Live from eTail 2004: “We Can’t Serve Everybody”

Hollywood, FL–After ZelnickMedia bought Lillian Vernon Corp. in July 2003, the parent firm has made it clear from the get-go that its goal for the 53-year-old gifts and housewares cataloger is to build the Lillian Vernon brand while maximizing the company’s direct channels—particularly online. During the eTail 2004 conference here, Jonathan Shapiro, the ZelnickMedia consultant installed earlier this year as president of Lillian Vernon, said that the company’s top priority in its efforts to build online revenue is to focus on its targeted customer base—but by using variations of traditional direct marketing tools.

“We can’t serve everybody,” said Shapiro, a former chief strategy officer for online advertising firm DoubleClick. “So we had to decide who to make the site easy for. We prioritize our site by economic value and potential near-term value. Rather than lifetime value where you look at 10 years out, we take a 12- to 24-month look in near-term value.”

The advantage of near-term value over lifetime value online is “you need only two-three years of transaction history,” Shapiro said, “so you probably have the data available that you need. And since you’re only projecting 12-24 months out, it’s easier to more accurately project, and you can apply it to more circumstances.”

Once you accumulate near-term data, “you’re still not done,” Shapiro said. “You still want to focus on customers who have the highest potential. And all the co-op databases, such as I-Behavior, NextAction, Abacus, can show you how much customers are spending with you and with the competition. And that gives us the opportunity to capture our share of the market.”

In improving its online experience, Lillian Vernon uses attitudinal segmentation, which gauges channel and media preferences, and the cataloger prioritizes those segments by economic value proposal, Shapiro said. What’s more, Lillian Vernon is asking why customers are coming to its site—for casual shopping, entertainment, or a problem resolution? “We want to figure out what customers are trying to accomplish when visiting our site,” he said. “We can tell down to the square inch in our print catalog how every piece of paper in that catalog has done. We need to be able to do the same thing online.”

To improve the online experience for its customers, who are women ages 35-55, Lillian Vernon conducted an extensive customer survey last year that took the average recipient 20 minutes to complete. “Testing is more than just measurement,” Shapiro said. “You need hardcore statisticians on staff or need to outsource so that you can create, design, and run statistically significant projectable tests. I want to be able to understand not only what happened, but why it happened.”

To do that, testing and customer research need to be hypothesis-driven and continuous, Shapiro said. “You can’t get caught in analysis paralysis,” he said. “So you form hypotheses. If, for instance, you want to test the effectiveness of different colors for the navigation bar, you test a high-value hypothesis first, and it has to be continuous.”