Planners deciding where to put the next EV charger, defibrillator, or transit stop rarely look at a flat list of candidate sites. Instead they build a layered web map: individual points get grouped into clusters as you zoom out, a heat layer shows where demand concentrates, and each site is colour-coded by a quantile classification so the top decile of demand always stands out — regardless of the underlying distribution's shape.
Quantile breaks are popular in choropleth mapping precisely because they guarantee a balanced-looking map — each colour class contains the same number of features — which makes skewed demand data (a few very busy spots, many quiet ones) far easier to read than a linear colour scale would.
A 3D city map lets you scatter candidate EV-charger sites, watch nearby ones merge into clusters as the cluster radius grows, colour each site by its quantile demand class, and blend in a live kernel-density heat layer.
Marker clustering groups points within a chosen radius into count-labelled bubbles; quantile colour coding ranks sites into five equal-count classes regardless of the raw demand distribution; the heat layer sums per-site kernels into a continuous density surface.
Adjust the number of candidate sites, drag the cluster radius from 0 (every pin visible) upward, tune the heat layer's opacity, and switch between quantile and linear colour modes. Regenerate for a fresh random demand surface.
Real siting tools like Folium's MarkerCluster and Leaflet.heat use exactly this two-layer trick — clustering for readability at low zoom, heat layers for density at a glance — so planners never have to stare at thousands of overlapping pins.