In 2005, the company began a 12-month evaluation of technology vendors, who were given an overview of the candy maker's manufacturing process. Supplied with a 10-flavor mix of Jelly Belly jelly beans, the vendors were asked to create and run a demo of their solution using Jelly Belly's information. The company eventually whittled it down to three final candidates and settled on St. Paul, MN-based Lawson Software, and implemented its M3 7.1 suite of applications.
"We chose Lawson because most of the vendors did not have an ERP solution that was flexible enough to handle our processes," says Schader. "Lawson demonstrated that its solutions could address challenges specific to our business and they were the only vendor that would implement the solution with their own consultants--the expense is not in the software, it's in the actual implementation."
The company implemented Lawson Financial Management, Supply Chain Management and Business Performance Warehouse, along with planning applications including Demand Planner and Supply Chain Orders.
The project began in March 2006 and the first hurdle was the cleaning up of Jelly Belly's master data. "With M3, all the data has to be entered correctly into the system to work. We did a data conversion from the old system into the new system and one of the things we learned was that our old system didn't enforce a lot of data integrity. Every master file had issues," says Schader. "We were finding so many inconsistencies in our customer master data, item master and vendor master that we had to put one person in charge of each master file to get everything corrected."
As part of the system rollout, Jelly Belly developed a change management plan that included employee training and ongoing communications to keep staff updated on project milestones. The company also supported employee buy-in by enlisting staff from various departments to conduct system testing and provide feedback throughout the project.
There were challenges along the way. During the implementation, Jelly Belly acquired another company and merged it into its Fairfield, CA, facility. Then, just as the system was ready to go live, Lawson upgraded M3 with a new user interface called Smart Client. Smart Client is an application that sits on a central server and delivers an interface that allows user-specific customization and can support different types of data displays including documents, graphics and video, replacing the tab-based data entry form style common to ERP systems.
"Smart Client is a much better interface than the previous version," says Rosman. "It allows a user to multi-task. It also makes it easier to train new people on the system. Even though we had training prior to this, everybody needed to be up to speed as quickly as possible, and Smart Client certainly helped."
Then there was the WMS. "M3 has some basic warehouse management functions, but it didn't have wireless capabilities. We were using over 100 handhelds on the floor, so Lawson brought in one of their partners, Data Systems International (DSI), to develop applications that allowed our handhelds to communicate with the system," says Rosman. "That was a major cost savings because we were able to use our existing legacy handhelds and did not have to change the proprietary wireless handheld network."
The implementation was completed last October--and the benefits were almost immediate. Jelly Belly has greater transparency to its supply chain. "The access to real-time transaction information had significantly improved coordination between departments," says Schader. "This increased visibility has extended into the financial analysis of our manufacturing cost accounting. We are now giving feedback to manufacturing on a daily, rather than monthly basis, so process adjustments can be made and variances can be addressed."
The company has greatly improved its demand planning and forecasting capabilities. Sales managers used to forecast every item on a monthly basis and inaccurate forecasting wreaked havoc on its supply chain. The system now calculates the forecast for many items with a sales history, leaving the sales managers more time to focus on selling.
"We're making significant progress," says Schader. "The quality of the data is markedly improved. We have a consistent sales history for many items and can statistically calculate what the demand should be. The sales staff can now focus on the large orders that move the needle, instead of them telling us they need five more cases."