Nightmare on Main Street

Jul 01, 2001 9:30 PM  By

Best fits will meet 70%-80% of your needs before modification

Here’s an operations and fulfillment professional’s nightmare: After months of research and a mega-monetary investment, instead of providing improved efficiency, your new software system delivers limited functionality, unexpected costs, and a failure to meet users’ expectations.

Unfortunately, it’s a nightmare that comes true all too often. Unless you manage the selection process with great care, the system — any system, whether catalog management, warehouse management, e-commerce, customer relationship management, or multichannel — can turn out to be a less-than-ideal fit.

Sweet dreams

Here is a game plan for selecting a system that will let everyone sleep peacefully:

Assemble your project team before you open your first brochure. The team should include representatives from all of the departments in your company who will use the system, so that the selection is not just the IT department’s take on the best platform. The team should include a delegate from top management who will speak for overall company policy.

This keeps the lines of communication open between the project team and the company’s top echelon, and it signals to other team members that system selection is a high priority. Name a project coordinator to keep all players on schedule and to serve as a liaison to vendors. Finally, appoint a steering committee to set the project’s direction and review its progress at least on a monthly basis.

Draft a company profile. Your company’s system requirements are multi-faceted and may change between the time the selection study begins and when the system goes live. It’s wise to draft a profile of strategic development prospects for the next few years.

The number of orders, size of the customer file, number of online transactions, number of products and SKUs, number of users in each department, and number of warehouses and selling channels will all determine the size of the system. The steering committee should detail its plans so that the project team can estimate growth over the next two years. Are there any plans for new catalog titles or other new selling channels? How will planned changes in merchandise mix affect system functionality requirements?

Define your application requirements. It’s shocking how often companies fail to write out a detailed requirements document. Almost inevitably, these are the companies that have problems later. The project team should give higher priority to functionality than to the hardware or software platform that the system runs on. Each department should define the functions it needs. Consult users and have them sign off on their lists. Prioritize requirements (must have, nice-to-have, future needs).

Write a request for proposal (RFP) based on applications requirements. Your RFP should instruct/enable the vendor to communicate in a format that will allow you to make an “apples to apples” comparison. The RFP should include background about your company and its management objectives. It should request details on pricing of all licenses, hardware, software, annual support, training and file conversion services, third-party systems, modifications to meet your requirements, and planned future enhancements. It should also ask for all contracts for review and a complete user list for reference checks. The RFP should include your timetable, including written replies and demos.

Ask vendors to supplement their narrative for each requirement with a standardized code that denotes the availability of that requirement. For example: If a system meets a requirement without modification, label it yes. If it requires modification, mark it yes-mod. If it doesn’t meet the requirement, label it no-mod. And if it doesn’t meet the requirement and no modifications can be made, mark it no.

Evaluate the vendor responses. We find it useful to create a decision matrix. We use a weighted score to evaluate each vendor’s offerings, assigning a point value to each priority level (must-haves, etc.) and the vendor response codes (yes, yes-mod, etc.). Multiply the priority points by the vendor response points for each requirement and total up the results. In our experience, the best fits generally meet 70%-80% of a company’s needs before modification. Systems that score lower require too much costly customization.

Arrange vendor demonstrations. First, schedule at least a full day for a demo. Do demos only after reviewing vendors’ written responses to your RFP. Script the demos with users, so vendors will focus on what you want to see rather than what they want to show you. Ask vendors to use your own business profile and products for the demonstration, so you can see how it will perform. Update your matrix with your demo results.

Check references and make site visits. This extra effort is critical. Ask vendors for a full client list and call companies in your merchandise niche at random. Pick both long-term and recent installations to see how support and enhancements have been handled over time. Draw up a standard list of questions to ask so you can compare apples to apples.

Make a final selection. You should now have a good picture of the available systems. Minimize modifications wherever you can; they’ll run up costs, increase risk, and slow down implementation. Look at the three-year cost of implementation and support. (Watch out for escalator clauses and other potential surprises.) Choose based on your decision matrix and the investment required.

Negotiate intelligently. Hire an intellectual property lawyer, preferably one who specializes in computer contract law. Make sure that any modifications to the system are detailed in the contract. Each modification should have an addendum in the form of a work order with a description of the modification, time estimate, and price. Both parties should sign each of these documents. Remember that anything not mentioned in the contract, even if it was in the proposal, is not binding.

System selection can be an arduous process. If you take time to approach the task with a solid game plan, you’ll be able to make an informed choice. Then, when the time comes for implementing the new system, it will be your dreams — and not your nightmares — that come true.

Curt Barry is president of F. Curtis Barry & Co. He can be reached at 1897 Billingsgate Circle, Suite 102, Richmond, VA 23233; by phone at (804) 740-8743; or by e-mail at cbarry@fcbco.com (Web site: www.fcbco.com).