FEATURE: Wake-up Call: Unintended consequences
I've been reading a lot recently about the large effect that unintended consequences -- those negative, unexpected results of actions we take -- have on our lives.
Sometimes these unintended consequences result from bad ideas or bad execution, as in, to paraphrase, "We didn't really mean to make Iraq safe for terrorists, we just messed up."
Often, however, unintended consequences occur while we're merely trying to make things better. My congressman, for example, established his reputation as an anti-nuclear-power activist in the 1970s, but he didn't realize then, as he does now, that eliminating nuclear power plant construction would lead to global warming, since all of the substitute power plants instead would burn coal. Thus, his position on nuclear power has "evolved" significantly, thanks to that unintended consequence.
These unintended consequences profoundly affect our professional lives in the supermarket as well. We originally installed scanners and urged the adoption of UPC codes to speed throughput at the checkstand. Now, however, we rely on the data those scanners produce to substantially run our businesses: Ordering, shrink, and marketing are all managed by relying on scanner data. So in a real sense, many of our current management systems are actually a positive unintended consequence of an improvement originally designed to speed one aspect of store operations.
I was reminded of this phenomenon recently when I was reviewing an analysis fielded by our friends at our sister firm Nielsen regarding the aseptic nondairy beverage category, which includes soymilk. Apparently, despite the fact that the aseptic nondairy beverage category is growing, its shelf space, SKU lineup, and brand assortment is shrinking daily. Why is this happening?
The explanation is simple. Aseptic nondairy beverages don't, as far as the industry seems to be concerned, actually comprise a category; they're an integrated part of the nondairy beverage category, which also includes refrigerated aseptic beverages -- and that segment has been decimated by the death of Parmalat, the unintended consequence of a completely unrelated financial scandal.
So when you apply category management principles to the aseptic nondairy beverage segment, you shrink both aseptic and aseptic refrigerated segments, based on a declining overall category. Thus, the decline in aseptic nondairy shelf space is driven by the decision to make aseptic and refrigerated one category.
The assumption seems to be that since shoppers who buy cow's milk are switching from aseptic to refrigerated versions, those who pick up aseptic nondairy beverages will do the same.
Redefining categories
But the real world doesn't work this way. As I see it, the Nielsen Homescan data I reviewed shows clearly that consumers who buy refrigerated nondairy beverages, and those who buy aseptic nondairy beverages, don't switch off between forms. In fact, 92 percent of Homescan panelists who reported making an aseptic purchase in food stores did the same thing the last time they purchased in the category.
Further, a significant percentage of consumers report that they would go to another store if their preferred aseptic product wasn't available at the usual store. These results ranged from 30 percent of consumers in food stores to 94 percent of consumers in mass stores reporting that they would shop elsewhere if their aseptic product weren't available.
What we see here is the law of unintended consequences at work. We're driving away our consumers by taking a growing segment, aseptic nondairy beverages, and minimizing it because we don't call it a category. We've allowed a completely unrelated financial scandal at Parmalat to be interpreted as a fundamental change in consumer behavior.
All we need to do to fix this is redefine the category. If we recognized aseptic nondairy beverages as a category, and saw growth in that category, we'd be discussing how much space, how many SKUs, and how many facings to add -- because it would be one of the few growing categories in the store. Instead, what we have here is a case of proper category management techniques, undermined by faulty category definitions, leading us to drive consumers out of our stores.
You might conclude the industry needs to do two things: to break the nondairy beverage category into two categories -- refrigerated and aseptic -- and create more, smaller categories. Well, yes and no. Yes, we need to make nondairy beverages two distinct categories. But the second generalization isn't so easy, because more categories aren't necessarily better categories.
Let's look at another example, from a few years ago. In the 1990s two powerhouses battled for share in the large and growing pasta sauce category: Lever (Ragu) and Campbell (Prego). Both ended up running ads that compared their product (rich, thick, and luscious) with the competitor's product (thin, watery, and runny).
By the end of the decade, they had each spent hundreds of millions of dollars in advertising, yet their shares remained basically unchanged. What's more, the category had declined 40 percent.
Some might be tempted to say this is proof that advertising doesn't work, but I argue that it's proof that advertising does work. Both Lever and Campbell did a great job of showing how bad the competitor's product looked, and their marketing campaigns persuaded consumers not to buy pasta sauce in a jar from anyone.
During this 10-year period, while the pasta sauce category was declining by 40 percent, the canned tomato category grew 70 percent and the frozen pasta entree category grew 100 percent. So did the advertising work? You bet! All of those images of runny pasta sauce persuaded me that I shouldn't buy it, regardless of the brand, and that if I wanted good sauce, I should make it from scratch, using canned tomatoes. And if I were in a rush, I could buy a frozen entree and it would be just as good.
Let consumers lead
So here, the same mistake was made -- because the category was not "pasta sauce," the category was "pasta for dinner."
The unintended consequence of defining pasta sauce as a category was that it focused the two competitors on each other, and allowed them to systematically persuade consumers that the category wasn't worth buying, since consumers weren't so much after "pasta sauce" as they were hungry for "spaghetti and meatballs for dinner tonight." Consumers got what they wanted in other ways.
So with aseptic nondairy beverages, the issue is that bad action is driven by a category defined too broadly, while in pasta sauce, bad action is driven by a category drawn too narrowly. So what's a retailer to do? How can these bad unintended consequences be avoided?
The best thing retailers can do in this case is to react like a consumer, like your customer. In both of these situations, the bad outcome would have been avoided if we had thought like our customers, and not like "industry professionals."
Category definitions can't be driven by distribution methodology or by organization charts. They need to be driven by the way consumers think about what they're buying. And when you don't know, ask some consumers what they think.
If one product can be substituted for another in most circumstances, they're in the same category, and if consumers don't consider them reasonable substitutes, they're not in the same category. But we need to let the consumer lead us.
This is the final and most important point. I would argue that if you cut 10 percent from your marketing budget (the 10 percent you like least), give half of it back to your c.f.o. to make him or her happy, and spend the other half on consumer research, you would understand your consumers much better and put yourself in a position where you have at least lowered the odds of doing something counter to the way your consumers think.
Sometimes these unintended consequences result from bad ideas or bad execution, as in, to paraphrase, "We didn't really mean to make Iraq safe for terrorists, we just messed up."
Often, however, unintended consequences occur while we're merely trying to make things better. My congressman, for example, established his reputation as an anti-nuclear-power activist in the 1970s, but he didn't realize then, as he does now, that eliminating nuclear power plant construction would lead to global warming, since all of the substitute power plants instead would burn coal. Thus, his position on nuclear power has "evolved" significantly, thanks to that unintended consequence.
These unintended consequences profoundly affect our professional lives in the supermarket as well. We originally installed scanners and urged the adoption of UPC codes to speed throughput at the checkstand. Now, however, we rely on the data those scanners produce to substantially run our businesses: Ordering, shrink, and marketing are all managed by relying on scanner data. So in a real sense, many of our current management systems are actually a positive unintended consequence of an improvement originally designed to speed one aspect of store operations.
I was reminded of this phenomenon recently when I was reviewing an analysis fielded by our friends at our sister firm Nielsen regarding the aseptic nondairy beverage category, which includes soymilk. Apparently, despite the fact that the aseptic nondairy beverage category is growing, its shelf space, SKU lineup, and brand assortment is shrinking daily. Why is this happening?
The explanation is simple. Aseptic nondairy beverages don't, as far as the industry seems to be concerned, actually comprise a category; they're an integrated part of the nondairy beverage category, which also includes refrigerated aseptic beverages -- and that segment has been decimated by the death of Parmalat, the unintended consequence of a completely unrelated financial scandal.
So when you apply category management principles to the aseptic nondairy beverage segment, you shrink both aseptic and aseptic refrigerated segments, based on a declining overall category. Thus, the decline in aseptic nondairy shelf space is driven by the decision to make aseptic and refrigerated one category.
The assumption seems to be that since shoppers who buy cow's milk are switching from aseptic to refrigerated versions, those who pick up aseptic nondairy beverages will do the same.
Redefining categories
But the real world doesn't work this way. As I see it, the Nielsen Homescan data I reviewed shows clearly that consumers who buy refrigerated nondairy beverages, and those who buy aseptic nondairy beverages, don't switch off between forms. In fact, 92 percent of Homescan panelists who reported making an aseptic purchase in food stores did the same thing the last time they purchased in the category.
Further, a significant percentage of consumers report that they would go to another store if their preferred aseptic product wasn't available at the usual store. These results ranged from 30 percent of consumers in food stores to 94 percent of consumers in mass stores reporting that they would shop elsewhere if their aseptic product weren't available.
What we see here is the law of unintended consequences at work. We're driving away our consumers by taking a growing segment, aseptic nondairy beverages, and minimizing it because we don't call it a category. We've allowed a completely unrelated financial scandal at Parmalat to be interpreted as a fundamental change in consumer behavior.
All we need to do to fix this is redefine the category. If we recognized aseptic nondairy beverages as a category, and saw growth in that category, we'd be discussing how much space, how many SKUs, and how many facings to add -- because it would be one of the few growing categories in the store. Instead, what we have here is a case of proper category management techniques, undermined by faulty category definitions, leading us to drive consumers out of our stores.
You might conclude the industry needs to do two things: to break the nondairy beverage category into two categories -- refrigerated and aseptic -- and create more, smaller categories. Well, yes and no. Yes, we need to make nondairy beverages two distinct categories. But the second generalization isn't so easy, because more categories aren't necessarily better categories.
Let's look at another example, from a few years ago. In the 1990s two powerhouses battled for share in the large and growing pasta sauce category: Lever (Ragu) and Campbell (Prego). Both ended up running ads that compared their product (rich, thick, and luscious) with the competitor's product (thin, watery, and runny).
By the end of the decade, they had each spent hundreds of millions of dollars in advertising, yet their shares remained basically unchanged. What's more, the category had declined 40 percent.
Some might be tempted to say this is proof that advertising doesn't work, but I argue that it's proof that advertising does work. Both Lever and Campbell did a great job of showing how bad the competitor's product looked, and their marketing campaigns persuaded consumers not to buy pasta sauce in a jar from anyone.
During this 10-year period, while the pasta sauce category was declining by 40 percent, the canned tomato category grew 70 percent and the frozen pasta entree category grew 100 percent. So did the advertising work? You bet! All of those images of runny pasta sauce persuaded me that I shouldn't buy it, regardless of the brand, and that if I wanted good sauce, I should make it from scratch, using canned tomatoes. And if I were in a rush, I could buy a frozen entree and it would be just as good.
Let consumers lead
So here, the same mistake was made -- because the category was not "pasta sauce," the category was "pasta for dinner."
The unintended consequence of defining pasta sauce as a category was that it focused the two competitors on each other, and allowed them to systematically persuade consumers that the category wasn't worth buying, since consumers weren't so much after "pasta sauce" as they were hungry for "spaghetti and meatballs for dinner tonight." Consumers got what they wanted in other ways.
So with aseptic nondairy beverages, the issue is that bad action is driven by a category defined too broadly, while in pasta sauce, bad action is driven by a category drawn too narrowly. So what's a retailer to do? How can these bad unintended consequences be avoided?
The best thing retailers can do in this case is to react like a consumer, like your customer. In both of these situations, the bad outcome would have been avoided if we had thought like our customers, and not like "industry professionals."
Category definitions can't be driven by distribution methodology or by organization charts. They need to be driven by the way consumers think about what they're buying. And when you don't know, ask some consumers what they think.
If one product can be substituted for another in most circumstances, they're in the same category, and if consumers don't consider them reasonable substitutes, they're not in the same category. But we need to let the consumer lead us.
This is the final and most important point. I would argue that if you cut 10 percent from your marketing budget (the 10 percent you like least), give half of it back to your c.f.o. to make him or her happy, and spend the other half on consumer research, you would understand your consumers much better and put yourself in a position where you have at least lowered the odds of doing something counter to the way your consumers think.