Apps
What's the likelihood that a loan applicant will default? What mix of sizes, styles and colors will sell best in a given store? Predictive analytics can provide forecasts that work like magic, but the results are only as good as the models that underlie them. Those models are developed by people like Kerem Tomak and his team at Macys.com, the online business unit of the Macy's department store chain.
Predictive analytics didn't exist a year ago at Macys.com but is quickly gaining momentum, says Tomak, vice president of marketing analytics.
His predictive analytics process, built using tools from SAS Institute, goes through four stages: discovery, data visualization, modeling and finally validation and testing. The data discovery and visualization steps guide the modeling approach, and being able to visualize the problem is critical, he says.
To continue reading, register here to become an Insider
It's FREE to join
Already an Insider? Sign in
Free Insider Guide