What is Media Mix Modeling?
Episode 01
Episode 01 Transcript
Hi, I'm John Colias, Senior Vice President of Advanced Analytics at Decision Analyst, here with the Media Mix Minute. The Media Mix Minute is a monthly video series dedicated to discussing media mix modeling. Our first topic today is "What is media mix modeling?"
Media mix modeling uses historical data on sales activity and media activity in order to quantify the return on investment of media activity. So the media activities such as—TV advertising, radio, print, online, search engine, mobile apps, and more—are simulated to increase, to determine the amount of increase, in sales caused by these media activities.
The return on investment can be done, the simulation can be done, not just for one type of media but for an entire media plan. So the amount of spending by media type across month of the year to fill out your annual budget. The simulation can be done for this entire media plan in order to optimize the spend and the media activities (to optimize your spending to increase your sales to the optimal level).
This can be done through 'what if' simulations. Try a particular combination of media, activities and spend levels to determine what the level of sales should be through the simulation. And then try a different combination of media activities and determine what the sales is for that. And through this iterative process, you can maximize the sales by optimizing your mix of media activity across the different types of media.
And this kind of optimization can be accomplished, not just through a manual procedure, but by applying a nonlinear programming algorithm. Which is a program which optimizes automatically the combination and allocation of the media activity, in order to reach the highest level of sales, to hit a peak of a sales hill. And ultimately that's the goal of media mix modeling, which is to allocate your spend across the media types in order to get the greatest reach, the greatest amount of sales.
Presenter
John Colias, Ph.D.
Senior VP Research & Development
As a leader with both university teaching and business consulting experience, John focuses on predictive modeling, prescriptive analytics, and artificial intelligence. As Senior Vice President, Research & Development, at Decision Analyst, John combines academic and business interests to help analytics professionals by offering cutting-edge analytic solutions tempered by business realism. He holds a doctorate in economics from The University of Texas at Austin, with specializations in econometrics and mathematical modeling methods.
