Convert

Convert your data to
loc_id.

Coordinates, postal areas, tribal lands, watersheds, and administrative codes resolve to one shared place key. Run the conversion here, run it on your own machine, or send us the file and we convert it for you.

Run it here

Hosted tools

Paste rows and get loc_ids back in the browser.

  • First 100 rows free, no account
  • Resolved CSV download
  • Larger batches through a funded account

Convert a sample

Run it yourself

Local tools

Install the MCP runtime and geometry packages, and convert offline.

  • Your data stays on your machine
  • No per-row cost once installed
  • Works in restricted and air-gapped environments

Get the local runtime

We run it

Managed conversion

Send the file and a short description. You get your rows back keyed to loc_id.

  • Messy, mixed, or undocumented location columns
  • Reference systems with no published crosswalk yet
  • Quoted per job

Describe your data

All three return the same key.

The hosted tools, the local runtime, and a managed job run the same resolution against the same maintained geography. Your original columns survive untouched, a loc_id column is added, and rows that did not resolve come back marked with the reason instead of being silently dropped. Start on whichever path fits today and move between them later without redoing the work.

What you can convert

Coordinates resolve to the administrative chain for the country they fall in. Reference codes resolve through the published crosswalk catalog, which covers postal areas, tribal areas, watersheds, and statistical geography alongside the administrative spines. Both lanes return the same key, so a file with a coordinate column and a file with a postal column end up joinable to each other.

Non-administrative geography is first-class here. A watershed, a tribal area, and a postal area each carry their own identity and a typed relationship back to the administrative spine, so a join across them keeps its meaning instead of collapsing into a nearest-county guess.

Which path to pick

Convert a sample here first. It costs nothing, needs no account, and tells you the match rate on your real data before you commit to anything.

Move to the local tools when the data cannot leave your machine, when the volume makes per-row cost the wrong shape, or when the conversion needs to run inside a pipeline you control. Ask for a managed job when the input is hard rather than large: hand-entered place names, mixed or undocumented code columns, or a reference system that has no published crosswalk yet.

After the conversion

A file keyed to loc_id joins directly against maintained data packs, downloadable geometry, and live feeds. The same key drives Explore maps, Research corpora, and the Agent and MCP lane, so the conversion is the only step that has to happen once.

Browse the data packs