Unlocking the Hidden Potential: Why Embracing Uncertainty in Marketing Mix Modelling Matters

Introduction

What if the most valuable insights from your Marketing Mix Model (MMM) are hiding in what we usually consider uncertainty or noise? What would happen if we could look at the results of the MMM models we are using today from a different angle?

The Power of Bayesian Hierarchical Models

Those of us who have made MMM models using Bayesian hierarchical models have seen that these models provide a wealth of information about each of the parameters we set up. By applying rigorous and widely validated statistical techniques, we choose the mean (sometimes the median) of the posterior distribution as the value of influence for a certain channel.

However, Bayesian analysis outputs a probability distribution of values, and the tails are frequently large with rare occurrences and exceptions. If we underestimate the information contained in these tails, we lose valuable opportunities.

Uncovering Valuable Insights

In the expression of those long tails, if we look through the proper lens, we can find very valuable insights. The basic idea for which most users utilize MMM models is to quantify the impact of marketing efforts accurately.

Embracing the uncertainty in our models not only enriches our understanding but also enhances our decision-making processes in marketing strategies.