One key advice I give to statisticians/data scientists looking to get started on Marketing Mix Modeling (MMM) is – Know your type of Data.
A big chunk of the projects in the industry involves dealing with tabular data.
This is particularly true of Marketing Measurement and Attribution Industry.
However, not many have the knowledge of Time Series Data, Cross Sectional Data and Panel Data.
All these types of data have a temporal component.
It is thus very important to know the subtleties because the type of regression that you would apply will vary slightly depending on the data and business objective at hand.
So here is a quick explainer:
1.Time series
We are collecting data about Brand ‘XY1’ sales over the years
Year sales
2017 50,000
2018 54,000
2019 65,000
2020 45,000
๐ Key Difference: Notice above that the data is collected for the same entity ‘Brand XY1’ but at different intervals of time.
2. Cross section data
Now we are collecting not only Brand XY1 data but also its rivals. However we are just interested in the year 2020.
Brand Year Sales
XY1 2020 45,000
EL3 2020 23,000
FT3 2020 56,000
TY4 2020 24,000
๐ Key Difference: Notice, we are interested in data at a particular point in time (i.e. yr 2020). This interest in particular point in time distinguishes Cross section from Time series.
3. Panel Data:
Panel data is what you get if you mix both the Time series and Cross Sectional data. Lets say that now you want to track brand XY1 and its rivals sales over the years. This results in data like below:
Brand Year Sales
XY1 2017 50,000
XY1 2018 54,000
EL3 2017 12,000
EL3 2018 23,000
FT3 2017 32,000
FT3 2018 35,000
In summary: In MMM, attribution is the name of the game in MMM. It is hence very important to know to which time period you are attributing. Developing a good understanding of the data is one of the key steps in specifying a good MMM model.
I would highly recommend Woolridge’s econometrics book and Rob Hyndman’s Forecasting book to anybody looking to get started on MMM. Link to the books in the resources section.
Resources:
- Introductory Econometrics Woolridge – https://economics.ut.ac.ir/documents/3030266/14100645/Jeffrey_M._Wooldridge_Introductory_Econometrics_A_Modern_Approach__2012.pdf
- Forecasting Principles Rob Hyndman – https://otexts.com/fpp3/
