Meridian is a comprehensive, open source marketing mix modeling (MMM) framework developed by Google to help advertisers analyze and optimize the impact of their marketing investments. Built on Bayesian causal inference principles, Meridian enables organizations to evaluate how different marketing channels influence key performance indicators (KPIs) such as revenue or conversions while accounting for external factors like seasonality or economic trends. The framework provides a robust foundation for constructing in-house MMM pipelines capable of handling both national and geo-level data, with built-in support for calibration using experimental data or prior knowledge. Meridian uses the No-U-Turn Sampler (NUTS) for Markov Chain Monte Carlo (MCMC) sampling to produce statistically rigorous results, and it includes GPU acceleration to significantly reduce computation time.
Features
- Bayesian causal inference–based marketing mix modeling framework
- Handles large-scale, aggregated geo-level or national-level data
- Integrated GPU acceleration for faster MCMC (NUTS) sampling performance
- Built-in calibration support using experiments and prior knowledge
- Automated ROI estimation and marketing budget optimization tools
- Privacy-safe approach using only aggregated (non-user-level) data
Project Samples
