Attribution & marketing mix modeling · Google

Meridian

Open-source Bayesian marketing mix modeling framework from Google, written in Python, for measuring channel ROI and optimizing budget.

Visit Meridian Alternatives

Meridian is Google's open-source marketing mix modeling library, released as a Python package rather than a hosted product. It uses Bayesian statistical methods, optionally incorporating geographic-level data, to estimate how much each marketing channel contributed to a business outcome and to simulate budget-allocation scenarios. It is the successor to Google's earlier LightweightMMM project and can ingest Google-specific signals such as reach and frequency data where available. Like Meta's Robyn, it is code, not a subscription: getting useful output requires a data scientist or ML engineer comfortable with Python and Bayesian modeling, plus a couple of years of weekly (ideally geo-level) spend and outcome data — there is no vendor support line if the model output looks wrong.

At a glance

Vendor Google
Pricing model Open source + paid options
Free tier Yes
Deployment Self-hosted
Open source Yes (Apache-2.0)
Best for Teams with a data scientist comfortable in Python and Bayesian statistics who want full control over their MMM rather than a managed vendor.

Pricing

Free, open-source Python library; the only cost is compute and the data scientist's time to build and maintain the model.

Pricing has not been verified yet — see the vendor's site.

Features

  • Bayesian marketing mix modeling in Python
  • Optional geo-level modeling for regional data
  • Incorporates reach and frequency signals where available
  • Budget optimization and scenario simulation
  • Successor to Google's LightweightMMM project
  • Model diagnostics and posterior-uncertainty reporting

Profile last reviewed September 21, 2026

Head to head

Alternatives

Robyn Open-source semi-automated marketing mix modeling package from Meta, built in R for data scientists to run themselves. Recast Bayesian marketing mix modeling delivered as a managed SaaS, combining statistical MMM with geo experiments.

Meridian in the index now

Terms to know

Marketing mix modeling A statistical technique estimating how much each marketing channel and external factor contributes to results. Regression analysis A statistical method for estimating the relationship between one outcome variable and one or more predictor variables. Causal inference The set of methods used to determine whether one variable actually causes a change in another, not merely correlates with it. Data scientist A practitioner who combines statistics, programming and domain knowledge to build models and extract insight from data.

Related guides

How to choose a marketing attribution or MMM tool Attribution, media mix modeling, and incrementality testing answer the same question three different ways. Pick the method before you pick a vendor. How to choose an attribution model Rule-based, data-driven, MMM and incrementality testing answer different questions — pick by what decision the answer has to support. How to measure marketing without third-party cookies First-party data, server-side collection, modeled conversions, clean rooms, MMM and incrementality tests — the methods that replace what cookies used to do. How to run marketing mix modeling What data an MMM needs, the build-vs-buy choice between open-source libraries and managed vendors, and how to validate the model with real experiments.