The e-retailer spends at least 50% of its monthly display ad budget on the highly targeted, data-driven—and often cheap—ad placements using programmatic platforms.
Online retailers can dazzle shoppers with site graphics, awe them with a bountiful product selection and impress them with recommendations that seem to read minds and souls. But if those pages load so slowly that consumers are left tapping their fingers on desks while waiting for information to appear, all that brilliance, and all the revenue it is designed to capture, can be for naught. Patience can quickly expire as frustration drives those potential customers to competing web sites.
Hoping to give online merchants a detailed assessment of the harm caused by slowly loading pages, Gomez, the web performance division of software firm Compuware Corp., has launched a tool that measures the revenue lost when consumers give up because they cannot shop quickly enough. Online merchants can use the service for free at www.compuware.com/fastcalc.
"Consumers expect immediate response regardless of where in the world they might be accessing a site from, what sort of device they are using and whose networks they might be on," says Bob Paul, Compuware’s president and chief operating officer. "Competitors are a second away, so organizations can`t give customers any excuse to click away, as this tool clearly illustrates."
Chief Technology Officer
The Compuware Financial Analysis Support Tool, or F.A.S.T. Calculator, is a web site with five fields into which online retailers input information about their sites. The first box asks for the industry the retailer is in (operators of travel and media sites have seven fields to populate) followed by questions about the number of monthly visitors, the percentage who make a purchase, the average transaction value and, finally, how many seconds it takes the home page to load. "Any retailer who`s been in a business a few months should have this information," says Imad Mouline, chief technology officer for Gomez.
After entering the information, the retailer clicks the "calculate" button at the bottom, which produces an estimate for the revenue missed monthly because of the page-load time. A chart also appears at the right demonstrating how the lost revenue grows over 12 months. Retailers can click another button for a full report, which requires the entry of address, phone and email information.
"The idea is simply to let the retailer know how much more they can make based on basic assumptions," Mouline says. "Instead of trying to just come up with some type of formula, we actually decided to look at real performance and data out there in the wild."
The data underlying the tool came from page response times and abandonment rates that Gomez collected from more than 150 web sites, 34,000 web pages and 150 million page views between Nov. 10 and Dec. 10. Gomez will continue to monitor web sites for the same information so the data that supports the calculator tool remains up to date.
A recent study conducted by Equation Research on behalf of Gomez illustrates how slow page load times or malfunctions during payment processing chase away consumers. The findings, based on a survey of 1,500 online shoppers conducted between Dec. 16 and Dec. 22, focused on peak shopping times such as Christmas and Valentine`s Day. 78% of survey respondents reported they had gone to competing web sites because of poor performance. Of those, 47% reported leaving a web site with a negative view of the retailer’s brand.
In fact, retailers who use the Compuware F.A.S.T. Calculator tool and receive a report will learn that the consequences of poorly performing web pages go beyond immediate revenue loss, especially as consumers stand ready to spread bad experiences via tweets on Twitter, complaints posted to other social networking sites or blogs and old fashioned word-of-mouth, says Eric Kushner, Compuware’s marketing strategy director. "There is a long-term revenue loss consideration," he says.
Reasons for poorly performing web sites vary. Insufficient server capacity can certainly slow web pages, but it is also possible for a web page to work well in, say, Boston but take 10 seconds to load for consumers located in San Jose, Mouline says. Additionally, the tendency of retailers to outsource parts of their web operation to third parties can add complications. For instance, a retailer might pay a third-party vendor to design and operate the ratings-and-reviews engine on a site, and the vendor might be responsible for slow page-loading time. "That composite nature of online retailing is increasing," Mouline says.
That also increases the risks of losing customers, because all the different parts of a retailing site are assembled in the shopper`s browser, not in the retailer`s data center. "The retailer`s IT department might see things humming along, and not see the problems the consumer does," Kushner says. “There are a lot of moving parts, so testing page performance from a consumer perspective is essential.”
After all, he adds, consumers do not care who is at fault for pages that take too long to load. And at the same time, consumers, who are becoming ever more accustomed to online shopping, are expecting more from retail web pages. "Consumers are expecting faster and more relevant web sites, and at the same time expecting richer experiences," Mouline says. "Retailers have to keep up."