< Back to articles

Should You Build a DIY MMM for Your Brand?

Read about the challenges of building an effective Marketing Mix Modelling in-house

Marketing Mix Modelling (MMM) is a powerful tool that can help brands analyse the impact of their marketing efforts using business KPIs. In fact, there are several reasons why a brand may choose to build an MMM solution in-house rather than rely on an external partner.

For instance, some brands may have unique data requirements that they feel cannot be fulfilled by an external partner. Building an MMM in-house also gives brands greater control over the model and the insights it generates. On the other hand, building an MMM solution in-house can be a costly and time-consuming endeavour. This blog post explores why it's best to rely on an experienced partner to provide a reliable MMM solution.

1.    Time is money: why building your own MMM solution can be a costly endeavour

Building an MMM is not an easy task. Facebook estimates that it takes between 12 and 22 weeks to build a relatively simple MMM, including data collection, model development, testing and validating. The whole process can take even longer if the data is not properly organised, cleaned and integrated. Building an MMM therefore requires a significant time investment.

2.    Automate for success: how to ensure your MMM solution is always up-to-date

Building an MMM model as a one off is not enough – an effective MMM requires automation. Data ingestion, processing and model updates should be automated where possible. Hence why the process needs to be designed so that new data is automatically collected and fed into the model for analysis. This ensures that the MMM model remains up to date and the latest data is used to generate insights.

3.    Insights that drive growth: why statistical results alone don't cut it

Once you have a model, you still need additional tools to generate insights. The statistical results alone are not enough – you need visualisations, reports and dashboards to help you make sense of the data. These will tools help you understand how each marketing channel contributes to overall sales and will help you identify which channels are underperforming. Without these insights an MMM model decreases in value.

4.    Expertise matters: how building an MMM solution requires a team of marketing measurement wizards

Building an MMM requires specific kinds of expertise – advanced statistical and data science skills are simply not enough. You need a good grasp of marketing, economics and consumer behaviour in order to build a reliable model. You also need to know which data sources to use, which variables to include and how to validate the model. Building an effective MMM therefore requires a team of experts with a variety of skill sets.

5.    Don't leave it to chance: the importance of testing your MMM solution across various brands and industries

In order to guarantee reliable results, MMMs need to be tested across a range of brands and industries. Relying on an inaccurate MMM can result in a lack of marketing efficiency and thus cause considerable additional costs. You need to ensure that the model is robust enough to be able to handle different scenarios and provide accurate insights across various brands and industries.


Building your own MMM solution can be a daunting task. It requires a significant investment in terms of time, resources and expertise. On top of that, the process needs to be automated, insights generated and the model tested to ensure reliable results. That’s why it's best to rely on an experienced partner who can provide a reliable MMM solution. In the long run, this will save you time, work and money and will enable you to make better-informed decisions.

Our Case Studies