Google Launches Meridian GeoX and Studio to Bring Causal Incrementality to Its Measurement Stack

Google announced three measurement products on May 5, two weeks ahead of Google Marketing Live 2026: Meridian GeoX, an open-source geographic incrementality tool; Meridian Studio, a Google Cloud-powered enterprise platform for managing marketing mix models at scale; and an upgraded Data Manager with a new map view and expanded API signals including store sales. Together, the announcements represent Google's most significant measurement infrastructure push since Meridian's global launch in January 2025.

The timing is strategic. With GML scheduled for May 20 and the upfront season in full swing, Google is signaling that its measurement ambitions extend well beyond last-click attribution and into the territory currently occupied by independent MMM vendors and incrementality testing platforms.

Meridian GeoX: Open-Source Incrementality Testing

The headline product is Meridian GeoX, an open-source geographic incrementality solution that lets advertisers run controlled experiments across regions to measure whether a media channel actually caused the business outcomes attributed to it. Unlike correlation-based modeling, geo experiments provide causal evidence — the gold standard for measurement confidence.

GeoX is designed to be publisher-agnostic and transparent: the codebase is fully auditable, and its signals feed directly into Meridian, Google's open-source marketing mix model. That integration is the key differentiator. Where standalone incrementality tools produce isolated experiment results, GeoX results calibrate and validate the broader MMM, creating a feedback loop between modeled predictions and experimental proof.

As AdExchanger reported, the practical value is that advertisers no longer need to choose between the scalability of MMM and the rigor of controlled experimentation. GeoX brings causal validation into the same ecosystem where budget allocation decisions are being made.

Google confirmed that GeoX will begin testing later in 2026, with broader availability expected to follow. The tool joins an existing portfolio of geo-experimentation resources that Google has maintained on GitHub, but formalizes them as a named product within the Meridian ecosystem with dedicated documentation and partner support.

Meridian Studio: Enterprise MMM at Scale

The second major announcement is Meridian Studio, a Google Cloud-powered enterprise platform that addresses the operational complexity of running marketing mix models at scale. While the open-source Meridian framework has been available since January 2025 and tested with hundreds of brands globally, Studio targets sophisticated measurement teams that need to customize, manage, and scale high-volume models using richer data signals.

Studio provides infrastructure for teams managing multiple models across brands, regions, or business units — a common pain point for agency holding companies and multi-brand advertisers. The Cloud backing means teams get managed compute, version control for model configurations, and access to expanded signal sets without maintaining their own infrastructure.

This is Google's play for the enterprise MMM market currently served by vendors like Analytic Partners, Ekimetrics, and Nielsen. By offering a Cloud-powered platform on top of an open-source foundation, Google is attempting to combine the transparency advantages of open source with the operational convenience that enterprise teams require. Over 20 measurement partners are already certified to support Meridian implementations.

Data Manager Map View and Store Sales Signals

The third piece is a Data Manager upgrade that adds a visual map view showing how data flows from platforms like BigQuery, Google Drive, HubSpot, and Shopify into Google's advertising ecosystem. The expanded API will also support store sales signals, giving advertisers a way to connect offline transaction data to their campaign measurement without custom engineering.

This is less flashy than the Meridian announcements but potentially more impactful for mid-market advertisers. The ability to visualize data connections and diagnose integration issues reduces the technical barrier to feeding first-party data into measurement models — a prerequisite for both attribution and MMM to work accurately.

The Strategic Picture

Google's measurement stack now covers three layers that the industry has increasingly recognized as complementary: attribution (Google Analytics 4 and conversion modeling), marketing mix modeling (Meridian and Meridian Studio), and causal experimentation (Meridian GeoX). As MediaPost noted, this positions Google to offer a complete measurement framework for what it calls the "AI-first era" of advertising.

The open-source strategy is doing real work here. By making Meridian and GeoX auditable and publisher-agnostic, Google sidesteps the obvious conflict-of-interest objection: that a platform grading its own homework cannot be trusted. Advertisers and their partners can inspect the code, validate the methodology, and run experiments that include non-Google channels. Whether that transparency is sufficient to overcome buyer skepticism remains an open question, but it is a more credible approach than proprietary black-box solutions.

The competitive implications are significant. Independent MMM vendors that have built businesses on the premise that Google's measurement tools are biased or incomplete now face an open-source alternative backed by Cloud infrastructure and a growing partner ecosystem. Incrementality testing platforms like Northbeam, Measured, and Stella face a free, open-source competitor that integrates directly with the largest ad platform in the world.

What This Means for Measurement Teams

If you are already using Meridian for MMM, plan for GeoX integration. The ability to validate your model's channel-level estimates with causal experiments is the single highest-value addition to any MMM workflow. When GeoX enters testing later this year, early adopters will have a meaningful advantage in model accuracy.

Evaluate Meridian Studio against your current MMM infrastructure. If your team is spending significant engineering time on model management, versioning, and scaling, Studio may reduce operational overhead. But assess the Google Cloud dependency and data residency implications before committing.

Use Data Manager Map View to audit your measurement foundation. Many measurement problems are actually data plumbing problems. A visual map of your data flows will surface gaps, broken connections, and missing signals that undermine model accuracy regardless of which tools you use.

Do not wait for GML on May 20 to start planning. Google typically announces direction at GML and ships in the following months. The measurement tools previewed on May 5 signal where Google's infrastructure is heading. Teams that start evaluating now will be ready to test when access opens up, rather than starting their assessment after everyone else.

The broader signal is clear: the measurement industry is converging on a three-layer framework of attribution, MMM, and incrementality as complementary rather than competing approaches. Google is betting it can own all three layers. Whether that bet pays off depends on whether advertisers trust a platform to measure itself — even with open-source transparency.