Today we’re diving into the resurgence of media mix modeling (MMM). With privacy concerns shaking up attribution methods like last-click, brands are turning to more holistic approaches. But is MMM really the silver bullet marketers have been waiting for?
—Elena
The 3 largest ad platforms are leaning into media mix modeling.
Amazon and Meta both released MMM tools to help their advertisers better measure results after attribution through cookies and pixel data became challenging. Now, Google’s said they’re following suit.
MMM is best used as one model of many.
Media mix modeling (MMM) has been around since the 1950s, but until recently, it was reserved only for big brands with massive budgets. Today, new models like Meta’s Robyn have democratized MMM. Of course, this new accessibility comes with both opportunities and challenges.
MMM has pitfalls. While MMM offers a holistic view of marketing channel performance, it's not without limitations. It struggles to measure long-term brand building and can be biased against channels like TV, which drive future demand alongside immediate sales.
Data and time. Quality MMM requires extensive historical data and a substantial time investment. Brands need to be aware of these prerequisites before diving in, or the quality of their analysis will suffer.
The future of MMM. Looking ahead, AI could enhance MMM by providing additional data sources and aiding in data cleaning, but human interpretation and decision-making will always remain crucial.
MMM is a valuable tool but using it as one model among many will provide the most balanced and realistic view of marketing effectiveness.
"Why Facebook, Google and Amazon Are Embracing Media Mix Modeling”
This article from John McDermott for Adexchanger recaps the latest MMM developments and explores concerns that the big ad platforms are once again ‘grading their own homework.’ Read the article.
Measurement enables optimization.
“You can’t manage what you can’t measure.”
—Peter Drucker, father of modern business management
