Alpha — GeoCanvas is in early development. Expect rough edges and breaking changes.

Automate and integrate

Drive GeoCanvas from your own scripts and notebooks.

You've built a map by hand — brought in your data, coloured it, shared the link. Now you want that to happen on its own: a fresh map every morning, a dozen maps from a dozen files, your latest numbers pushed up without anyone opening a browser. The good news is that anything you can do by clicking in GeoCanvas, you can also do from a script.

Why automate

When clicking doesn't scale

Doing something once by hand is quick. Doing it every day, or for fifty datasets, or the moment new data lands, is a job for a few lines of code. Creating a map, adding a layer, colouring it by a number, flipping on the share link — every step in the editor has a twin you can run from a script or a notebook. You write the recipe once and let it run.

What you can automate

Things worth handing to a script

A few of the jobs people reach for first:

  • Publish a fresh map every morning. Point a script at your latest export and have it rebuild the map on a schedule, so the link your team opens is always current — never last week's numbers.
  • Make many maps at once. One per region, per client, per week — generate them all in a loop instead of repeating the same clicks fifty times.
  • Refresh data the moment it's ready. Hook GeoCanvas onto the end of whatever produces your data, so a new file turns into an updated, shareable map without anyone lifting a finger.

The flagship NYC taxi-fares map — 263 zones coloured by their average fare, from about $12 in Manhattan up to $67 around Newark Airport — is exactly the kind of map you'd stamp out from a script and refresh whenever the underlying numbers change.

How

Meet the Python SDK

The friendliest way in is the Python SDK. It speaks GeoCanvas in plain Python, so creating a map, adding a layer and turning on sharing each take a line or two — no plumbing to learn. It drops straight into a notebook, where you can run a cell and watch the map appear next to your data.

The full step-by-step lives on the Python SDK page — install it, copy the starter script, and you'll have a map building itself in a couple of minutes.

Happy to keep clicking for now? That's every other page in this guide: bring data in with Upload your data, keep it tidy in the catalog, and publish with Sharing maps. Automation just teaches a script to do those same clicks for you — at whatever scale and on whatever schedule you need.