Selecting the customer service level
From Supply Chain Management Encyclopedia
Russian: Выбор уровня обслуживания клиентов
One approach to evaluating the optimal service level for the firm is to adopt the profit maximization economics model which states that, in a perfectly competitive market, the firm should produce output at that point where marginal revenue equals marginal cost. Replacing the number of units provided as the indicator of output with service as the indicator of output suggests that the model may be applied to a variety of customer service variables including fill rates, on-time delivery, lead-time, and goods damaged during shipment. Let us assume that the values on the horizontal axis in Figure 1 represent the fill rate with a viable range of 90% to 100%. Marginal revenue increases at a decreasing rate as the service level output increases. As the firm increases its fill rate, supply is increased, and therefore the unit price declines. Since demand for input factors increases, marginal cost increases at an increasing rate as the fill rate increases. The third line in Figure 1 represents profit (or revenue minus cost). Profit is maxmimized where the slope of the cost curve equals the slope of the revenue curve. This occurs at an approximate 95% fill rate.
From a managerial perspective, a number of variables go into evaluating the cost and revenue curves. On the cost side, improving the on-time delivery rates, for example, may include information technology infrastructure investments which would be fixed over a range of improved service levels. The firm may be required to invest in better forecasting tools or better ways to improve communications among business functions (e.g., sales, production, and distribution). Some costs might be variable and rise according with improvements in on-time delivery. For example, a larger proportion of shipments may require speedier and hence more expensive modes of transport. The creation of revenue curve is often problematic due to the challenges of evaluating customer response to different levels of various logistics customer service variables. Furthermore, the cost and revenue curves may not be constructed, but rather replaced with income statements showing profit at various service levels. The net goal of profit maximization remains unchanged.
In a competitive industry, the revenue curve faced by two competitors should look rather similar. However, in practice it is often the case that different firms possess different cost curves. This would be due to dissimilarities in strategic decisions over supply chain investments, skill sets, learning rates, and openness to change, among others. In Figure 2, Firms 1 and 2 operate in one market and both face the same revenue curve in relation to the service level (again, we may assume that the sevice level is represented by a fill rate). The cost curve for Firm 1 is shifted downward relative to that of Firm 2. From the profit curves, Firm 1 should select a profit maximizing fill rate of 95%. Firm 2 should select a profit maximizing fill rate of 96%. Firm 1 would want to operate with a profit maximizing service level lower than the profit maximizing service level of Firm 2. However, Firm 1 may select a “match the competition” heuristic to determine its customer service level and thus select a non-optimal service level. Suppose that Firm 2 continually seeks ways to open a gap between itself and its competitors by implementing fashions to shift its cost curve downward. Rivalry of this sort for Firm 1 is troublesome as it elects to compete on its service level and not on its underlying cost structure. Efforts are required to shift cost curves, not merely move along them.
In a simple example, let us assume that each of the revenue and cost curves may be represent by straigtforward formulas where x=fill rate with a range of 0 to 100:
Taking the first derivative of the cost and revenue formulas and setting them equal yields:
Solving for x yields 94.9, which equals an approximate fill rate of 95%
A more realistic yet simple example better illustrates the business process underlying the evaluation of the optimal service level. Suppose the current late delivery rate of a firm is 5%. As seen in Table 1, the firm generates €100 million in billed revenue at this late delivery rate. The late delivery penalties are evaluated as the late delivery rate multiplied by the billed revenue. Net revenue equals the difference between the two. The cost of goods sold equals 35% of the billed revenue as a late delivery penalty does not detract from the production value of the product. A gross margin of €60 million with €40 million in fixed expenses yields a profit before interest and taxes of €20 million. A shown in Table 1, the billed revenue will increase as the late delivery rate declines. Acquiring these billed revenue values may not be a simple matter. Sources of input could include interviews with sales and logistics staff and with customers. The additional expenses to reduce late delivery might come from several sources. For example, the firm may use free on board origin pricing (which means that the customer pays the freight charge). However, customers might only be willing to pay standard delivery rates. The firm, in order to decrease the late delivery rate, may absorb the difference between standard and premium (i.e., faster) freight charges. Furthermore, decreasing the late delivery rate may require investments in improved forecasting tools and information technologies. As seen in Table 1, these additional expenses were determined to be €3 million and €8 million respectively for late delivery rates of 4% and 3%. Profit is maximized (€24.1 million) at the 4% late delivery rate.
|Table 1: Evaluating the Optimal Service Level (€'000)|
|5% Late delivery rate||4% Late delivery rate||3% Late delivery rate|
|Billed revenue||100 000||110 000||115 000|
|Late penalties (late delivery rate × billed revenue)||5 000||4 400||3 450|
|Net revenue||95 000||105 600||111 550|
|Cost of good sold (35% of billed revenue)||35 000||38 500||40 250|
|Gross margin||60 000||67 100||71 300|
|Fixed expenses||40 000||40 000||40 000|
|Additional expenses to reduce late delivery rate||3 000||8 000|
|Total expenses||40 000||43 000||48 000|
|Profit before interest and taxes||20 000||24 100||23 300|
- ↑ Ballou, R.H. (1992), Business Logistics Management, Prentice Hall, Upper Saddle River: NJ