Simulation Example
Episode 08
Episode 08 Transcript
Hi. I’m John Colias, Senior Vice President of Decision Analyst’s Advanced Analytics Group, continuing our series on Media Mix Modeling. Today we’re going to talk about simulation, and how simulation is used to measure the return on investment of media activity.
As we mentioned in an earlier episode, media mix modeling produces an equation and that equation has inputs and outputs. And the inputs are the units of activity for TV advertising and other types of advertising. And in the output is the units of product sold. And this equation is embedded into a simulation program (can be embedded into one) where the inputs are actual dollars spent on media (different types of media), and the dollars are divided by the cost of media. For example, TV, amount spent on TV advertising dollars could be divided by the cost of TV GRPs. And it’s converted to units, and then the units are input into the equation and we simulate the output, which produces units (again of products sold) converted to dollars, converted to revenue, by multiplying times of price (or if it’s a product category an average price) and then the net revenue is produced (the cost of the media subtracted from the revenue).
So that’s the simulation program. So how do we use this to, to generate a return-on-investment measure? We do two simulations. First, we simulate an entire media plan, which would include, for example, as in the diagram, the, the spend on TV advertising of a $100,000 in January and internet advertising (in this simple example of $40,000 in January) and the rest of this spend for the rest of the months of the year. And this first simulation includes the entire media plan. And the output again is the net revenue that’s produced by that, that scenario with that full media plan.
The second simulation would be reducing the TV advertising to zero, if we want to measure return on TV advertising. And then again producing net revenue. So we take the net revenue from the first simulation minus the net revenue from the second simulation and divide it by the cost of TV advertising (the total amount spent on TV advertising) and that produces an average return on investment for TV advertising. So that’s an example of how, in media mix modeling, we use simulation to measure the return on investment for media activity.
Thank you.
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.
