The call for an audit of Facebook’s metrics comes a week after the social network acknowledged inflating its video metrics.
(Page 2 of 2)
Often, the interviews uncover information that web site operators would not know otherwise. For instance, when ForeSee interviewed customers of a government site, it learned that repeat users gave the site an acceptable score, yet first-time users gave it extremely low ratings. Further probing revealed that first-time users had a difficult time knowing how to find information on the site. “The site was organized the way governments are organized,” Freed says. “It wasn’t organized by the user’s orientation.”
The operators changed the home page to ask users to identify which group they belonged to-individual citizen, business or government employee-then presented information organized by the user. Satisfaction scores from first-time users doubled immediately, Freed says. “They were organizing by functionality and not by user,” he says. “Many retailers do the same thing.”
Similarly, a retailer client of ForeSee had scheduled a re-design of its web site targeting the look and feel of the site. Managers felt the site hadn’t kept up with industry leaders in design elements. But in conducting customer satisfaction surveys, ForeSee identified a different problem: site functionality. Customers complained they had a hard time finding products and that once they did find products, it was difficult to find information about the products. “They re-directed their re-design from look and feel to functionality,” Freed says, “and created a site that was easier to navigate into the details.”
However retailers come to their conclusions about what they need to change on a web site, one of the benefits of analytics is that it provides a solid foundation on which to argue for change. “One of the biggest problems when it comes to making changes at web sites is lack of confidence by managers,” Plymale says. “They have no way of knowing if they make the change if things will get better or worse.”
Analytics can provide managers with the data to back them up when they propose a fix-even if the fix would be politically unpopular within the organization, Plymale says. He cites the example of a retailer where customers were inadvertently putting multiple copies of the same product into their shopping carts, then becoming confused at checkout time. “The shopping cart software had won awards, so no one wanted to touch the award-winning software,” he says. “But that’s where the problems were coming from.” With the analytics data to back them up, managers changed the software and saw an increase in sales.
Plymale also points out that analytics generally will point out small changes that can be made and measured, which gives web site managers a way out if there is a problem. “People in large companies are looking for safe ways to drive changes,” he says. “This way, if something goes wrong, they can always scale it back.”
Using data from WebCriteria’s Site Analyst product, Outrigger plans to re-design its site with prices posted upfront. It expects higher customer satisfaction and more sales as a result.
The cost of analytics depends on volume and on whether a retailer buys a service hosted by a vendor or a software license to perform analytics in-house. On a hosted basis, a small retailer will pay about $2,000 a month and the largest retailers with high volumes could pay $30,000 to as high as $50,000 a month, says Berk of Jupiter Research. On a licensing basis, a retailer will pay $1,000 a year for an entry level program up to $15,000 a year for a top-of-the-line program, Berk says. At the high end, vendors also charge 18-22% in annual maintenance and upgrade fees and require the customer to purchase certain professional fees. While the prices may seem high, Berk notes, “There’s a lot of revenue involved with the large retailers.”
That’s doubly true when it’s a multi-channel retailer, looking to the web site not only for sales but also to drive sales into stores. And that may be the new frontier of web site analytics. “Whether you’re converting online shoppers to offline buyers is difficult to measure,” Freed of ForeSee Results says. “We look at future behavior, tying customers’ online experience to their offline behavior. Retailers are seeing this as an opportunity to know what their customers are doing in all channels.”