Wal-Mart Stores Inc. is working on ways to analyze its Web site in the same manner it does its thousands of stores nationwide.
The world's largest retailer is using Omniture Inc.'s SiteCatalyst hosted Web analytics service to study the relationship of merchandising to sales on the site at www.walmart.com.
“What they're trying to determine and influence is product conversion — what kind of merchandising translates to the largest conversion,” said John Mellor, vice president of marketing at Omniture, Orem, UT.
Wal-Mart's segmentation of its site is challenging from an analytics standpoint because this requires listing the same product in various digital shelves, or categories. For instance, a CD selling on the site may list under categories like “Music,” “Daily Special,” “Gifts Under $20” or as a cross-sell with CD players.
So it is easy to know the overall conversion of the CD on the site. But it is more useful to glean the conversion by category. That task, however, becomes more complex when trying to track hundreds of thousands of products in multiple categories that change weekly.
SiteCatalyst helps mine data in real time to quantify and visually reflect the effectiveness of Walmart.com and its marketing objectives. Accessed online and customized, it requires no hardware or software installation. SiteCatalyst users include Hewlett-Packard Co., Time Warner, Gannett Co., Maytag Corp., Microsoft Corp., VeriSign and Timberland Co.
Specific to Walmart.com, SiteCatalyst includes data points like historical site activity — what happened on the site — as well as actionable information — what is happening on the site. Its Custom Insight Architecture identifies factors about site visitors, generating customized reports on such online activity.
For Walmart.com, the two main issues under analysis are product placement or exposure and merchandising.
Walmart.com can ask a question like, “When a product is discounted more than 2 percent, does the site need to show a lifestyle image or can it use a simple product image since the item sells itself?” SiteCatalyst helps extract the answer.
Also, Walmart.com can analyze the same information on a campaign level. The online division, for example, may want to know how a product sells by category and by campaign. In other words, how does the CD sell when it is offered as a “Gift Under $20” and promoted via a 150-by-90-pixels banner on a particular affiliate network the week after Thanksgiving?
Correlating all this data for numerous items lets Walmart.com place a value on each campaign variable like “How well do the promotions work the week after Thanksgiving?” and “How well do my 150-by-90 banners convert?”
There are many takeaways for teams at Walmart.com, Mellor said.
For buyers, it answers the question, “How well does a particular CD sell, and should the buyers purchase more?” Merchandisers may learn how well products sell in “Gifts Under $20” promotions. Campaign managers find answers to, “How well do 150-by-90 banner ads convert on AOL.com?”
Finally, Walmart.com creative managers can ask, “Does the blue 150-by-90 banner ad convert better than the red?”
The goal is to deliver analysis in a way that makes the data understandable to Walmart.com executives with daily access to SiteCatalyst.
“We've given Walmart.com the ability to analyze their online world in a similar way that they analyze their offline world,” Mellor said.
