Line Extensions: Fine Young Cannibals

While catching up on my post-vacation reading today, I saw a Brandweek article about candy market segments according to Hershey. While the part about the consumer segments (running the gamut from “loyal indulgers” to “detached occasionalists”) was the topic that initially caught my eye, after reading the article, what really piqued my interest was that this marketing philosophy came about as a result of product cannibalization. According to the article: “Hershey’s old approach involved introducing flavor extensions of core brands, such as Reese's White Chocolate, Reese's Caramel and Reese's Big Cup, which had cannibalized sales. Hershey relied heavily on flavor extensions in 2006 and ad spending declined 13% from the previous year, per Nielsen... After Hershey's sales increased 5.3% between 2001 and 2005, revenue rose just 2.6% during 2006 and 0.1% last year.”

I’ve often wondered what marketeers use for a formula to determine line-extension cannibalization, since it’s always a danger when going that route. While the goal is to steal the business from the other guy’s brands—line extensions rarely carve a new niche—the strategy can backfire and eat its own. (Cut to the mind of a typical consumer perusing the cookie aisle: “Oh look green-filled O-brand cookies! I will buy them because... [steal] I was going to buy cookies for a St. Patty’s Day party and these are perfect! [new niche] I’ve always wanted to eat a cookie with green filling and I must have them! or [cannibal alert] I was going to buy regular O-brand anyway and green filling is so much more festive than white!”)

In my own product-development days, brand extensions seemed to involve a “cross your fingers” prediction method, a me-too product to answer someone else’s line extension, or sometimes just a little something to temporarily create buzz for flagging sales. Not that the last two aren’t sound business reasons. However, marketing rarely shared the strategy...something I found a bit frustrating on more than one occasion. But in this case, the numbers indicate it may have been more of a wishful-thinking approach.

Of course as a scientist, I’d be lax in not noting the concurrent decrease in ad spending, and we always know to question cause and effect when we have more than one variable. But I’m off to hit the IFT show, and those are both topics for another day.

―Lynn A. Kuntz