Companies fall into a trap when they collect Web site statistics.
They report statistics, not analyze them, said Matthew Bailey, president of SiteLogic Marketing.
Bailey, addressing an audience at SES NY, said a company must first know what its goals are, and then identify key performance indicators to track such as sales leads or downloads. “Then focus on a goal and ask how are we reaching that goal,” he said. “Start asking those kinds of questions rather than how many unique visitors we have each month.”
Bailey advocates the use of “segmentation” to track a company’s Web site performance. What exactly does that have to do with Star Trek?
To make his point, Bailey referred to the original Star Trek television series, which was aired in the 1960s. He said the starship Enterprise started with a crew of 430 people. By year five, 59 crew members had died; 43 of whom, or 73 percent were wearing red shirts. The others were in yellow or blue shirts. “We have a little bit of knowledge, but we don’t have any indicator what we can do with it,” he said.
After a little more digging, Bailey said one would learn that 57.5 percent of the crew members who accompanied Captain Kirk on a mission died. If they accompanied the captain when he met an alien woman, the survival rate soared to 84 percent. Comparing the circumstances when red shirt crew died is an example of segmentation, he said, advising the audience to apply the concept to Web site analytics. For instance, a site that sells both MP3 players and digital cameras should track results for each product to gain additional insights on what works and doesn’t work.
“Segmenting your visitors, where they came from, what they are looking for…trying to find the goals and motivations of your visitors and segmenting them to get a little bit of intelligence,” he said.
Bailey says he’s surprised by how many companies still tracking the number of “hits” they receive. “It’s so ’90s,” he says. “We have to get beyond reporting these types of numbers…it drives me insane,” he said, imploring the audience to dig deeper into the data that’s collected.