Part 1: Introduction to MMM
What is Marketing Mix Modeling and how it has developed over the years
Introduction to MMM
Lesson 1.2 - History of MMM
In this lesson (4:23)
- Why MMM was created
- How it has developed throughout the decaded
- Why MMM is suddenly becoming popular
The dawn of MMM
The term “Marketing Mix” was first introduced to us by Neil Borden, who stated that “An executive is a mixer of ingredients, who sometimes follows a recipe as he goes along, sometimes adapts a recipe to the ingredients immediately available, and sometimes experiments with or invents ingredients no one else has tried”.
This happened during the "Mad Men era" and the rise of advertising agencies, which had drawn a lot of attention to the power of marketing.
But while the Don Drapers and other marketing rockstars had remarkable talent in creating impactful marketing campaigns, companies noted that media had a huge impact on campaign sales as well. The problem was no one could distinguish the effects from each other!
Scientific breakthrough
In early 70s, a group of statisticians at the University of Chicago wanted to go further and quantify this impact scientifically, and so they created the first statistical models to measure the impact of marketing on sales. And so Marketing Mix Modeling was born!
The early models were based on simpler regression analysis, which aims to estimate the relationship between outcome variable, often sales or traffic, and explanatory variables, such as marketing activities or external factors.
The most common regression analysis is linear regression, where you find a line that fits most closely the data.
The second coming of MMM
The story of MMM takes a bit different turn in the 2000s with the rise of digital marketing.
Digital media, such as social media channels, digital display and Search Engine Marketing, featured new ways to track the consumer behaviour online and target the activities based on it. Attribution models challenged the status quo of marketing measurement, and MMM became more of a luxury for the biggest companies that had the resources for it.
Cloud computing and machine learning enabled innovative startups to automate majority of the analyses in an MMM project, transforming MMM into a cost-effective & continuously updated software.
Today, subscription-based MMM provides a shortcut into the powerful insights advanced time series analysis unlocks, but without hefty price tags or violating privacy regulations!