The Top 500 retailer buys Campus Deals, which offers mobile coupons to college students.
Refereeing a Redesign
Each department has its own views on how a site should be designed. Retailers score with redesigns when they let customers make the calls.
A retail web site redesign is akin to a bricks-and-mortar shop hiring an interior designer to redecorate a retail store. There’s a lot involved, from a new color palette to refreshed navigation and site search, to new product detail pages and web site architecture. It’s exciting. It’s a big deal. And, too often, retailers say, it’s more complicated than it should be.
Many online merchants who have been through an e-commerce redesign express the same woes. Everyone-from technology staff to upper management to the marketing department-wants a say in how the site should look and function, and many times, those opinions clash.
On top of that, there are budgets and deadlines to consider. Pulling off a successful site redesign is not easy, but it can be done. And retailers who’ve done it offer a few tips: Put the customer first, establish clear objectives in the beginning, and designate a redesign leader that no one, not even the CEO, can trump.
Testing, testing, 1, 2, 3
Katie Fernands, Internet marketing manager at Colonial Candle, says staff had many different ideas when the retailer set out to make some significant changes to several web pages of its retail site. To help it make decisions, the retailer turned to its customers.
“What designers know about web page design for search engine optimization and what marketers want are sometimes at odds,” Fernands says “So when we make changes, we test different versions and then the winner gets to stay. We let the consumer decide.”
For example, the retailer of candles and home fragrances recently used the free Google Web Site Optimizer tool to test three versions of a landing page. Over several months, about 100,000 visitors were sent to the various versions. Colonial Candle experimented with fonts, images, graphics sizes and calls to action, and set up the optimizer so that traffic was equally divided among the pages.
The results surprised both the marketing and design teams. For example, while web design 101 teaches that consumers don’t want to scroll down web pages, guess which page customers favored: the longest one.
In fact, that lengthy page garnered 20% more conversions than the original and generated incremental revenue of $20,000 over a four-week test period. Fernands says her team will continue to test as it redesigns other parts of the site, starting with its 10 most-visited pages.
Observing consumer behavior is one way to figure out what shoppers want; another is to simply ask, says Colin Campbell, founder of Lavish & Lime, a retailer of eco-friendly and natural products. The retailer used a free online survey tool called 4Q from web analytics company iPerceptions Inc. when it started doing research for its redesigned site, which went live last month.
The poll asked consumers why they came to the site and if they achieved their objectives. When consumers said they didn’t accomplish their goals, the survey asked why. Campbell says those answers told his company a lot, and helped him prioritize his redesign budget.
“Much of it concerned silly, easy-to-add things that we didn’t think of,” Campbell says. “People wanted to know the ages that were appropriate for products.”
And, because the site attracts environmentally conscious consumers, Campbell says shoppers wanted additional information about distributors and how goods were made. For example, many consumers wanted to know for certain that parts were not made in China, and that products did not contain Bisphenol A, commonly referred to as BPA, which is used in many plastics.
A retailer’s own employees also can offer useful insights. Manjula Higginbotham, support and implementation manager at e-commerce software company AmeriCommerce, as well as part-time e-commerce manager at children’s furniture e-retailer Pippin McGee, likes to talk with staff in small groups when planning a redesign.
After she gathers employee input, she sees if customers feel the same by conducting online surveys and employing web analytics. When chatting with employees, she says she’s careful to get everyone’s opinion, even that soft-spoken customer rep who is typically the last one to speak up.
“I will leave out of some gatherings individuals who I know are strong in their opinions, or managers and team leaders who might intimidate others,” Higginbotham says. “I take notes and then I run a customer survey.” Higginbotham says customer data is a great way to solve internal disagreements about design. Sometimes, she adds, popular theories fall flat.
When Pippin McGee redesigned its site in April 2007, for example, Higginbotham says most employees felt strongly that the retailer needed to overhaul its navigation. The 3% to 5% of site visitors surveyed over a four-month period, however, didn’t agree.
“About 60% of the 400 to 600 respondents commented on how absolutely wonderful and clear navigation was,” Higginbotham says. “We were considering spending thousands to completely change it. I didn’t think customers would so happily talk to us. I was amazed at the simplicity of getting key information. It’s really easy: Listen to your customers, they love to talk to you.”
Higginbotham also says such surveys can save current programs that management may want to cut. For example, Pippin McGee pays a monthly fee to display high-resolution images, and overwhelming positive feedback about the pictures helped justify the cost, she says.
Making the cut
Few retailers can afford to make all the changes they would like on a site, and Higginbotham uses analytics to set priorities. Pages with the highest bounce rates and product pages for the most searched-for wares move to the top of the to-do list, she says. “Analytics not only point out the areas to implement changes in, but also help with working out what should be prioritized if everything cannot be implemented,” Higginbotham says.
While data from many consumers is one way to prioritize, another is to closely watch how a small number of individuals navigate a site.
Online jewelry, tableware and collectables retailer Lenox used a program called Listening Labs from vendor Creative Good to watch consumers search the web for the types of products it sells, see if they landed on the Lenox retail site, and determine if they could easily find what they wanted once they arrived. Lenox watched 16 consumer sessions.