Segmentation at eBay: Reaping the Rewards

Segmentation has traditionally been a key component of eBay's business strategy. Through segmentation research, eBay has continuously identified opportunities to introduce new services and products based on an in-depth understanding of customer needs. What follows are 5 questions with Meg Sloan, Director - Market Research at eBay, about the benefits of effective segmentation.

CMB: Over the last few years, you have done a lot of segmentation work. Why is segmentation so important to eBay?

Meg: Segmentation is important to eBay for a variety of reasons. eBay is a very big ecosystem - in the U.S. alone we have around 26 million active users of our Web site. Realizing that nobody who comes to eBay is the same, we need to have an understanding of how our customers are different and then tailor the experience to them. Precisely because eBay is such a huge ecosystem, we recognize that we cannot be successful trying to be everything to everyone. Segmentation, in combination with our own internal data, allows us to understand and think about what we are for whom, how we are going to move the business forward, and what we choose to do in terms of changes to the site or the business model in general.

Segmentation is also very important to our sellers. We recognize that as a marketplace our sellers are the people who deliver great experiences to buyers. So at our ecommerce forumin 2007 we shared our segmentation data with our top sellers, in return empowering them to then deliver experiences that our customers really value.

CMB: Can you describe how segmentation has evolved at eBay over the past few years?

Meg: Certainly. Segmentation has definitely evolved over the past few years, and it will be a constant evolution for us as we get more and more sophisticated in terms of how we use segmentation. If we go back to 2003, we really had a single basis for segmentation, focusing mostly on behavioral-based, survey-based segmentation, as well as industry standards like VALVE and RFM - and because of this its shelf-life was limited; it just wasn't stable over time because the way people use the web is ever-changing . Over the years, we have recognized that one basis for segmentation is not enough of a view for such a complex business in order for us to make the segmentation actionable. So one of the things we have done with Chadwick Martin Bailey is to understand segmentation across multiple bases. The goal now is to create the most relevant and actionable segmentation for our business by combining internal and external data (such as behavior, attitude and value) with multiple survey bases.  
 
CMB: How has segmentation changed the way that eBay currently conducts business? 
 
Meg: It has definitely changed the way that we think about our customers. A few years ago we thought about them as a more homogenous set. We thought about what people came to eBay for and what they did - but we did not have a complex enough understanding of who our customers are and what motivates their purchase and what kind of attitudes people have about shopping in general.  Now we have a much deeper understanding of our customers and this knowledge informs many aspects of our business, ranging from thinking about new ideas, proposing new initiatives and implementing service standards. The majority of our employees, and especially our most senior executives, also know what segment they personally belong to and will frequently reference this information when putting forth their thoughts.
 
One significant result of the segmentation work with Chadwick Martin Bailey is the understanding that a large portion of our customers absolutely love to shop and view the experience as a form of entertainment. Our site will increasingly cater to this group of shoppers who view shopping as fun and love to get great deals. Customers can get a preview of things that are coming at www.eBay.com/playground, which reflects our understanding that for many of our customers shopping is a fun hobby. We have introduced a new feature called "DealFinder" that allows people to find great deals that are ending now on really popular items - something else that segmentation clearly tells us that our target customer values.
 
CMB: eBay has a wonderfully large database of customers - and one of the big benefits of a large database is the ability to map people in the database to membership in a particular segment. How are you using the information in the database today and how do you plan to use this information in the future?
 
Meg: One of our goals was to have the ability to take our attitudinal segments and apply these, with a scoring, to our internal data. I am happy to say that with the help of Chadwick Martin Bailey and great analysts at eBay we were able to get this accomplished. Now we are beginning to incorporate the analysis into a variety of different aspects of our business. For example, we will be able to analyze our customer support data including the attitudinal segments, so that we understand the type of customer who calls, the type of problem that he or she is experiencing, the resolution that we provided and how this relates to the customer profile in our segmentation. In the end we will be able to tailor our response to our customers in a way that strikes a chord with them based on their specific shopping profile. 
 
CMB: If you could think ahead a couple of years, and reflect on the segmentation legacy at eBay - what would your biographers say about the role of segmentation at eBay and, in particular, your role - and the role of Chadwick Martin Bailey - in driving it? 
 
Meg: The one thing we have not touched on and that I am very proud of is the approach that we take to segmentation. From the beginning, we took a "do it right vs. do it fast" approach. We started with the end objectives (i.e., specific business decisions we wanted to make) in mind and certainly calculated every decision based on that. We also approached the project in a very global way - with a global team - making decisions that made sense for our biggest markets. Because of the approach we were able to expand segmentation beyond the original set of 3 countries to a set of 8 countries.
 
The success of the approach has meant that we could apply it not only to buyers, but also to the other major components of our company: sellers and the motors business. It has taken us about a year to feel the full impact of the segmentation and the scoring of the data warehouse - and we are only beginning to feel the true impact. The results are telling though because our approach is careful, decision based and global.  And, last but not least, we have a lot of fun on the way, which is very much the eBay way of doing things.
 
Meg, Thank You very much!

Chadwick Martin Bailey