The value of SKU rationalization: the pooling effect under suboptimal inventory policies

Managing product variety is a widely recognized challenge. Several approaches to this rely on the "pooling effect", the reduction of uncertainty that occurs when individual demands are aggregated. This can occur through reduction of number of products orSKUs, through postponement of differentiation, or in other ways. These approaches are by now well-known and widely applied in practice. However, theoretical analyses of the pooling effect always assume that one has an optimal inventory policy before and after pooling. If this is not the case, how does that affect the value of pooling? This paper analyses the benefits of pooling in terms of costs and service level under optimal and suboptimal policies and proposes a simple framework to analyze the trade-off between implementing pooling and improving inventory policy. We show there is always a range of current inventory levels within which pooling is better and beyond whichoptimizing inventory policy is better. We analyze how this range varies with the problem parameters and illustrate these findings using highly erratic empirical demand data.