How testing builds a better home page
Producing the most successful web pages takes testing and optimization, but the best results require more than what retailers see in most simple A/B tests. “Many online marketers run home page tests and results always come back 50/50,” says James Roche, president of on-demand testing services provider Offermatica.
While most online retailers make their initial foray into live testing with just two versions of the home page, boosting the scope of testing can produce lifts in conversion and ROI “almost overnight,” Roche says. Using multivariate testing, Roche says he has seen marketer clients gain revenue-boosting test results within a few weeks.
Among his tips for online marketers launching home page testing programs, Roche advises marketers first off to avoid making too many changes at once. “Instead of changing everything in version B of a home page test, change just one element at a time,” he says. “If you change the headline, layout, graphics, copy, product or promotions all at once, you won`t know which changes drove the increase or decrease in traffic or conversion.”
Roche also advises online marketers testing home pages to measure whether the changes improved visitor penetration deeper into the site, moving them toward the shopping cart, rather in terms of direct conversion off the home page. “The goal of a carefully designed home page is to drive visitors farther into the site, not to convert them into buyers on the spot,” he says.
Roche also notes that because a more consistent test population provides more rigorous testing results, marketers should show different pages to different, defined audiences, such as visitors from specific campaigns, or visitors new to the site. “Target your tests,” he says. “Not everyone visiting your home page got there in the same way. Some typed in your URL, some came from a paid search link, and others followed an affiliate link.”
Back...
