Don't let this make you nervous, but if you're building or extending a supply chain management system designed to automate transactions with buyers and suppliers, the people whose functions you're trying to reduce or eliminate are the ones who will ultimately make it either work or fail.
You know these people. But you may have succeeded in avoiding them since you made it to a high enough position on the help desk that you no longer have to fix mysterious system problems caused by smiley-face magnets on their PCs.
They're data entry managers, accounts receivable managers and customer service specialists, all of whose work represents the expensive, slow part of the supply chain.
But they're also the final fail-safe for your supply chain system—the human middleware that's flexible enough to adapt to new processes and customers even before you tie them into the network. They resist change to their own processes, but they can adapt a lot more quickly to new customer problems than to the software you're building.
They don't mind keying in order data, even when they do it with a green-screen application whose codes and formats they spent years learning. That's their job, and by doing it, they can connect customers who work with incompatible systems, customers who insist on faxing in orders or those who just get the orders wrong.
Market research firm Killen & Associates Inc. in Palo Alto, Calif., estimated earlier this year that one-fifth of all business transactions—most of them hand-keyed—have errors that can delay payments by at least a month. That can cost a $1 billion company more than $32 million per year, the report said.
A lot of those errors come from miscoding by your own people, but most of them come by way of miscodes from other companies—errors you can't possibly correct automatically.
 |
|
|
Based on the technology and best-practices information available, there are basically three ways to minimize those errors:
• The macro tech approach. This embraces standards throughout your whole system and helps your buyers and suppliers do the same. It will reduce friction and errors for everyone.
Try the XML-based set of standards being developed by RosettaNet, a nonprofit consortium made up largely of technology companies whose specs are among the most widely regarded.
Standards-based exchanges are your Holy Grail. Enjoy them when you can get them, but don't count on getting them all the time.
• The micro tech approach. Add a layer of supply chain event management (SCEM) software onto your system. These applications are designed to sit on the servers and monitor the process by flagging late orders, logging shipments and looking for discrepancies in orders and other errors. They're being built into applications from SAP AG, J.D. Edwards & Co. and others and are sold as stand-alones by companies such as HighJump Software Inc. in Eden Prairie, Minn. That minimizes the errors when they occur.
• The fallback factor. The third approach is to get those data entry managers and departmental order-takers on board not only to ease the political process of rolling out your system, but also to give your system an automatic fail-over capability that can take into account random customer errors, new customer problems and other integration glitches for which IT can't possibly plan.
Get too automated, and you can't deal efficiently with these people. But in this economy, you can't afford to ignore them.
So build your own people into the workflow plan to deal with problems the SCEM can't handle, and build the time and money they cost into your budget projections. You'll still vastly reduce the amount of expensive human intervention required, but you won't be caught unprepared when something changes too fast for your system to handle.
After all, people don't just cause system failures; sometimes, they can fix them, too.
Kevin Fogarty is a former Computerworld editor. He can be reached at kevinfogarty@yahoo.com.
Missing Links
Stories in this report:
- Bad News Everywhere
- Beyond Paper Clips
- CIOs Catch On to SCM Standards
- Case study: Unilever Crosses the Data Streams
- Courting the Dispossessed
- Dirty Data
- Just in Case
- Kinks in The Chain
- Other Challenges That Lie Ahead
- Patching the Supply Chain Together
- Supply Chain Management
- Tech Check: Getting Demand Planning Right
- The Weakest Link
- Tips for Success
- Vendor Choices: Know the Differences
- Your Crystal Ball
- Covisint's Stalled Start
- Data quality should be a boardroom issue
- Gaining better visibility
- Internet-based collaboration beats airport hassles
- Managing the people supply
- Perkins Takes Smart Approach to Online Parts Catalog
- Supply chain uncertainty requires better IT tools
- The cost-cutters
- Tips from the field: Deploying demand forecasting
