# google/meridian

Meridian is an MMM framework that enables advertisers to set up and run their own in-house models.

Repository: https://github.com/google/meridian
Canonical: https://ross.abutalabs.com/products/google-meridian
Homepage: https://developers.google.com/meridian
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T19:30:23+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 67
- inputs: {"age_days": 945, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1511, forks 286 (observed 2026-08-28T04:04:55.945231+00:00)

## What it is
Meridian is Google's open-source marketing mix modeling (MMM) framework built on Bayesian causal inference, letting advertisers run in-house models to measure channel impact and ROI. It supports geo-level and national data, experiment calibration, reach/frequency optimization, and budget allocation guidance.

## Use cases
- measure marketing ROI across channels
- optimize marketing budget allocation
- run in-house marketing mix modeling
- calibrate MMM with experiments
- model reach and frequency for video planning
- analyze how marketing channels drive revenue or KPIs

## When to choose
- you need privacy-safe, aggregated-data marketing measurement without user-level data
- you want a customizable Bayesian MMM with GPU-accelerated large-scale geo modeling
- you're migrating from LightweightMMM
- you need budget optimization and experiment calibration in one framework

## When to avoid
- you need real-time or user-level attribution rather than aggregated MMM
- you lack Python 3.11-3.13 and ideally a GPU
- you want a fully managed turnkey solution rather than a modeling framework

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, analytics, data-visualization
- domain: data-science, analytics, artificial-intelligence
- platform: python
- tags: marketing-mix-modeling, bayesian-inference, budget-optimization, causal-inference, privacy-safe-measurement, marketing, linux, macos, gpu

## Member repositories
- google/meridian (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:55.945231+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:32:23.153439+00:00, confidence not recorded.
  - readme: https://github.com/google/meridian (fetched 2026-08-28T04:04:55.945231+00:00, sha a4fc1379287f)
  - homepage: https://developers.google.com/meridian (fetched 2026-08-29T11:36:26.542967+00:00, sha e916d655c9e2)
- Data as of 2026-08-30T08:39:29.467469+00:00.