and Stanislaw Ulam, attempted to model a complex situation that still adhered to the laws of physics in order to forecast neutron behaviour. 2,3 P. Bruce McDonielBruce is the Vice-President of Database and Analysis at Summit Marketing Group in St. Louis, Missouri -an integrated marketing service company. Bruce has been in the database marketing industry for over ten years holding positions at GMAC and Mellon Bank. He is currently completing an Executive MBA at Washington University. J. Patrick MonteleonePat is currently the Senior Modeling Analyst at Summit Marketing Group. Pat has 12 years of experience in industry and academia and has held several positions in financial service marketing. Pat also holds a PhD in Management and Decision Sciences from Saint Louis University.Abstract This study is the first in a two-part series that proposes the use of simulation and optimisation modelling in the direct marketing environment to improve marketing results. The first part of this series outlines the benefits of probabilistic simulations of the business process to improve forecasting information. The second part focuses on how an optimisation model, after simulation of the business process, can produce significant gains in direct marketing efficiency.The problem addressed in this paper focuses on creating a forecast for a marketing effort, comparing different marketing allocation scenarios and determining which one produces the most revenue per marketing dollar. The study proposes a solution to this direct marketing problem by making contributions in two areas: 1) A forecasting method using Monte Carlo simulation techniques; and 2) an adaptation of return on promotion as the key indicator of marketing results and its use to compare alternative mailing scenarios.A simulation methodology is proposed and a simulation model is presented. The simulation model is then compared to deterministic methods that use point estimates and evaluated against actual results.

No abstract