Optimizing marketing planning and budgeting using Markov decision processes: An airline case study

Many companies have no reliable way to determine whether their marketing money has been spent effectively, and their return on investment is often not evaluated in a systematic manner. Thus, a compelling need exists for computational tools that help companies to optimize their marketing strategies. For this purpose, we have developed computational models of customer buying behavior in order to determine and leverage the value generated by a customer within a given time frame. The term “customer value” refers to the revenue generated from a customer's buying behavior in relation to the costs of marketing campaigns. We describe a new tool, the IBM Customer Equity Lifetime Management Solution (CELM), that helps to determine long-term customer value by means of dynamic programming algorithms in order to identify which marketing actions are the most effective in improving customer loyalty and hence increasing revenue. Simulation of marketing scenarios may be performed in order to assess budget requirements and the expected impact of marketing policies. We present a case study of a pilot program with a leading European airline, and we show how this company optimized its frequent flyer program to reduce its marketing budget and increase customer value.