Models to Make Your Head Spin
With the title "Agent-Based Modeling Workshop: Simulations: The Future of Marketing & Media Mix Planning," I should have known that I was getting into an academic discussion.
Agent based modeling (ABM), for the uninitiated, is a computer-based tool that will help you better plan. This is how Peter Storck, Presiden of Points North Group, laid the landscape on a very esoteric topic, but apparently pragmatic approach to marketing. This is the same approach taken to for military simulations and predicting the path of communicable diseases. So why not use complexity science for marketing?
Storck walked through a visual reminiscent of the SIMS, or perhaps Second Life, in which players represent classes of people, and we observe how they interact with the brands in question in an enviroment, such as an office. Predicted interactions are determined by insight gleaned form large amounts of data that are provided by the client, such as syndicated data, client data, secondary sources, heuristics.
For example, we might know that one class of consumers is more likely to influence others, or that another class only drinks Pepsi when she wants a break.
From here, Storck then handed the floor to Damon Ragusa, CEO of ThinkVine, and a statistician through and through. He walked through complex methodology that makes one's head spin. He verified that the data and scenarios are both replicated and calibrated (a good thing) and that as a result we are able to determine future behavior and potential future behavior.
