The Corner

Re: Why Does Arnold Bread Have Forty Different Kinds of bread?

Tyler Cowen links to a post in which William Gadea wonders why Arnold Bread offers so darn many products:

Is there really anyone in this world who loves the Arnold 10-grain, but can’t stand the 7-grain or 12-grain?

A slice of Arnold 12-grain has about 35 percent more calories and twice as much fat as a slice of Arnold 10-grain. I bet that might matter to some consumers. 

Gadea goes on with this:

More importantly in business terms, is the advantage of addressing these additional slivers of taste (if indeed people can make distinctions between the varieties – I can’t) really outweigh the additional expense of producing 40 separate packages, 40 separate categories of inventory, and 40 separate (at least slightly different) production processes?

My guess is no.

It’s not like this is the first time anybody has asked that question. There is a huge management discipline that focuses on this issue, and it attempts to evaluate product-line proliferation as the trade-off of incremental revenue versus increased complexity costs for production and distribution. But Gadea doesn’t have to bother with any knowledge of how the production economics of a bakery and its distribution network is affected by product count, or why different kinds of product proliferation might be more or less costly than others. He doesn’t have to bother with analyzing the elasticity of total product-line sales and margin to changes in product count. Or with doing any work at all, really. He can just guess the answer.

#more#Gadea continues:

The motivator here isn’t making the customer happier, it’s the oft-neglected fourth ‘P’ of marketing: placement. Even if the supermarket carries only half the varieties that Arnold offers, all of a sudden they are hogging a big part of the bread aisle.

Let me give you an example how a conversation between an Arnold account manager and the bakery department buyer at a major retailer normally does not go:

Arnold account manager: We’ve now doubled our SKU count by adding 20 new products.

Retailer buyer:  OK, then we’ll give you twice as much shelf space.  Never mind that huge space allocation model we built to assign space based on product space elasticities. Also, never mind all of those randomized experiments we ran to test the effect of adding space to the bakery category at the expense of canned goods. We won’t need to see any test market data on the sales of your new products, and we won’t try adding some of your products in a few stores to see if it helps our overall bakery department GMROI. Which of course reminds me to forget all of the slotting fees that other CPG companies are paying us for the shelf space that you want. I’ll give it to you because our space allocation rule is “always take half of the products offered by any major vendor.” Because we’re morons. 

Consumer package goods (CPG) vendors are in an endless war for retail shelf space. They will add products to try to “hog the shelf,” just as they will invest in joint promotions with the retailer, buy TV advertising, try new packaging ideas, and so on. They often will just cut to the chase, and pay the retailer cash money (slotting fees) for it. Similarly, the retailers are constantly trying to figure out how to range and merchandise departments, manage their overall shelf capacity and so on. They try to allocate shelf space to the products that will make them the most money. Sometimes they believe it to be in their interest, given the coordination costs and costs of maintaining expertise, to outsource some shelf decisions to CPG “category captains.”  The trade-offs here are multi-dimensional, constantly evolving, demand judgment, and are never made perfectly; but both sides are working hard to drive up their own economic profits, and nobody who stays in business is a child about them.

Determining what product to sell and how much space to give each product is extremely challenging, both analytically and organizationally. This is why retailers and CPGs are constantly undertaking line reviews, and are pretty much always either rationalizing SKUs (cutting products from the line) or adding products to meet perceived unmet needs. It is a much harder problem than just wandering into a supermarket and deciding “who could possibly want 40 kinds of bread?” without even bothering to read the labels.

Jim Manzi is CEO of Applied Predictive Technologies (APT), an applied artificial intelligence software company.


The Latest