Now Can the Candy Man Sell on the Web?

It’s hard to execute a technology project effectively when most of the vendors who could help you are on death row.

That’s the situation the Jelly Belly Candy Company, the family-owned California candy maker, found itself in last year as it tried to expand its sales effort on the Web. Ryan Schader, director of corporate strategy, and Dan Rosman, IT director, were frustrated with the company’s inability to do e-commerce back in January 2001. Online sales did not amount to much. They were only 1% of its $140 million in annual sales. Jelly Belly’s e-commerce setup amounted to a single NT server running an out-of-the box Web store package. Its most remarkable attribute? The amount of hate mail it generated.

Even at 15,000 visits a day, the online store was seizing up, dropping customers’ connections as they shopped. It was a customer-service nightmare.

Obviously, the storefront wasn’t up to snuff. It did not take data from Jelly Belly’s warehouse management software. It couldn’t give customers a history of their orders or remember their preferences of flavors. It didn’t let customers ship purchases to more than one address—tough on corporate or family gift-giving. In short, it didn’t do much.

“The customers complained bitterly,” says Schader. “They said, ‘You have an enormous brand, and we expect more from you.’ “

Schader and Rosman went back to the drawing board. In a revamped site, they wanted the sales process on the site to be tied into its enterprise planning and warehouse management systems, so customers could see when jelly beans could be shipped. They wanted a customer portal that could be tailored to each customer, greeting them by name when they logged in. They wanted features that let them keep track of each customer’s buying habits so they could market to their tastes.

Plus, if they could take orders from customers directly, they could then take orders from distributors. At the time, the company mailed out 10,000 price lists a month to distributors and retailers, and took orders by fax.

Unfortunately, they began their search for help just as Web consulting firms such as MarchFirst and Razorfish were entering death spirals. “We didn’t want to be stuck with a vendor that wasn’t going to be around next year,” says Schader.

It didn’t help matters that no one at Jelly Belly was running the project full-time. The selection process dragged on for eight months. “The most important lesson from our experience is that companies should dedicate at least one resource or more to the project,” says Schader. In the end, the team came up with four choices: a joint offering from Accenture and Microsoft, and alternatives from ATG, BEA, and Intershop.

Even after selecting a winner—Accenture and Microsoft—the dot-bomb Grim Reaper wasn’t done. The Web design agency Jelly Belly chose to create the templates for the site—Clear Ink of Walnut Creek, Calif.—went under, though it was relaunched in January.

Speed finally arrived. Accenture and Microsoft did help Jelly Belly execute the first phase of its e-commerce strategy—a new site for selling to consumers—in a little more than two months. Now, the site is managed by Accenture and hosted by AT&T. Only three Jelly Belly people were involved in the technical rollout—two in testing and one part-time in development. And the pre-packaged software and hardware involved are priced at under $250,000.

The site was live on Dec. 17, in time for Jelly Belly to get some leverage out of a tie-in to the Warner Bros. film, Harry Potter and the Sorcerer’s Stone.

Daily Web sales are now up 35%, without an increase in traffic. Of course, Jelly Belly won’t see a real return until the phase of the project that ends up serving retailers and distributors is finished. But even that is 80% based on work for the first phase.

And the hate mail already is subsiding.