Ep.1 - Advanced Marketing Analytics with Marketing Mix Modeling: Science behind Marketing Investment

There is science behind advertising investment, and this science is called "Marketing Mix Modeling" (MMM).
It is a Marketing Analytics Framework that relies on the application of advanced statistical methods (econometrics) on historical data to understand the impact of each individual sales driver, measure ROI and predict future performance.
Learn about the world of Marketing Mix Modeling, how it fits into your entire marketing strategy and how it has become an essential tool for data-driven decision making.
Outline of episode 1:
0:00 Intro
2:34 What are Marketing Mix Models
3:45 MTA vs MMM
7:31 What to use Marketing Mix Models for
9:22 What are the factors to consider for these analyzes
11:42 The MMM analysis process
15:01 How much data do we need?
15:45 How to structure the data
17:00 Intro to AdStock and Diminishing Returns
20:00 Learning algorithms
21:45 Analysis output: Measurement
25:18 Final notes
TEMPLATE: bit.ly/excelmmm
Join the Marketing Mix Modelers Community: https: // groups/marke ...
Find out on:
Get started with MMM : bit.ly/3C0YEfT
Growth Agency Website: bit.ly/HybridaYT
Linkedin: https: //www.linkedin.com/in/gabriele -...
Source: www.saxifrage.xyz/post/econom...

[@ThiagoFernandes-sk3nk]

Is there a video where you explain how to achieve de MMM decomposition into different variables? Your videos are amazing!

[@cassandra4533]

Unfortunately we don't have a decomposition over time tutorial yet.
but
If you add new columns next to your dataset and multiply each transformed media variable (after adstock and diminishing returns) by its coefficient (from the linear regression) you can get the contribution of each variable and analyze it in a decomposition chart

[@kostaeliseev9960]

you are a ninja! thanks!

[@cassandra4533]

thank you a lot man

[@dotndotful]

Might be a stupid quesion but how do you choose alpha & beta values for the transformation?

[@cassandra4533]

Hello! That's the one of the trickiest parts. In traditional MMMs analysts would just try and try again until they hit a combination they would like. In Cassandra we apply Machine Learning to automatically find the optimal hyperparameters while addressing for the otherwise present human bias.

[@saumyasurendran6928]

Hey can u suggest any book to study MMM in detail?

[@cassandra4533]

Hey there, unfortunately we don't have a specific book to suggest.
There's are quite a few papers and documentations available online, we'd suggest to do some researches and start from there

[@adityamankar8910]

Hello
I need to connect with you regarding a project I am working upon. How to contact you?

[@cassandra4533]

hello, please reach out on linkedin: www.linkedin.com/in/gabriele-franco-hybrida/

[@adityamankar8910]

@@cassandra4533 please accept the request