Brands Turn To Marketing Mix Modelling To Measure Impact In A Data Constrained Era

There have been several reasons for the sharp growth in MMM. Most obviously, privacy regulations and platform changes have made individual‑level tracking much harder, eroding the reliability of user‑level attribution. At the same time, marketing has become exponentially more complex. Once brands only had a handful of traditional channels to choose from. Now they face a sprawling ecosystem of search platforms, social networks, streaming environments, retail media and connected TV. In this environment, MMM offers marketers an evidence‑based way to understand which parts of the mix actually drive the bottom line, at an aggregated level, and over time.

Why MMM is so important

MMM offers marketing leaders three immediate and important gains:

1. Clarity in a complex environment

   MMM helps answer questions such as: What is the marginal impact of another rand spent in TV versus search? How does out‑of‑home work alongside social? How do promotions or seasonality interact with media spend? All vital in data-driven budget decisions.

2. A move to optimisation

   When user‑level tracking is constrained, modelling at aggregated level becomes essential. While MMM doesn’t try to follow every click, it can track how total marketing activity correlates with key outcomes over time, allowing brands to reallocate budgets for maximum impact.

3. Long‑term growth

   MMM is not a short-term tool. Its value lies in reshaping the overall investment strategy, assessing how much to spend, where to spend it, and in what combination. For brands with bigger budgets, that can translate into millions in incremental revenue or savings once the model is properly embedded.

Adoption challenges

Despite its promise, there are some adoption challenges that marketing decision-makers should be aware of.

Like most things in marketing today, having access to the right data is essential, but many organisations get tripped up with this. Not because they lack data, but because it’s fragmented, inconsistent or stored at the wrong level of detail. Digital, out‑of‑home and in‑store activity often sit in separate teams and systems. Stitching all of this together into a coherent, model‑ready dataset is a major piece of the initial work. And it requires close collaboration between marketing, data teams and sometimes CTO‑level stakeholders.

Secondly, marketers should have a realistic time expectation. A proper MMM implementation typically takes one to three months to build, then several more to deploy, refine and track. Leaders should allow for six months to a year to unlock the full potential of the system.

Third, there are a number of excellent MMM tools out there, some of which are open source. However, the expertise to design, calibrate and interpret models must also be factored in.

Lastly, there are a number of human challenges to overcome. For instance, CFOs must be prepared to fund a long‑term initiative. Marketing leaders must also be willing to act on what the model says, even when that means reducing spend on favoured channels. We understand that people naturally stick to what they know, even when what they know is not working, and that conservatism can quietly stall adoption.

Overcoming the barriers

Successful adopters of MMM treat it as a change management programme, not a technical project.

On the data side, that means starting with a realistic assessment of what’s available and where the gaps are, then prioritising integration and quality over perfection.

Setting expectations is equally critical. Senior stakeholders need to understand from day one that MMM is a strategic compass, not a real‑time steering wheel.

We’ve seen companies insist on building in isolation, treating MMM as pure IP, only to end up with models that simply don’t make sense. A more pragmatic approach is to combine internal data knowledge with external modelling and strategy expertise, at least for the initial build and validation.

Finally, communication and education are essential. Demystifying MMM for non‑technical stakeholders, sharing early case studies and inviting questions all help reduce the fear factor.

Looking to the year ahead, we are planning for a sharp increase in demand for MMM, particularly among enterprise and mid‑ to upper-market brands. As media continues to fragment and privacy constraints tighten, relying on intuition and last‑click metrics will become exponentially more risky. MMM will not replace every other form of measurement. Nor will it deliver instant answers. But as a disciplined way to understand what truly drives growth, it is rapidly moving from experimental to essential.