eBay – Big Data with SAP HANA

Photo: iStockphoto
Photo: iStockphoto

Remember when you had to actually drive to the store to buy something, or fight crowds during the holiday shopping season? Pre e-commerce, there were simply no other options.  Shopping was a time-consuming and stressful endeavor: it was difficult to find sought-after products and even more challenging to bargain shop for the best deal. At the time, these issues weren’t perceived as challenges – but in retrospect it’s painful to think of life without the convenience of online shopping.

Small businesses were also limited, able to sell only to local consumers. It’s a wonder how they survived. Luckily for over 100 million eBay members, and the 30,000 new users who join each day, these are problems of the past.

eBay: the world’s largest marketplace

As the world’s largest marketplace, eBay has been helping small businesses like Wheel & Sprocket achieve greatness, and users like Jack Sheng receive over 2 million pieces of user feedback, for nearly two decades. Business is certainly booming for eBay, but analyzing a billion transactions a day has its challenges.  eBay wanted to provide timely intelligence to sellers, improving predictability and communicating areas of opportunity, so the company needed a better way to sift through its massive 100 petabytes of data to separate the signals from the noise.

Next page: Real time insights with SAP HANA

SAP HANA helps eBay identify health of its marketplace

Previously, eBay needed more than 300 analysts studying data on a full-time basis just for the North America marketplace. Most of the process was manual and largely depended on luck and experience. Consequently, ‘signals’ would be identified too late, resulting in missed opportunities and delayed corrective action. After partnering with SAP, eBay will benefit from real-time insight into the health of its marketplace – made possible by the early detection system powered by Predictive Analytics on the SAP HANA platform. SAP HANA is able to analyze 500 metrics and select the best model automatically to determine true positives with 100% accuracy at 97% confidence. The reliable automated detection system will free up bandwidth so eBay analysts can spend their time on strategic tasks and eBay can focus on what it does best: connecting buyers with things they need and love.