Development of Enhanced Marketing Flexibility by Optimally Allocating Sales Campaign Days for Maximizing Total Expected Sales

In retail businesses, a sales campaign is typically organized for a segment of consecutive days within a certain period so as to maximize the expected sales in the period. However, there is no theoretical foundation to claim that it would be better off to organize a sales campaign this way. By introducing the marketing flexibility by allocating N campaign days over a given period of M days, one may expect the increase of the total sales without any additional cost. The purpose of this paper is to explore this potential by establishing a systematic approach to estimate the expected total sales over a certain period, given that on which days in the period would be designated for sales campaign. A machine learning technique is employed to estimate whether or not a day belongs to a “Good Sales Day (GSD)” defined by a two dimensional ABC-analysis. Depending on whether or not the day is a GSD and whether or not the day is under sales campaign, four expected daily sales values are then obtained. An optimization problem is then formulated so as to maximize the expected total sales.