About AMA

Executive Summary
There are ample reasons for geographic variation in brand share to persist. This comment highlights intentional strategies and feedback loops.

Trade promotion, advertising, coupons, and special events are likely to be differentially allocated on the basis of market position and may have different effectiveness for proportional programs. For example, the same manufacturer discount rate per case will result in more money in a high-share area and better retailer support because the retailer anticipates more sales.

Distribution is highly correlated with share, and distribution can be easier to get and maintain in strong markets. A tight relationship and a positive feedback loop go a long way to sustain differences in geography across time.

The interaction between retailers and manufacturers at the category level can be strong and persistent and can be viewed in market-level data as a market effect if the retailers have different strength in different markets. For example, there is one national retailer in which Pepsi’s share of regular soft drinks is higher than Coke’s share across regions. There is another retailer in which Coke’s share is higher than Pepsi’s share across regions. This difference has persisted for years. The same effect can be observed in a close category (low-calorie soft drinks) but not in the bottled-water category.

Biography
Michael W. Kruger is Executive Vice President of Analytics Research and Development at Information Resources Inc. Mike joined Information Resources Inc. in 1989 after working at Quaker Oats, Beatrice, and SAMI/Burke. He develops algorithms for syndicated products that use point-of-sale and consumer panel data. He received a PhD in 1984 from the University of Michigan.