W3: Slides (MMM MMO)

Question 1

Define ‘Media Mix Modelling’ (MMM) and ‘Media Mix Optimization’ (MMO)

Answer 1

MMM is a technique measuring the impact of marketing and advertising campaigns to determine how various elements contribute to a goal like driving conversions. MMO involves adjusting budget allocation across marketing activities to optimize a key performance indicator (KPI).

Question 2

What is the goal of Media Mix Modelling?

Answer 2

To measure the correlation between total sales/revenue and advertising spending or impressions, and to incorporate external influences such as seasonality and promotions.

Question 3

Explain the business problem addressed by Media Mix Modelling

Answer 3

It seeks answers to strategic questions like how a firm’s marketing activities correlate to a KPI and what the optimal mix of marketing activities for a given KPI is.

Question 4

What is the difference between Media Mix and Marketing Mix Modelling?

Answer 4

Media Mix Modelling focuses on the role each media type plays in overall campaign performance. Marketing Mix Modelling takes a holistic view, considering media channels, marketing activities, pricing strategies, product distribution, and even macroeconomic indicators.

Question 5

What are the key components of a Simple Linear Regression model?

Answer 5

The model includes a dependent variable (like sales), independent variables (like price), a constant term, coefficients for each independent variable, and an error term.

Question 6

How does Multivariable Regression differ from Simple Linear Regression?

Answer 6

Multivariable Regression includes multiple independent variables, allowing for a more nuanced understanding of how various factors together influence the dependent variable.

Question 7

What are some extensions to the Simple Media Mix Model?

Answer 7

Extensions can include non-linear responses, temporal effects, and mixed models which allow for region-specific or time-specific deviations in marketing activity impacts.

Question 8

Explain the Media Mix Optimization Problem

Answer 8

It involves allocating a marketing budget across multiple channels to maximize a KPI like revenue, subject to constraints like total expenditure not exceeding the budget.

Question 9

Difference between Media Mix and Attribution Models

Answer 9

Media Mix Modelling is a top-down approach using time series data and can account for offline variables. Attribution Modelling is bottom-up, using individual user data and primarily focusing on digital advertising.

Question 10

Reflections on Media Mix Modelling

Answer 10

MMM is challenging in causal terms due to omitted/lurking variables and marketing spending not being randomly allocated. It’s more suited as a prediction model rather than for causal interpretations.