Better search results lead to more sales for Fanatics
October 18, 2016 01:22 PM
When a team wins a championship, online professional and college sports apparel retailer Fanatics Inc. has to launch a lot of new products in a very short period of time so it can capitalize on the spike in traffic that comes from shoppers looking to immediately buy their favorite team’s championship apparel.
About a year ago, Fanatics No. 38 in the Internet Retailer 2016 Top 500 Guide, decided to change from an in-house product indexing and search platform because pushing out new product listings quickly to its 300+ e-commerce sites was difficult, and there was a delay in taking out-of-stock products off sites.
“When the Cleveland Cavaliers won the championship game (Game 7 of the 2016 NBA Finals), it was kind of an odd event,” says Ramana Thumu, vice president of engineering, product and design at Fanatics. “We needed to launch hundreds of products on hundreds of sites across multiple platforms.”
Fanatics needed to do that while dealing with a large spike in traffic from shoppers looking to score the latest championship gear.
“For us to deal with this huge surge in traffic, we need to rely on a stable back-end platform,” Thumu says. “We want to make sure we show you the right products in the right order for the right customer.”
Thumu turned to search software provider Elastic and its distributed search engine Elasticsearch to help it do just that. A distributed search engine is a decentralized search engine that allows companies to better handle large amounts of data.
“We leaned on Elastic’s back-end platform to tie into the huge spike that we see during this time of year,” he says. “We wanted to leverage closed-loop data to improve the product finding experience. Elastic made it extremely easy for us to index and organize the data.”
Elastic vice president of marketing Jeff Yoshimura says a distributed search engine is built to handle large amounts of data and searches, something Fanatics needed given the more than 300 online stores it operates.
Elasticsearch is an open-source tool, which means retailers can customize it to their needs. For instance, Fanatics integrated its own internal search data with Elastic’s search platform to make for a speedier search experience for Fanatics customers running searches, Thumu says, a process that took roughly four months from start to finish.
In testing, shoppers found the products they were searching for 10-15% faster, on average, than they could on Fanatics’ old system because better indexing meant shoppers were shown more relevant results.
Better search results led to a “few point” increase in the conversion rate for shoppers who used the site search tool, Thumu says. “At the core of it, it’s a conversion metric for us,” he says. Thumu declined to further specify just how much conversion improved.
Yoshimura says Elastic charges its clients an annual subscription fee to use its software depending on the amount of data that the client needs to index. He declined to provide specific price ranges.