Keith Anderson: Besides exactly matching products, one of the fundamental, but not obvious ways we add value is through the matching of products that are not exactly identical. There is just one example—private label products, which are the products that the retailer sells under their own label or a private brand are often positioned as national brand equivalent, but they’re not exactly the same product. From a data perspective, matching those products is possible with a unique identifier like UPC. You have to have fairly sophisticated matching algorithms that either use product characteristics or visual matching that are associated with the products.
Some of the other companies we consider peers in the industry put a lot of the burden of that matching on their customers to the extent that we can automate that to enable analysis of what we call “like-for-like comparisons”. It’s a comparable product but not identical. It becomes instrumental in understanding the architecture of your assortment in a specific category versus competition, particularly as it relates to your own private label or brands that you may have chosen to carry exclusively and that your competitor doesn’t carry. Those are two very fundamental ways that retailers use our data in itself. A growing number of our retail customers do integrate our data directly into one of the price optimization engines. It’s not the only data set on any level that influences the prices that they set, but it’s definitely one of the critical inputs.
Sramana Mitra: I’m going to ask you to give me some macro trends. What’s happening as far as pricing on the Internet goes? What are the big trends?
Keith Anderson: It’s an interesting time. It’s definitely a cat-and-mouse game between shoppers and retailers as well as retailers and manufacturers. It’s all driven by increasing transparency. Some of the pressure is put on different retailers to manage their profitability and the brands to manage their brand equity and their margins. You may have experienced these tactics as a shopper yourself, especially at higher price points for discretionary goods like televisions, which are heavily researched categories online. Shoppers tend to do four to five visits before they convert to a buyer.