Number of Supply Chains

The last strategy we’ll cover is based on a company having more than one supply chain, depending upon the number and types of products that are passing along the chain and other variables. For a product with a complex bill of material (many parts that combine into many components to make the final product), a manufacturer may be bringing in materials from many suppliers. And these materials might range from low-priced commodities to fragile or sophisticated materials that require special shipping and handling. Suppliers might range from small specialized firms to raw materials giants larger than the manufacturer. Some are key accounts; some might be occasional buyers. The finished products may be sold through several different channels—c-commerce, printed catalogs, commercial, and retail. These variables may combine in different ways, each suggesting its own type of supply chain strategy. Next we’ll explore how product types, functional versus innovative, often require different supply chain strategies.

In “What Is the Right Supply Chain for Your Product?” Marshall L. Fisher distinguished two types of products, functional versus innovative, that requires different supply chain strategies. Functional products that change little from year to year have longer life cycles (perhaps more than two years), relatively low contribution margins, and little variety. Because demand for them is stable, they are fairly easy to forecast, with a margin of error of about 10 percent, very few stock outs, and no end-of season markdowns. The appropriate supply chain for these products should emphasize predictability and low cost with performance indicators such as the following:

• High average utilization rate in manufacturing.
• Minimal inventory with high inventory turns.
• Short lead time (consistent with low cost).
• Suppliers chosen for cost and quality.
• Product design that strives for maximum performance and minimal cost.

However, make-to-order functional products, such as replacement parts for customized equipment, usually have long lead times (six months to a year). Innovative products have unpredictable demand, relatively short life cycles (three months for seasonal clothing), and high contribution margins of 20 to 60 percent. They may have millions of variants in each category, an average stock out rate from 10 to 40 percent, and end-of-season markdowns in the range of 10 to 25 percent of regular price. The margin of error on forecasts for innovative products is high-40 to 100 percent—but the lead time to make them to order may be as low as one day and generally is no more than two weeks. The supply chain for innovative products should emphasize market responsiveness rather than physical efficiency, with performance indicators such as the following:

• Excess buffer capacity and significant buffer (or safety) stock of parts or finished items.
• Aggressive reduction of lead times.
• Suppliers chosen for speed, flexibility, and quality (rather than cost).
• Modular design that postpones differentiation as long as possible.

Innovative products, with their high margins and unpredictable demand, justify the extra expense for holding costs. (Fisher also proposes, however, that manufacturers of innovative products can look for other solutions to the problem of unpredictable demand, such as aggressively reducing lead times and producing products to order rather than for inventory.)

Here is a conundrum… What happens when a product can fall into either category? Fisher says that some products can be either innovative or functional. Automobiles fit that description, with a low-priced, no-frills car like a base model Chevrolet Cobalt or Hyundai Excel representing the functional end of the spectrum and a Porsche representing the other end. Similarly, coffee can be functional—as anyone who has worked in an office knows, in which case it should be available quickly at a low price with perhaps cream and sugar as options. At a high-end coffee shop, on the other hand, patrons are willing to endure longer lead times and pay more money for their coffee, but they want variety in return.

The idea that the same type of product can be either functional or innovative implies that one company might have more than one supply chain. And that’s the contention of Jonathan Byrnes, a professor at MIT. Writing in the Harvard Business School’s Working Knowledge, Byrnes asserts that one supply chain is not enough; two, three, or more would be preferable. “One size fits all” supply chains may have been sufficient in the past, he believes, when that was the competitive norm, but new information technology makes it possible to have multiple, dynamic chains that can accommodate different product and information flows.

Byrnes breaks products into three categories: staples, seasonal products, and fashion.

• Staples (which are much like Fisher’s functional products) have steady, year-round demand and low margins. White underwear is an example. Byrnes advises stocking staples only in retail outlets in small quantities and transporting them in truckload quantities. (A full truck is more cost-effective for the shipper than a partially loaded vehicle.)
• Seasonal products could include outdoor patio furniture, holiday decor, etc., for which the demand is more predictable since it is tied to the holiday season.
• Fashion products are like Fisher’s innovative items, with unpredictable demand. Zara, the Spanish clothing manufacturer, has two supply chains, one for staples and the other for fashion clothing. To get the fastest response time, Zara uses European suppliers for the fashion items. But for the more predictable demand items, it uses eastern European suppliers that have poor response time (not a concern) and lower cost.

In addition to varying the supply chain by product type, Fisher recommends several other variables to consider—store type and time in season or product cycle. Demand varies considerably over the life cycle of many products. The same item might have infrequent demand at first, more stable demand in its maturity phase, and falling demand at the end of its life cycle. With more than one supply chain, the nucleus firm can move its products from one chain to the other in response to changing variables, such as type of channel or life-cycle stage.