Behavioural economics emerged as an academic field challenging classical economics’ assumption of rational actors. Its purpose is to better understand decision-making and predict human behaviour, how we can be swayed by macro issues like politics, or micro issues like the number of products we have to choose between. In short, behavioural economics helps us predict illogical human behaviour.

By adding the power of behavioural economics to next-generation MMM, data teams can extract real commercial value from their data. On one hand, MMM answers the classic questions of ‘how much did the client spend, and what incremental lift did each channel deliver’? On the other hand, behavioural experiments and testing asks a different but complementary question of ‘how did we influence that choice’?

Moving the two disciplines closer

In practice, this means using behavioural insights to design specific nudges, offers, frames and experiences, then feeding what is learned back into MMM. Rather than treating behaviour as a black box where information is locked away, the two approaches are designed to work side by side. Behavioural work explores why and how people respond, while MMM quantifies the incremental effect and optimises the mix.

The goal, however, is to get to a point where they become one solution, as opposed to seeing them as two different solutions. By feeding behavioural inputs into a MMM learning loop, marketers can make continuous incremental improvements.

In a UK retail project, Incubeta incorporated nudges like discounts and two-for-one offers directly into the MMM as explicit variables. The team fed in discount levels and promotional calendars, enabling the model to answer key questions such as what sales impact comes from a 15% discount? Or how does switching to a two-for-one special change results? For CMOs, this reveals not just channel performance, but how choice architectures (such as pricing frames, messages, and bundles) drive outcomes and guide smarter budget decisions.

Understanding the skills merge

Setting up teams to leverage these opportunities is more than just having the right tech on hand. It requires the right mix of skills, processes and collaboration. To get an organisation to where it needs to be requires commercial behavioural experts working hand in hand with data scientists and AI specialists.

In this model, behavioural economists identify why a particular bias or pattern exists. Data scientists then use AI and modelling to identify which customers are currently exhibiting that bias, so that targeted nudges can be designed and deployed at scale. MMM then closes the loop by quantifying the incremental impact of those nudges on sales and other KPIs.

To achieve this, CMOs need a deliberate way of stitching together behavioural insight, experimentation and MMM into a single measurement journey. That typically means partnering with a specialist or hiring for a small number of commercial behavioural specialists, ensuring your data science partners can translate their hypotheses into model‑ready variables.

Behavioural economics‑infused MMM demands comfort with probabilistic outcomes. That doesn’t mean every marketer needs a statistics degree, but it will require structured up‑skilling around concepts like incrementality, bias, lag effects and uncertainty.

Unlock your data and choose your partners

Finally, CMOs need to whip their data into shape. Legacy data infrastructure can make it incredibly difficult to capture behavioural signals in a structured, model‑ready way. In some organisations, behavioural work lives inside UX, product, or even HR ,and rarely connects to marketing analytics. What’s more, there is also a risk that behavioural techniques are used in one‑off campaigns instead of being wired into a repeatable test‑and‑learn loop that MMM can exploit.

The opportunity to use behavioural economics and MMM together to build a far more realistic view of how people choose, and what really shifts those choices, is enormous. But it absolutely depends on designing the right collaborations, insisting on integrated experimentation and modelling, and ensuring your team’s comfort with probabilistic, behaviour‑driven decision‑making. In almost every instance, the best results will come from finding a specialist partner who can help get you started, help you realise results, and then deciding from there if you want to build out your own in-house teams.