Custom Attribution Modeling Using GA4 + BigQuery | Part 2 - Using Markov Chain

This is part 2 in the series on BigQuery-GA4 custom attribution modeling. It's highly recommended that you watch part 1 first:
https://youtu.be/qastDPgPVLU?si=paShYw7l6_T64bI7

In this second installment in the series, I discuss how to leverage Markov Chains method, a probabilistic model, for attribution. Markov Chains have both merits and de-merits, as does any other method, when it comes to custom attribution modeling. But it is one of the most widely used methods in marketing measurement, not just limited to MTA but even MMM (marketing mix modeling) in the form of MCMC.

The approach is broken down into 2 main parts:
1. Calculation of transition probabilities
2. Removal effects
and finally the attribution score of each channel based on the removal effects.

I have also shared a sample code in the doc, but if you are interested in a very robust application of the Markov method to attribution modeling, get in touch with me over LinkedIn (See bio) or comment below.

Here's the guide doc:
https://docs.google.com/document/d/1_yluEi-7O5F6nmkR3LGNXbsw6cuvr7mKeSBog1jGSG8/edit?tab=t.0

Medium Articles that served as inspiration:
https://medium.com/@diptim_99684/a-beginners-guide-to-channel-attribution-modeling-in-marketing-using-markov-chains-with-a-case-3a9793945395

https://medium.com/@aditya2590/multi-channel-attribution-modelling-with-markov-chains-fbf3bdab2ca8

https://medium.com/@akanksha.etc302/python-implementation-of-markov-chain-attribution-model-0924687e4037

Видео Custom Attribution Modeling Using GA4 + BigQuery | Part 2 - Using Markov Chain канала Marketing Analytics With Kisholoy