From brilliant to actionable - It takes technical brilliance and constant questioning to achieve the truly actionable in marketing ROI

Mark Garratt
United States

Rafael Alcaraz
The Hershey Company, United States

Steve Cohen
United States


In this paper, we will talk about the way that analytics can evolve to serve the demands of resource allocation in a dynamic corporate setting. This is an on-going engagement, susceptible at any time to changes in strategy and priorities. We will describe the initial context, the mathematical tools developed, and the conversations that we started and are still developing. The contribution of this paper is not so much in the advanced mathematics of optimization but in how the authors have learned to straddle the space between the problem-solving environment of academia and the “moving target” of business.

It is well-known in the world of analytics that the last thing that anyone needs is another “tool” aimed at a problem that may be too specific or that is no longer relevant. Our thesis is that optimization is a process rather than a set of methods. The optimum can never be achieved and is a quixotic goal. What does happen is that the conversations started by marketing mix analysis and leading to optimization force everyone to adjust and clarify their language so that the results delivered have some usefulness. This means that the problem solvers, rooted in their academic training, must become generalists and learn how to trade-off precision for a widening scope. The corporate clients, meanwhile, must move beyond the role of being passive receivers of vendors’ solutions and start to actively frame their problems in ways that are new to them and new to their senior management. In the best of all worlds, we look for convergence between what senior management wants to see happen and the kinds of solutions and metrics that research can offer them.