Data scientists regularly have to demonstrate impact: How well does your work justify your current and future compensation? 

Large companies may have advertising budgets in the hundreds of millions of dollars so that even small efficiency improvements can be highly valuable. 

Enter the marketing mix model. The MMM ingests a time series of advertising spend and impressions across channels and simultaneously estimates ROI across channels, providing opportunities for optimization. 

Google recently made their MMM available to the public, and I introduce the model in this video: https://youtu.be/X-4BvG1IDiA

Despite being 'the holy grail' for marketing teams, MMMs are not without controversy. Marketers often chase higher demand with increased spend (endogeneity) and spend across channels often trend together (multicolinearity). It's not clear how well MMM solves these problems. 

While the video I link above introduces Google's publicly available model; future videos I'll share on the channel will deep dive further, where I simulate data with known relationships and known ROI and see how well the Meridian model is able to estimate the true ROI. 

If you're interested in this type of free content, the best way to support is following here and subscribing on YT. 

#datascience 

#datascientist 

#marketingmixmodels

#mmm 

#marketingdatascience