New product market analysis and development pipeline management: The marketing and R&D interface

Proper R&D and marketing interfacing has been shown to improve the success of a firm. There is, however, a lack of quantitative tools managers could use for this purpose. I have developed a set of tools in this dissertation to assist managers to address this problem. The first tool is a normative model for structuring new product development (NPD) pipelines under uncertainty. This pipeline problem is observed in situations when no single development approach will necessarily lead to a successful product and managers choose to fund simultaneously multiple approaches at each development stage. The structure of the optimal pipeline is driven by cost, probability, and the expected profitability. I illustrate the workability of the model by applying it to a number of real-world scenarios in the pharmaceutical industry, and by comparing its normative pipelines recommendations against actual pipelines. I also present general qualitative insights for one-stage and two-stage NPD. The results suggest, in general, that the pharmaceutical firms I studied employ narrower pipelines for their new drugs development than they should, and thereby they underspend on R&D. The second tool is a category level model that captures the diffusion patterns of all new products within the same category. It explicitly models a consumer's pre-trial and post-trial utilities for each product and captures decision-making process for products bought jointly by two parties. The model has been implemented using data from pharmaceutical industry. Judging by both fit and 6 months-forward forecasting, I found our model often performs better when joint decision making structure is implemented, and both models performs better than a benchmark model. In most cases, the post-trial utility is larger than pre-trial utility, lending support to the common practice of spending on activities needed to get potential users to try a new drug. The third component of this dissertation demonstrates how prelaunch scenario analysis could be done. Asides from its main objective (marketing-R&D interface), each component of the dissertation could be used separately for other interesting marketing problems.