Marketing Mix Modelling in practice 2400-ZEWW321

1. Introduction
• Course plan: meeting agenda
• Course completion requirements
• What actually is MMM?
• MMMs - econometrics or analytics? Comparison with other forms of media effectiveness measurement
• The role of an MMM project in brand strategy
• What questions does econometrics answer?
• The econometric project process
• MMM providers - market characteristics, career path, working model, team roles

2. Introduction to R, tidyverse, environment setup – part_1
• Installing R and RStudio
• Useful keyboard shortcuts
• Installing packages
• Using RStudio
• Principles of writing code in R
• Intro to Tidyverse

3. Introduction to R, tidyverse, environment setup – part_2
• Libraries within tidyverse - readr, tidyr, tibble, dplyr, lubridate, purrr, ggplot
• Comparison of ggplot vs plotly
• Modeling in R - examples of essential functions for modeling

4. Data used in MMM and its structure
• Panel data and time series
• Format of the modeling database
• Data frequency and its implications
• Choosing the appropriate dependent variable
• Methods of aggregating sales value and volume
• Data included in models using the example of FMCG category
• Numeric vs weighted distribution
• Price, promotion information
• External factors: Covid, Seasonality, Macroeconomics/Demographics, Trade, Weather, Holidays
• Media measurement methods and data providers
• What is GRP?
• Impressions or Clicks?
• What is AdStock?

5. Modeling and model interpretation (1)
• Graphical analysis of variables
• Types of models: multiplicative, additive
• Order of including variables
• Interpretation of parameters: elasticities, semi-elasticities, marginal effects

6. Modeling and model interpretation (2)
• Response Curves, diminishing returns
• Media: c-shape vs s-shape
• Arctangens transformation
• Modeling in practice - criteria for variable selection (statistics vs. business)
• Spurious regression and multicollinearity
• Statistical tests used for MMM model validation
• Holdouts

7. Model-based analyses and drawing business conclusions
• Decomposition of the dependent variable
• Baseline vs Incremental
• Dynamic vs static Waterfall
• Calculating ROIs

8. Marketing investment optimization and creating recommendations
• Optimization on concave curves - example of a simple algorithm
• Sample recommendations

9. Course completion
• Presentation of individual projects created by students (or oral exam)

10. The course concludes with the awarding of Certificates to Students:

11. The best students will be invited to a simplified recruitment process for the Choreograph Poland department, where they will have the opportunity to gain valuable experience through international projects carried out for the biggest brands.

During the classes, a series of practical exercises will be carried out, including working with data using RStudio, creating basic models, conducting key business analyses, and creating recommendations.
The course has been prepared in collaboration with Choreograph WPP Media.
Szacunkowy nakład pracy studenta:
Typ aktywności K (kontaktowe) S (samodzielne)
wykład (zajęcia) 0 0
ćwiczenia (zajęcia) 30 0
egzamin 1 0
konsultacje 2 0
przygotowanie do ćwiczeń 0 22
przygotowanie do wykładów 0 0
przygotowanie do kolokwium 0 0
przygotowanie do egzaminu 0 20
… 0 0
Razem 33 42 = 75