Thus far I have described how to analyze the cost behavior of a business unit as a whole. In practice, however, a business unit usually produces a number of different product varieties and sells them to a number of different buyers. It may also employ a number of different distribution channels. For example, a shipbuilder constructs both liquid natural gas tankers and containerships while a bank lends to sophisti cated high net-worth individuals as well as to middle income customers. Any of these differences may give rise to segments in which the behavior of costs in the value chain may be different. Unless the firm recognizes differences in cost behavior among segments, there is a significant danger that incorrect or average-cost pricing will provide openings for competitors. Thus cost analysis at the segment level must often supplement analysis at the business unit level.
Chapter 7 discusses the identification and analysis of segments in more detail. Differences in cost behavior among products, buyers, channels, or geographic areas is one of the key bases for the existence of segments, and hence cost analysis is an essential input to segmenta tion. The value chain for segments generally parallels that of the whole business unit. However, segment value chains may differ in some re spects that affect cost. For example, the large sizes of a product line may be produced on different machines than small sizes and require different handling, inspection, and shipping procedures. Similarly, they may require different purchased inputs. Identifying important differ ences in the value activities for different segments is a starting point in segment cost analysis.
A firm should analyze the costs of those product lines, buyer types, or other portions of its activities that
- have significantly different value chains
- appear to have different cost drivers
- employ questionable procedures for allocating costs
In practice, a firm may want to select representative product varieties or buyers to illuminate differences among segments, rather than analyze every product variety or buyer in complete detail.
The process used to analyze cost behavior for segments is the same as that used for business units. The value chain for the segment is identified and costs and assets are assigned to it. Then the cost drivers of each activity are determined and quantified if possible. While the process remains the same, however, some complications often arise in practice. The prevalence of shared value activities among segments (see Chapter 7) requires the allocation of costs among segments. Stan dard cost systems often employ arbitrary measures as the basis for allocating cost to segments, such as sales volume or other readily measurable variables. While these measures have the benefit of simplic ity, they often have little to do with the true contribution of the segment to overall costs. For example, allocating the costs of a value activity to domestic and international buyers by volume of sales will usually seriously understate the true cost of international sales, because interna tional sales often make disproportionate demands in terms of time and attention. The costs of support activities and the costs of indirect primary activities appear to be most susceptible to misallocation. Such misallocations result in incorrect costs and inappropriate prices for product or buyer segments.
The costs of value activities shared among segments should be allocated based on each segment’s actual impact on the effort or capac ity of the value activity. Such measures will capture the opportunity cost of using a shared value activity in one segment instead of another. In technology development, for example, allocation should probably be based on the estimated time spent by engineers and scientists on particular product lines rather than on the products’ respective sales volumes.
It is not always feasible or necessary to allocate the costs of shared activities to segments on an ongoing basis. The required analysis for strategic purposes does not require a high degree of precision, and periodic studies can suffice. To allocate R& D costs, for example, engi neers can be interviewed to determine the percentage of their time spent on various products and buyers over a period of time long enough to eliminate distortions. Some firms may also be in a position to com pute time allocation by sampling engineering change orders or requests for product modifications flowing to the engineering group from the sales force. Similar methods of approximation can provide the basis for allocating effort to segments in almost any shared activity.
Source: Porter Michael E. (1998), Competitive Advantage: Creating and Sustaining Superior Performance, Free Press; Illustrated edition.