In its second-largest acquisition, Amazon buys the company for $970 million.
(Page 2 of 3)
"It's really caused us to refocus our efforts on making sure all our products are lined up correctly," he says. Because of the complexity of feeding data in the correct format to nearly 10 comparison shopping sites, eHobbies uses technology provider Your Store Wizards to manage those feeds.
Retailers also must realize that the data they feed to Google Product Search determines whether a retailer shows up, and what information consumers see, in several parts of the Google search results page, says Steven Roth, vice president of performance marketing at Channel Intelligence, which facilitates retailer sales through shopping sites and web marketplaces.
Besides feeding into the OneBox, the array of product images and data in natural results, Google also uses that product data for Product Extension ads that appear above the natural listings and display a plus sign that lets consumers see more products from the advertiser listed. A retailer can specify which products, such as best sellers, appear for each keyword; if the retailer does not specify the products, Google makes its own selection, Roth says.
Google also uses the same product data feed to populate the Product Listing Ads it's testing. Those ads appear at the top of the right column with product images and about 40 or so characters of text. Retailers should front-load their product names with the key terms they want consumers to see in the short description area of those ads, Roth says. Unlike standard Google search ads, which charge per click, retailers only pay a fee for Google Product Listing ads when the consumer makes a purchase.
Providing Google plenty of product data is particularly important for products like furniture and apparel that are not as easy to precisely describe as a digital camera that has a model number, says Wingo of ChannelAdvisor. For instance, while the product title of an item might be "outdoor rattan chair," the attributes should include terms like "wicker" and "patio" that consumers might search for.
And getting the feed right is important for all comparison shopping sites, not just Google's. Altrec.com Inc., a retailer of outdoor gear and apparel, improved traffic 150% from shopping sites by moving the designation "men's" or "women's" from the last word in a product title to the first.
Most shopping engines pay close attention to the last word in a product title, because that's often the name of the item, such as in the phrase "blue men's coat." Putting the product name at the end boosted results, says Derek Kershner, search marketing analyst at Altrec, which uses Mercent Corp. to manage its feeds to some 25 shopping engines.
Time for a change
The challenge from Google is forcing established comparison shopping engines to respond with new features and policies. And newcomers are presenting additional options to e-retailers.
Among the established players, NexTag has developed an ROI-tracking system that promotes items that are converting well for retailers and provides them with feedback on those not meeting their targets, says Mark Bradley, senior vice president of NexTag.com Inc.
PriceGrabber is testing a system that lets retailers bid above the rate card, which lets them grab top spots for items selling well. Shopping.com has introduced a tiered pricing structure, so that retailers pay less per click for lower-priced items.
That's aimed at retailers' biggest complaint about pay-per-click shopping sites: They can get burned if too many clicks don't turn into sales.
"The most important thing is to have a program in place to suppress products that are getting too popular," says Rick Backus, co-founder of CPCStrategy, which manages comparison shopping feeds for retailers, typically charging a flat fee in the range of $1,000 to $2,000 per month. Comparison shopping sites typically move to the top of popular searches items that generate a lot of clicks—which means more revenue for them.
The result is that a less popular product is likely to be seen only by consumers who put in search terms closely related to the product. But a popular product is viewed by many consumers, including some not interested in that product, leading to lower conversion rates, "After 50 clicks without a sale, remove a product," Backus says.
Addition by subtraction
Many retailers follow Backus' advice, sending shopping sites only items converting well.
Dillon Chemical LLC uses software from SmartFeed Inc. that shows graphically which products are making money on each shopping site, and which are costing the retailer of cleaning and janitorial supplies more than it's earning.
Products show up in green on the SmartFeed dashboard if they are hitting the goals owner Dan Dillon sets, while others are displayed in red. Dillon, which sells at CleanItSupply.com and DillonChem.com, paid SmartFeed about $500 for a customized add that pulls in Dillon's cost for each product.
That way, Dillon says, he can see for each product how many clicks it's generating, the average cost per click, dollar sales from those clicks and his return on investment. Each day he sorts his products by return on investment, starts with the products faring the worst and decides which ones should be pulled, or whether an enhanced photo or description could improve results for a particular SKU.
Using SmartFeed, for which he pays $100 per month, Dillon has been feeding products to about 10 shopping sites this year, typically sending about 3,000 of the 6,600 SKUs on his sites. The results: more than $53,000 in sales for an expenditure of $4,000 in click fees. "That spread works for us," Dillon says.
Traffic to the Dillon sites has also gone up 70% since December, which the retailer attributes to the comparison shopping engines. And Dillon is not the only e-retailer to find that shopping sites can be a good way to acquire new customers. Twenty-two percent of e-retailers ranked marketing through such sites as among their most effective acquisition tactics in a survey last year by Forrester Research Inc. and e-retail trade group Shop.org. That put comparison sites fifth in acquisition effectiveness after paid and natural search, affiliates and catalogs.