Introduction
Media Mix Modeling (MMM), also known as Marketing Mix Modeling, has been used for decades because of its effectiveness in guiding marketers and advertisers during portfolio planning, scenario modeling, and understanding the long-term interactions between channels and the impact across different marketing tactics.
While this legacy measurement approach is still effective for these purposes, with the rise of commerce media (a recent outlook for 2026 from IAB forecasts a 12.1% growth in commerce media spend) and proliferation of retail media networks (RMNs), advertisers now have more access to deterministic purchase data and closed-loop feedback than ever before.
With this structural shift, more sophisticated measurement is required for brands to make effective media investment decisions.
In fact, relying primarily on MMM for retail media can lead to improper budget allocation and hinder business growth.
Why MMM Worked
MMM has been used for decades to measure the effectiveness of different marketing channels and campaigns in generating sales.
By using historical, aggregated data to quantify the impact of different marketing channels, this approach succeeded particularly in environments where direct, granular, or fully trackable data didn’t exist, such as when sales only occurred offline. MMM can incorporate proxy metrics and external factors to estimate the incremental effect of marketing despite imperfect sales data.
It was the default option for reach-based media channels like traditional TV, radio, and OOH that don’t generate trackable clicks or direct conversion paths. MMM was designed to measure these channels by looking at correlations among changes in spend, reach, and business outcomes over time. Many MMMs also compared marketing tactics to other business developments like pricing action implications and gain/loss of distribution on total sales.
As privacy and technology changes reduce addressable measurement signals, MMM often becomes the default option for assessing performance. By using aggregated historical data, it can quantify channel impact even when impression-to-conversion visibility is limited.
Why Traditional MMM Needs to Evolve for Retail Media
- Always-on Activation Creates Invisible Impact
Retail media is designed to be always on. Budgets rarely turn off or fluctuate in ways that classic MMM can detect. MMM needs spend variation to reveal impact—no variation means no measurable incrementality. The result is simple: the model sees constant activity, so it infers no effect and pushes into the baseline.
Buying dynamics make this harder. Traditional media is planned on CPMs, impressions, or GRPs—clean inputs for MMM. Retail media introduces CPC, CPA, redemption-based buying, and algorithmic auction dynamics that shift multiple times in a day. These modalities don’t translate cleanly into the exposure‑based inputs MMM expects, so their performance is structurally undervalued.
The consequence: Retail media looks inefficient in MMM, and brands underinvest in one of their most effective commerce-driving channels. - Diverse Commerce Signals Obscure True Contribution
Most retail media networks sit on top of fragmented commerce ecosystems. A platform like Instacart connects more than 2,200 retail banners on Instacart Marketplace with 310+ grocery ecommerce sites.MMM, however, is usually fed only point‑of‑sale data from a limited set of brick‑and‑mortar retailers.As a result, the model is blind to large portions of the actual impact. When a RMN drives national, omnichannel sales—including delivery, pickup, marketplace transactions, and cross‑retailer spillover—MMM simply cannot parse the signal from the noise. The lift is real and substantial, but it has nowhere to land inside the model.
The consequence: MMM does not attribute enough to retail media because the underlying sales impact is diversified across data sources it doesn’t ingest or that have too small an impact on sales to achieve the needed sufficiency for the model. - Estimates Are Informing Decisions When Facts Are Available
MMM was designed to estimate sales impact in environments where direct measurable impact did not exist. Retail media now provides the opposite: closed-loop, SKU-level, transaction-level measurement.With this direct evidence available, correlational modeling is not only unnecessary, it is outdated. While MMM infers impact, closed-loop reporting and incremental sales lift testing prove it causally, taking into account all other factors. Additionally, MMM typically aggregates retail media into a single tactic, obscuring campaign-level performance nuances.
The consequence: Brands can end up making strategic investment decisions based on estimates instead of the factual, causal evidence retail media measurement provides.
The reality is, MMM is not going away. What’s needed is the adoption of a set of best practices and standards for retail or commerce media measurement that define how to measure the true business impact of commerce media investments. One example is the ‘Guidelines for Incremental Measurement in Commerce Media’ released by IAB.
“The future of commerce marketing is here—where advanced causal closed-loop retail media measurement reveals the true incremental sales impact of campaigns. By grounding strategic decisions in this level of data-driven accountability and iROAS clarity, brands can invest with confidence, optimize media with precision, and accelerate growth. I’m proud of our work with the IAB to showcase the impact of closed-loop measurement as the top performance-driven standard for commerce media.”
— Kimberly Sugden, former Senior Marketing Manager, PepsiCo
What Should CPG Marketers Do Now?
- Reposition MMM in the Measurement Stack
MMM still has value, but not as the primary decision engine for retail media. It should guide long‑range portfolio planning, budget guardrails, and high‑level mix scenarios—not dictate the precise level of investment allocated to closed‑loop channels. Retail media operates with deterministic signals, causal testing, and verified transaction-level outcomes. MMM cannot see that full picture with just input and outputs, so it should complement as a secondary results source, not compete with, primary platform‑level measurement.
- Adopt Measurement Built for Closed-Loop Environments
Retail media networks provide what MMM was created to approximate: a deterministic link between exposure and purchase. CPG brands should treat incrementality testing—control/exposed designs, randomized experiments, geographic lift tests—as the primary foundation, not optional enhancements. These methodologies are purpose-built for environments where media and commerce are tightly integrated. They provide the causal clarity that MMM is not designed to produce for retail media.
- Audit Current Decision Logic
If MMM consistently under-credits retail or commerce media relative to platform-reported or experimentally validated outcomes, the issue is almost always a measurement mismatch rather than a media performance issue. Retail media is visible to closed-loop systems and invisible or distorted within MMM structures built for traditional media. Large gaps between MMM outputs and causal retail media results are a signal to evolve your methodology—expanding inputs, adjusting model specifications, or redefining how commerce signals are captured.
“As media channels evolve from inferred exposure to observable outcomes, measurement has to evolve with that. The most effective marketers will no longer ask ‘which methodology is best?’, but ‘which methodology is appropriate for the signal, channel, and decision at hand?’”
— Collin Colburn, VP Commerce & Retail Media, IAB
The Bottom Line
In a world where commerce data is now observable, continuous, and verifiable, CPG marketers shouldn’t rely on MMM as the primary lens for evaluating retail media. While MMM still plays an important strategic role in long‑horizon planning, commerce media is different. Retail media networks offer deterministic purchase signals, closed-loop reporting, and causal lift testing that reveal the true incremental value of spend far more accurately than correlational models ever could.
Don’t undervalue and undercount the impact of one of the most powerful growth engines in the CPG playbook with a legacy measurement model. Marketers who modernize their approach will unlock the full business impact that commerce media is already capable of delivering.
This content is for informational and educational purposes only and is not intended to predict or guarantee advertising performance or outcomes.
