Objectives and Cost-Benefit Analysis of Customer Service

To measure and monitor the service delivered to customers, there are three main aspects of the ‘perfect order’: these are ‘on-time’, ‘in-full’ and ‘error-free’.

To calculate the service level, each aspects is calculated in percentage, and then the aggregated perfect order achievement is determined by the following equation:

Perfect order achievement = (on-time)% x (in-full)% x (error-free)%

Then, the objective is set for each element and for the aggregated measure.

Providing services imposes an extra logistics cost; however, it is vital to deliver proper services to customers. The following image shows that by increasing the service level, the cost of service level increases in a steeply rising curve.

On the issue of finances, the Pareto law (otherwise known as the 80/20 rule) is often cited within a business context. As a general principle, 80% of a business’s revenue comes from only 20% of customers – and 80% of total costs derive from 20% of customers (but unlikely to be the same 20%). This principle will give us a basis for developing an effective service strategy, as well as giving a baseline for prioritising.

Service levels based on a normal distribution

Let’s use a scenario where customer demands are not deterministic and may change based on a normal distribution (or bell curve) shape. If you’re not familiar with what a normal distribution is, or need a refresher, this simple guide is a useful aid for understanding the graph and calculations to follow.

In an inventory and sales case, suppose that customers’ demands can be covered in the range of (x̄-3σ, x̄+3σ), where x̄ is the average demand and σ is the standard deviation of demand – this means that we have a 99.7% service level. In other words, only 0.3% of customer demands will be lost.

However, if we can cover only demands between the range of (x̄-2σ, x̄+2σ) because of available inventory, it means that only 95.4% of demands will be satisfied, so the service level is 95.4%.

Finally, we can cover about 68% of customer demands if the inventory level is determined based on the (x̄-σ, x̄+σ); the reason being that 68% of distribution will be inside the mentioned limitation.

References

_{Christopher, M. (2016), Logistics and supply chain management (5th ed.). Pearson.}