Scan-downs, also called Pay-for-Performance, are programs in which manufacturers pay retailers allowances based on units sold rather than cases shipped. These programs theoretically eliminate forward-buying and diverting because data comes from the checkout. Here's how the program works: The retailer puts up a shelf-talker near the product and/or lists it as a temporary price reduction in the weekly ad or store flier. After the allotted time period (usually two to four weeks), the retailer bills the manufacturer, based on scanner movement data. A fee is applied, usually called a merchandise incentive payment, or an 8-cent-per-unit "handling" fee.
Several third parties are in the business of collecting and ensuring accuracy of this data. They include Efficient Market Services, Scanner Applications, Catalina Marketing and Information Resources Inc. Some of these companies collect scan data directly from the store's system. Others obtain it from the retailer, but verify its accuracy. Most scan-downs, however, are still handled directly between manufacturers and retailers. What are the issues here? First of all, most retailers don't take scan-down data directly from their scan systems. Invoices from retailers are usually developed on a personal computer within the office of the buyer who made the deal. As long as they're on some kind of computer printout, the manufacturer assumes it's "real" scanner data. It's often not.
One of the ways manufacturers could double-check, you'd think, would be to compare scan-down invoices with Information Resources Inc. or A.C. Nielsen movement reports. This is difficult to do, however, because most product movement reports are based on equalized volume (for example, X ounces of regular cola). Raw (unprocessed) retailer data is in units (for example, X liters, X 6-packs, X 12-packs of regular cola, etc.). So manufacturers really don't have a way to check the numbers. Do retailers alter the numbers? Sometimes. But the manufacturers I know say they were losing money before, due to retailer noncompliance on other trade promotions. They'd pay X dollars off per case for a six-week period if the product were displayed, run in the ad or otherwise promoted. The incidence of no display or obituary listings in the ads, however, is high. Manufacturers think they might be better off with the scan-downs, even if the numbers are inflated. And the savings associated with eliminating forward-buying and diverting might offset the extra they pay for scan-down promotions. Do the retailers have any rationale for doing what they do? Yes, they say (and some of the third parties agree). The quality of their scanner data is not perfect -- and in the case of scan-downs, it errs against the retailer. Some retailers say it's fair to add 10% to 20% to the scan-down to make up for cashier error. What's really fair? Who knows? These issues will apply to frequent shopper promotions. As the industry progresses, we'll eventually have the data coming direct, with no intervention. Until that time, I guess we'll have to live with questions related to scan-downs.
