04 — Donor cluster locator

04 — Donor cluster locator

Question: Where do my donors physically cluster in Australia?

Outputs: answer.md (state breakdown + top-10 by count + top-10 by rate + donor-belt + donor-desert) + answer.html (two side-by-side MapLibre maps: left coloured by donor count, right coloured by donor rate per 1,000 adults, panning synced between panels, postcodes that appear on both lists outlined in dashed purple, CARTO light_all basemap).

Files in this folder

File What it is
prompt.md Instruction set.
sample_donors.csv 500 synthetic donors.
CLAUDE.md Auto-load for Claude Code.
expected_output.md Shape of a good markdown answer.

Nothing else is bundled. ABS boundaries + adult-population per POA come from the public endpoints listed in the prompt.

How to test

cd starter/examples/04_donor_cluster_locator
claude
> "Run CLAUDE.md."

Why “top by count” ≠ “top by rate”

The interesting moment in a good answer: the top postcodes by donor count are not the same as the top postcodes by donor rate per 1,000 adults. Vaucluse (2030) might top the count list, but a smaller high-density inner suburb often tops the rate list. That delta is the story — it tells you whether your brand has volume traction vs high-density traction.

The two-panel map design makes this visually obvious. Postcodes that appear in both lists get a dashed purple outline.