Allocation of scarce promotion resources among competing brands in a portfolio is a key strategic planning decision in pharmaceutical marketing. Typical business questions include:
- How effective were specific promotion tactics in the past as measured by return-on-investment (ROI)?
- Can the promotion budget for particular tactics for specific brands be modified going forward in order to efficiently increase revenue?
Marketing mix (MMx) – a series of statistical, regression models - estimates the level of return associated with each level of promotion spend. The model generates response curves, for each tactic and brand across the portfolio. The response curves are the inputs into a non-linear programming (NLP) model that solves the portfolio optimization business problem.
At Bayer Healthcare, we integrated MMx statistical models with an NLP optimization model. Our solution approach identified opportunities to both grow top-line revenue and profitability by optimally investing in various promotion tactics.
Moshe Rosenwein is interested in the application of optimization models and methods to business problems across many application areas: pharmaceutical marketing and sales, customer relationship management, e-commerce, call center operations, telecommunications network design, and supply chain optimization. He spent his 34-year career at AT&T Bell Laboratories, Medco Health Solutions, Novartis, Eisai, and, currently, Bayer. In his current role, Rosenwein is responsible for optimizing the allocation of promotion resources – including sales force personnel and digital media – across the Bayer portfolio.
Rosenwein received his B.S.E. in Civil Engineering on 1980 and a Ph.D. in Decision Sciences from the University of Pennsylvania (Wharton) in 1986.