Reproducible U.S. geography for academic and policy research

Versioned boundary sets (states, counties, congressional districts, voting precincts) joined to the Census American Community Survey, with full provenance metadata, citable URLs, and spreadsheet / GeoJSON / Parquet export. Every analysis is reproducible — your collaborators get the exact same boundary edition a year later.

What it looks like

A typical workflow for research, academia & think tanks teams. Each line is one operation; together they take a few minutes and produce a published map plus a downloadable dataset.

# A policy researcher —
# "Compare median income by county across 2010-2020 Census editions."

mcp.call("list_shape_versions", { collection: "us-counties" })
  → tiger-2010, tiger-2015, tiger-2020, tiger-2024
    (each tag is a specific Census boundary edition)

# census_acs takes one year + one boundary version per call.
# Loop the editions to build a comparable time series:
for (const [year, version] of [[2010, "tiger-2010"], [2015, "tiger-2015"], [2020, "tiger-2020"]]) {
  await mcp.call("census_acs", {
    collection: "us-counties", version,
    year,
    variables: ["B19013_001E"],   // median household income
    slug: `counties-mhi-${year}`,
  });
}
  → 3 workspace datasets, one per edition, each version-pinned

mcp.call("compare_shape_versions", {
  collection: "us-counties",
  v1: "tiger-2010", v2: "tiger-2020",
})
  → which counties changed boundaries (e.g. Connecticut's 2022 planning-region switch)

# Publish three sibling reports — a small-multiples panel in your content
# management system from the three embed iframes (Maps MCP doesn't render
# multi-panel server-side yet).
for (const year of [2010, 2015, 2020]) {
  await mcp.call("create_report", {
    slug: `county-mhi-${year}`,
    name: `U.S. county median household income — Census ${year}`,
    public: true,
    config: {
      collection: "us-counties", version: `tiger-${year}`,
      datasets: [{ slug: `counties-mhi-${year}`, field: "B19013_001E",
                   palette: "sequential",
                   classify: { method: "natural_breaks", bins: 7 } }],
    }
  });
}
  → /v/county-mhi-2010  /v/county-mhi-2015  /v/county-mhi-2020

mcp.call("build_geojson", { slug: "county-mhi-2020" })
  → reproducible map file for the paper appendix

What you can build

• Provenance-anchored maps — Every report URL pins a specific boundary-edition tag. Re-running a year later reproduces the same figure.
• Multi-edition Census comparisons — Pull 5-year American Community Survey estimates for 2010, 2015, 2020 — one call per edition, each version-pinned. Stack the embedded maps side-by-side in your manuscript.
• Redistricting-aware analyses — compare_shape_versions tells you which congressional districts changed between Congresses; gate any time-series analysis on that diff.
• Open-data export for the paper appendix — Each published report exposes data.geojson, data.csv, and data.json — citable as supplemental material. Reviewers can re-render your figures in their own mapping software or run independent checks.
• Collaborative workspaces — Pro tier supports sub-workspaces — one per co-author or per cohort. Each has its own data, all under one bill.

The tools you'd actually use

How it differs from the alternative

vs. raw Census boundary files + IPUMS NHGIS: same data, but the joins are pre-built and every boundary edition is provenance-pinned. vs. ArcGIS Online: open licensing, citable URLs, spreadsheet / GeoJSON / Parquet downloads. The disaggregate_acs tool does area-weighted interpolation in one call — no manual database SQL. Free for 50 calls per month, forever — enough to ship a paper.

Pricing for research, academia & think tanks teams

Single researchers fit on Free or Starter. Labs and centers run on Pro ($50/month) for sub-workspaces per project or co-author. Full pricing details.

The same toolkit also covers grassroots organizing platforms & campaign tooling · healthcare & life sciences · civic tech & journalism · real estate & property tech · retail, consumer brands & site selection · insurance & risk modeling · logistics, delivery & field service · pollsters & survey research firms — one U.S.-geography surface for any team whose question comes down to where.