23% of e-retail transactions on Thanksgiving and Black Friday came from mobile devices, according to payments security firm ThreatMetrix. However, 15.5% of retailers say ...
IRCE 2011 Report: An online wine retailer goes granular with marketing data
E-retailers can squeeze more out of pay-per-click budgets if they analyze performance data.
If e-retailers take a blanket approach to pay-per-click campaigns, they’ll soon find they are underpaying for good performers and overpaying for lousy ones, said Michael Mothner, founder and CEO of search marketing firm Wpromote, this morning during the Internet Retailer Conference & Exhibition in San Diego. “If you do that, you end up with inefficiency, and in the end you have a mediocre campaign,” he said. “Wallowing in mediocrity is not a fun place to be.”
The solution, he said, is for e-retailers to invest in pay-per-click based on a deep understanding of performance analytics. That’s what Wine.com tries to do, said Cam Fortin, director of business development at Wine.com Inc., No. 239 in Internet Retailer’s Top 500 Guide, during the session, entitled “Breathing new life into old PPC campaigns.”
Fortin says half of new customers to Wine.com come to the site through paid search, which makes understanding the dynamics of its search investments important. For example, the e-retailer bids differently for generic search terms that are likely to generate interest from potential new Wine.com customers, rather than brand name terms that are used by existing customers. “We’ll spend more money to get someone new in the door rather than on someone we’ve sold to before,” Fortin said.
New customers, for example, will search terms like “buy wine,” “wine delivery,” or “top rated red wine.” Existing customers will search for terms like “Wine.com” or “Cristalano Brut Cava.”
E-retailers need to analyze data across all pay-per-click investments, such as each web site where a display ad renders, Mothner said, which he admitted can be exhausting. Using the analytics provided by the search engines or ad networks helps. “Let Google work its magic,” he said, “then apply really smart analytical thinking to it.”
For Wine.com, data showed variations in how and when consumers accessed Wine.com via an iPad tablet computer. Typically, Wine.com lowered its paid search spending on evenings and weekends because fewer Wine.com customers shopped during those periods. A deep dive into the analytics, however, showed that iPad users shopped more frequently during those times. Wine.com raised its investment in paid search for ads exposed only to iPad users. “Being able to think about delivering a unique message based on the device is something we never really thought about before,” Mothner said, “but it opens us up to a whole new way of thinking about customized messages.”
To Mothner and Fortin, performance data reveals trends that greatly influence how to allot a budget. For example, consumers who use a desktop computer to find Wine.com cost less per click, but convert at a 6.49% rate. Whereas those that access Wine.com via iPad cost more to get but convert at a 9.54 % rate. Wine.com now runs different campaigns for each segment. “When you have the opportunity to do so, look at your data as granularly as possible,” Mothner said.