google-meridian
Google's open source mixed marketing model library, helps you understand your return on investment and direct your ad spend with confidence.
Decision gist · record as of 2026-08-14
Yes, if you have marketing data and the infrastructure to run it. Meridian is actively maintained, has no security vulnerabilities, and solves a real business problem (channel attribution and budget optimization) with a principled Bayesian approach. The main barriers are the unclear license (verify before production use), heavy compute requirements (GPU strongly recommended), and a large dependency footprint (15 runtime packages including TensorFlow and tfp-nightly). Install in a fresh virtual environment and plan for GPU resources.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10+.
- GPU strongly recommended for practical model training; CPU-only operation will be slow.
- TensorFlow and tfp-nightly are heavy dependencies.
License · maintenance · safety
(unclear) — License status is unclear—no SPDX identifier or explicit license text is declared in the package metadata. Before adopting Meridian in a commercial or proprietary project, verify the actual license terms in the GitHub repository to understand usage and distribution rights.
last release 2026-07-22 (23 days) · last repo commit 2026-08-14 · 1,495 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 96,163 downloads/mo, #13,226 on PyPI
Alternatives
Verify before relying
pip install google-meridian
# For GPU (Linux with CUDA):
# pip install google-meridian[and-cuda]
import meridian
# Load data and instantiate a model
model = meridian.MMM(data, ...)- Whether the unclear license permits commercial use or redistribution—check the GitHub repository for actual license terms.
- Exact GPU memory and compute requirements for typical marketing datasets beyond the T4 16 GB test case.
- Whether tfp-nightly (a pre-release dependency) introduces stability risks in production workflows.
What it is and what it does
Meridian is Google's open-source framework for marketing mix modeling (MMM), a statistical technique that isolates the causal impact of marketing channels on sales and other KPIs using aggregated, privacy-safe data. It answers questions like 'How much revenue did each channel drive?' and 'What is my marketing ROI?', then uses those insights to optimize future budget allocation. The framework is built on Bayesian causal inference and uses a No U Turn Sampler (NUTS) for MCMC sampling, which is compute-intensive but well-suited to GPU acceleration.
Meridian handles both geo-level and national-level data, provides built-in visualizations, and supports calibration with experiments and prior information. It requires Python 3.10+, 15 runtime dependencies (including TensorFlow, scipy, pandas, and xarray), and is designed to run on GPUs for practical performance. The package is actively maintained, has no known vulnerabilities, and is documented with tutorials and a detailed API reference.
Use it for
- Measure the incremental revenue impact of each marketing channel (search, display, social, TV, etc.) to understand which channels drive the most value.
- Calculate marketing ROI and compare efficiency across channels to inform budget reallocation decisions.
- Optimize future media spend by using Meridian's budget allocation tools to simulate different spending scenarios.
- Validate MMM results against controlled experiments to calibrate model assumptions and improve forecast accuracy.
- Analyze geo-level marketing performance to identify regional differences in channel effectiveness and tailor strategies.
- Support privacy-safe marketing analytics without relying on cookie or user-level data.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you have marketing data and the infrastructure to run it.
Meridian is actively maintained, has no security vulnerabilities, and solves a real business problem (channel attribution and budget optimization) with a principled Bayesian approach. The main barriers are the unclear license (verify before production use), heavy compute requirements (GPU strongly recommended), and a large dependency footprint (15 runtime packages including TensorFlow and tfp-nightly). Install in a fresh virtual environment and plan for GPU resources.
Install
google-meridian on PyPI
Before you install
Low friction installation via pip; active maintenance with a recent release (23 days old). Requires Python 3.10+ and 15 runtime dependencies including TensorFlow, tfp-nightly, and tf-keras, which are compute-heavy. The project recommends a GPU (tested on T4 with 16 GB RAM) for practical use, though CPU operation is possible.
Requires Python 3.10+. GPU strongly recommended for practical model training; CPU-only operation will be slow. TensorFlow and tfp-nightly are heavy dependencies.
License in practice
License status is unclear—no SPDX identifier or explicit license text is declared in the package metadata. Before adopting Meridian in a commercial or proprietary project, verify the actual license terms in the GitHub repository to understand usage and distribution rights.
Quickstart
pip install google-meridian
# For GPU (Linux with CUDA):
# pip install google-meridian[and-cuda]
import meridian
# Load data and instantiate a model
model = meridian.MMM(data, ...)Verify before relying
- Whether the unclear license permits commercial use or redistribution—check the GitHub repository for actual license terms.
- Exact GPU memory and compute requirements for typical marketing datasets beyond the T4 16 GB test case.
- Whether tfp-nightly (a pre-release dependency) introduces stability risks in production workflows.
Package facts
| License | Not declared unclear |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 15 packagesarvizaltairbidictimmutabledictjoblibmatplotlibnatsortnumpypandasscipystatsmodelstensorflowtfp-nightlytf-kerasxarray |
| Maintenance | Actively maintained 23 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 96,163 / month, #13,226 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Programming Language :: Python :: 3Programming Language :: Python :: 3 :: OnlyTopic :: Other/Nonlisted TopicTopic :: Scientific/Engineering :: Information Analysis |
Evidence: google_meridian-1.7.1-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “MMM framework”
- google-meridianMeridian is a Bayesian marketing mix modeling (MMM) framework that…
- pymc-marketingPyMC-Marketing provides Bayesian statistical models for marketing…
- Pinterest-Generated-ClientProvides auto-generated Python bindings for Pinterest's REST API,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Information Analysis packages
A drop-in replacement for Python's standard `re` module that adds advanced regex features like nested sets, fuzzy matching, lookaround in conditionals, and full Unicode case-folding while maintaining backward compatibility.
pyarrow provides Python bindings to Apache Arrow's C++ libraries for efficient columnar data processing, serialization, and interoperability with pandas, NumPy, and other Python ecosystem tools.
NetworkX provides data structures and algorithms for creating, analyzing, and manipulating graphs and networks, supporting everything from simple undirected graphs to complex directed and weighted networks.
Connects Python applications to Snowflake data warehouses using the DB API 2.0 specification, enabling SQL queries, data transfers, and warehouse operations.
ContourPy calculates contours of 2D quadrilateral grids using C++11 algorithms wrapped in Python, offering serial and multithreaded implementations without requiring Matplotlib as a dependency.
Snowpark Python provides APIs to query and process data directly in Snowflake without moving data to your local system, with support for both native Snowpark and pandas-compatible interfaces.
Install it if you use Snowflake and want to process data without moving it to your application layer.
See also pymc-marketing · causalml · pymc3 · pymc · emcee · pgmpy · pymc-extras · tensorflow-probability · arviz-stats · policyengine-core