From raw public data to the stories that change how decisions are made.
An open-source policy data platform. 13 federal and state data sources. 42 indicator variables. 72 Wisconsin counties. 1,652 census tracts.
| Tract | Score | Tier |
|---|---|---|
| Variable | β Coefficient | Std Error | p-value |
|---|---|---|---|
Policy Data Chat
Policy Data Infrastructure
Policy Data Infrastructure is an open-source national policy data platform. It ingests data from 13 federal and state sources, stores indicators at multiple geographic levels in PostGIS, runs a statistical analysis engine (OLS regression, tipping points, ICE segregation indices, bootstrap confidence intervals, Blinder-Oaxaca decomposition), and generates evidence-based policy narratives.
The platform connects candidate policy positions to the data that measures the problems they address. 85 policy positions from progressive candidates are crosswalked to 38 equity dimensions, enabling automated links between data and advocacy. Built for grant reviewers, policy analysts, and advocates who need real numbers — not dashboards for the sake of dashboards.
| Source | Type | Vintage | Coverage | Variables |
|---|---|---|---|---|
The entire platform is open source under the Apache-2.0 license. The Go backend has 13 datasource adapters, a 7-stage pipeline with validation gates, a statistical engine with 10 analysis functions, a narrative template engine, and an HTMLCraft deliverable builder. 380+ automated tests. The Python layer handles data ingest, evidence card generation, and factor analysis.
Stack: Go 1.24 (pgx/v5, Gin, Cobra) · PostgreSQL 16 + PostGIS 3.4 · Python 3.11 · Alpine.js · Caddy · Hetzner VPS