Google Meridian and the future of marketing measurement: What marketers need to know about Google’s open-source MMM 

With Google’s launch of Meridian, open-source marketing mix modeling (MMM) is back in the spotlight. But at FELD M, our experience shows that Meridian might not be the perfect choice for every marketer. Even when it is, MMM alone doesn’t provide the full picture. Instead, the most robust insights come from combining (triangulating) multiple methods: MMM, attribution, and incrementality testing.

Why MMM is back on the agenda

As privacy regulations tighten and third-party cookies disappear, attribution models that once seemed like the “holy grail” are losing their edge. Fragmented user journeys, walled gardens (closed digital ecosystems like Google, Meta, or Amazon that limit data sharing), and data silos make it harder than ever to connect marketing touchpoints and outcomes.

This has sparked renewed interest in MMM, which doesn’t rely on user-level data and can incorporate both online and offline channels.

But MMM is just one piece of the puzzle. Attribution and MMM are not alternative measures for the same purpose, but are meant to answer different questions. Real-world marketing effectiveness requires a layered, pragmatic approach that triangulates three methods: attribution, MMM, and incrementality tests — an approach that balances the strengths and limitations of each method.

Psst: We'll also be talking about this topic at the upcoming TDWI in Munich in two weeks.

Understanding the measurement toolbox: triangulation in practice

Triangulation means integrating three core measurement methods to get a more complete, credible view of marketing effectiveness:

• Multi-touch attribution (MTA): Delivers granular, user-level insights by assigning credit to each touchpoint in the customer journey. MTA can be powerful for optimizing campaigns in real time and understanding channel interplay. But it is increasingly limited by privacy constraints and data fragmentation, and often struggles to account for offline or long-term brand effects.
• Marketing mix modeling (MMM): Provides a strategic, privacy-friendly, cross-channel perspective. MMM quantifies how different marketing levers and external factors (like pricing and seasonality) drive business outcomes. It is ideal for budget allocation and long-term planning, but less suited for day-to-day campaign steering, and depends on high-quality, aggregated data.
• Incrementality tests: The gold standard for measuring causal impact. By comparing test and control groups, incrementality tests (such as geo-experiments or conversion lift studies) reveal the true effect of marketing interventions. However, they require careful design, can be resource-intensive, and are typically run only for selected campaigns or time periods.

Each method answers different questions, and each has its own strengths and limitations. The real value comes from combining them—using triangulation to validate findings, challenge assumptions, and fill in the gaps.

We also have an upcoming event on this topic, if you'd like to join:

Triangulation offerings on the market

A growing number of vendors now offer triangulation solutions or integrated measurement platforms. The available solutions can be grouped into three categories:

1. SaaS solutions that offer all three methods and their triangulation.
2. SaaS solutions specialized in one or two methods, typically attribution and MMM, or MMM and incrementality testing.
3. Open-source solutions, such as Google’s Meridian and Meta’s Robyn.

Which category is the best fit depends on a company’s marketing mix, overall goals, and measurement maturity.

At FELD M, we have hands-on experience with all of these platforms and help clients select, implement, and interpret the right mix for their needs. Here’s how we support you with custom MMM solutions.

Want to get started with MMM but not ready yet? You can find an example case study of how we helped our client Bexio prepare for MMM here.

Deep-dive: What is Google Meridian?

Google Meridian is an open-source MMM platform designed to help marketers measure and optimize the impact of their campaigns in today’s privacy-first landscape. Meridian stands out for its integration with Google data (like YouTube reach and search query volume), Bayesian modeling, geo-based hierarchical analysis, and open-source flexibility.

It’s a strong option for brands with significant Google investment and in-house analytics expertise.

However, like all MMMs, Meridian is not a plug-and-play solution. It requires clean, historical data and a clear understanding of both its capabilities and its limits. Compared to SaaS solutions, it also requires data science and data engineering capabilities. 

While Meridian makes it easier to use online channel data, particularly from Google, it faces the same challenges as other solutions when it comes to integrating offline channel data, such as OOH, print, or TV.

What are the pros and cons of Google Meridian?

Like any measurement solution, Google Meridian comes with its own set of strengths and limitations. Understanding these can help you decide whether it’s the right fit for your organization or if another approach might be more suitable.

Strengths:

• Open-source and transparent: You have full access to the code and methodology, allowing for customization and independent validation.
• Deep integration with Google data: Meridian provides direct access to YouTube reach, Google Search query volume, and other Google platform metrics, offering richer insights for brands heavily invested in Google channels.
• Customizable and flexible: Technical teams can adapt the framework to specific business needs, rather than being locked into a black-box solution.
• Scenario planning and Bayesian modeling: The platform allows for uncertainty quantification and incorporation of prior knowledge about media performance. The hierarchical modeling can even be done on geographical levels.
• Incrementality tests: Meridian offers the option to integrate results of incrementality tests to validate results or calibrate the model.
  
  

Limitations:

• Google-centric: Meridian’s strengths are most apparent for organizations with significant spend on Google properties; it is less effective for brands with a heavy focus on non-Google or offline channels.
• Requires data science expertise: Implementing and customizing Meridian is not plug-and-play; it demands technical skill in data preparation, modeling, and interpretation.
• Limited by data quality and availability: As with any MMM, success depends on having high-quality, consistent data over time.
• Operational complexity: Organizations without mature analytics processes may find the setup and ongoing maintenance challenging.

Is Meridian right for you?

Whether Meridian is the right choice depends on several factors.

If your marketing mix leans heavily on Google channels and you have strong in-house analytics resources (or a team like FELD M at your side!), Meridian can deliver valuable, privacy-friendly insights and support smarter budget allocation.

Other important factors include organizational leadership, ownership, a test-and-learn culture, stakeholder enablement, and the availability of key resources.

For organizations with more diversified channel strategies, limited technical capacity, or a lower level of maturity, Meridian might not be an optimal choice.

In general, any MMM should be supported by other marketing measurement methods, especially incrementality testing. The triangulation of all three approaches to marketing measurement should be the guiding principle.

The FELD M perspective: process, culture, and sustainable measurement

Triangulation is a practical necessity in today’s marketing environment. But it’s also an ambitious undertaking. Integrating MMM, attribution, and incrementality testing requires more than just technology. It demands:

• High data quality and harmonization
• Technical expertise and analytical rigor
• Clear documentation and transparent communication
• A test-and-learn culture, with ongoing validation and adaptation
• Change management and organizational enablement

We help our clients build measurement frameworks that are robust, actionable, and tailored to their business context. Our approach is grounded in independence and transparency, cutting through vendor hype to deliver measurable value.

Explore our data strategy and measurement framework.

Want to learn more?

If you’re considering Meridian, triangulation, or want to future-proof your measurement strategy, FELD M can help you navigate the options independently and pragmatically.

Explore our marketing measurement services

Book a free 30-minute call

 

Or join us at our upcoming event on July 3, 2025:

Bridging the data gap in marketing measurement

With three expert speakers, we’ll focus on the practical challenges of marketing measurement, data quality, and AI. Alongside learning about practical approaches, you’ll also have the chance to ask your questions during the Q&A session.

View event agenda

About the authors:

Dr. Isabelle Kes has advised FELD M clients across industries on data strategy and marketing measurement for over a decade. She also teaches at Hochschule München (Munich University of Applied Sciences) and IHK Nord-Westfalen (North Westphalia Chamber of Commerce and Industry).

Alexander Eiting is a consultant at FELD M and PhD researcher at TU Braunschweig, specializing in marketing measurement and data privacy. He lectures at Universität Münster.