Data Science Investments a "Core Driver" of Wayfair's E-Commerce Success

Jamie Grill-Goodman
Editor in Chief
Jamie Grill-Goodman profile picture

E-commerce retail Wayfair's Way Day event on April 25th resulted in the biggest revenue day in its history, according to co-founder and CEONiraj Shah and the company expects the event will grow further in future years. During Way Day, customers in the U.S. and Canada were able to shop at Black Friday level prices across tens of thousands of products with free shipping on everything.

Wayfair ran several initiatives to drive awareness of Way Day with shoppers, including TV, direct mail and PR, and Niraj said the company's teams have been able to test and learn from promotional approaches and site features during the cyber five period and other sales events, enabling the business to give shoppers an experience "that we think was highly engaging and brought them the best of Wayfair."



"Data science is a term that is often overused in e-commerce," said co-founder and co-chairman Steve Conine on the company's first quarter 2018 earnings call. "At Wayfair, we believe we are at the forefront of using data to improve the decision making of our employees and ultimately to enhance the experience of our customers and our suppliers."

Conine claimed Wayfair's investments in data science have been a "core driver" of its success to date. The e-com retailer boasts a team of around 1600 engineers and data scientists, built up over the last four years.

"We believe we offer driven people the chance to work on complex and rewarding problems in an environment where they develop quickly and drive impact across our business," said Conine. "We place great importance on our ability to develop algorithmic technology in-house at a fast pace and we see this as a competitive advantage of our business."

Wayfair's long term investment in business intelligence, technology infrastructure and custom software provides its data scientists the ability to act on real-time marketing, customer product and logistics data.

"We have been able to build multiple platforms in-house at a strong ROI as a result," he noted.

Wayfair's data science team’s work is powered by the retailer's rich customer and product data within the home category in which it sells.

"On average, we capture and store four terabytes of data every day and over the course of a year, we track approximately 40 billion customer actions on our site, putting us in a great position to generate data driven insight to improve the experience of our customers," said Conine.

"One of our unique advantages is that our data science team impacts the entire customer experience. Even before a customer is familiar with the Wayfair brand, we have proprietary marketing measurement models that prioritize the type of ads and creative content that are likely to be most engaging to specific groups of potential customers across digital, direct mail and television marketing. When customers reach our site, our algorithms consider aspects around product diversity, style, popularity and more to offer an engaging experience every visit."

Conine explained that it's the collaboration between data scientists and functional teams across the business which results in quick iteration, implementation and measurement for the retailer. He also pointed to Wayfair's pro-data culture, claiming it helps makes this team work "much more natural than it would otherwise be" resulting in a "rare" level of integration within the industry.

"Data science and machine learning influence a wide variety of our business areas, such as onsite personalization, dynamic pricing, algorithmic merchandising, demand forecasting and advertising technology among many others," he noted.

Wayfair offers e-commerce shoppers' features like "search within photo" that use deep learning to find Wayfair products that are visually similar to photos that shoppers submit themselves. Once an order is placed, Wayfair's models kick in to determine the most effective way to fulfill the order, taking into account the retailer's proprietary logistics network.

"Deep analysis of data is essential to how our business operates and to how we win with customers," he said.