eBay Drives Business with Analytics

With more than 97 million active users globally, eBay is the world's largest online marketplace, where practically anyone can buy and sell practically anything. Founded in 1995, eBay connects individual buyers and sellers, as well as small businesses—with the staggering total value of goods sold on eBay at $62 billion – more than $2,000 every second. At the 2011 U.S. Tableau Customer Conference, David Stone, senior manager of analytics platform at eBay, discussed with us how Tableau is helping eBay “use analytics to drive the site.”

Tableau: How do you use Tableau at eBay?
David: We use analytics to drive the site. You can’t stand in an eBay store and watch customers walk around. So, the visibility you have is in the web logs coming off of eBay.com. And so by looking at those, we not only see what customers are doing, we see more than a regular retailer would see.
   
Tableau: How much data are we talking about?
David: The data is five terabytes a day that we load to our data warehouse to do the analytics. I think the number is 18 terabytes is the Library of Congress. So, we do that every three days, basically. We’re over two trillion rows' worth of data in a single table in the logs.
   
Tableau: What was going on when you started to explore Tableau?
David: I was doing a lot of back-end database stuff. And it felt like we had a lot of data and there was an opportunity to have more applications, to use that data more. And so I worked with some guys that were doing BI solutions for end users. And as we talked to these end users, they were asking for a lot more than just another report. They wanted their own data to a large degree. They wanted a democracy of the data, this sort of thing.
   
Tableau: How did Tableau help you to make data at eBay democratic?
David:

I think that equated to self-service and there's a strength that Tableau has on that self-service side. Obviously, there's visualization, you know, it was born with that in mind. But the strength from a self-service standpoint was we could give people their own database space. They could go to our enterprise data warehouse and insert select into their own set of tables. They could experiment with the data. And then they needed a front-end; they needed an alternative, really, to Excel.

And so that's where Tableau started kicking in. The exciting part was just the uptake. It was easy for someone to get online, and they could start learning right away. In a day or two, someone was proficient and could get more done in Tableau than they could get done in Excel.

   
Tableau: How do you feel Tableau compares to Excel?
David: A couple of things I was noticing when users were limited to Excel and the million-row limit, it caused them to look at the top three categories instead of the top 40,000 categories, and there's so much less data. And so at eBay that's important because there is more action out in the long tail accumulated than there is in the top three, for instance.
   
Tableau: What’s your implementation been like at eBay?
David: In the beginning, we hoped that anyone who was using Excel would have the option to also have Tableau on their desktop if they wanted, and so we're just coming to have that be possible. So, we had 300 licenses, and now we've increased that to this pool where basically anyone at eBay and marketplaces who wants to get a license and try it out can go do that. It’s open for anyone who has an idea and wants to be able to get some results and test out the data they see with the end users and then iterate a couple of times. And then from there maybe get that information into something that's more enterprise industrial strength.