Marketing mix modelling, or MMM, to give it its rather alliterative acronym, is a powerful technique that measures the impact of various marketing activities.
With the end of cookies drawing ever closer, this tried and tried method of gauging the effectiveness of marketing campaigns has been in resurgence.
But it’s also been evolving and Nick Yang of data consultants fifty-five takes us through how developments in tech and AI make it even more fit for purpose, indeed helpful, in our soon to be cookieless world…
Imagine planning an outdoor event. What would you do if the weather forecast was only available once a year? How would you account for unexpected weather events?
This analogy draws parallels to traditional Marketing Mix Modelling (MMM), a tool that offers valuable insights but lacks transparency and agility.
While it has been effective in providing a general overview of marketing activities’ outcomes, it has faced challenges, particularly in the era of increasing data privacy concerns and rapidly changing consumer behaviours.
Evolving Landscape of MMM
Marketing Mix Modelling dates back to the 1950s and has evolved alongside the marketing landscape. It originally helped advertisers measure the impact of their campaigns across various channels.
However, the advent of cookies and browser-based tracking in the 2000s shifted marketers’ attention towards user-level attribution.
Recent years have witnessed a significant shift due to global privacy regulations like GDPR and CCPA. With only a fraction of users agreeing to share additional data via Cookies, and Google is scheduled to end this type of tracking completely in 2024 (while Apple already has), MMM is having a resurgence.
As consumers demand greater control over their data and tech giants prioritise privacy, Next-Gen MMM has evolved, poised to revolutionise how businesses approach marketing performance measurement in line with privacy regulations.
Traditional MMM vs. Next-Gen MMM
Next-Gen MMM represents a leap forward from its predecessor. Traditional MMM had limitations: infrequent reporting, lack of granularity, and high costs.
It operated as a “black box” solution, with minimal insights into its methodology.
In contrast, Next-Gen MMM capitalises on data science advancements and offers:
- Privacy-first Approach: In an era of heightened data privacy, Next-Gen MMM aligns with regulations, ensuring ethical data
- Holistic Decision-making: It provides a comprehensive view of marketing effectiveness, including offline marketing & sales, enabling better resource allocation.
- Dynamic Forecasting: Next-Gen MMM accommodates advanced scenario planning capabilities, facilitating agile decision-making.
- Deeper Insights: With greater granularity and frequency of modelling outputs, it delves into detailed insights that drive strategic investments.
- Actionable Recommendations: It goes beyond data analysis, providing practical suggestions for effective investments.
- Accessible Dashboard: The solution can integrate into any BI solution to ensure that it is accessible to all stakeholders, fostering collaboration.
Additionally, with advances in data science techniques, Next-Gen MMM is now a more cost-effective option than it was previously, which makes it an attractive investment for businesses seeking additional clarity and agility in their marketing strategies.
Making the transition
Transitioning to Next-Gen MMM requires a strategic approach.
While each business is different and needs to tailor their tracking and analysis tools to their needs, we at fifty-five have pioneered a simple, four-step methodology to help set up and deploy an effective solution, encompassing three central pillars for optimal results: data & insights, artificial intelligence and expertise.
It is critical to pull relevant information from rich and trusted first and third-party sources – but to really make the most of it, your model needs to combine expertise and artificial intelligence to drive deeper analysis and automation that delivers actionable, pragmatic recommendations.
To get there you need to:
- Understand the core datasets that are most relevant for your brand and ensure their reliability
- Use these datasets to build a fully customised, machine learning enriched set-up tailored to your requirements
- Ensure that your model is fully transparent and, if possible, set up entirely in-house to enforce your ownership over your data
- Integrate your internal teams into the process so that they can manage and maintain the MMM solution themselves in future
It’s important to take each step as it comes, rather than trying to solve everything at once. Following this process should deliver actionable outputs that empower your team to run simulations, adapt strategies, and make informed decisions in real-time to drive ROI uplift.
When should you be using Next-Gen MMM?
No two companies are the same and therefore every solution should be different. Your tailored model could include anything from performance and real-time monitoring to experiments, forecasting and scenario planning.
Identifying the applications that will be most important to your organisation will shape your solution, but once you are up and running you can continuously adapt your initial set-up to take account of evolving requirements.
At fifty-five, we work with companies to customise their Next-Gen MMM, listening to their needs and challenges to define which modules are needed and analysing the results to ensure that the model is delivering exactly what you need.