Skip to content

CONCEPT Cited by 1 source

Historical pickup heatmap

Definition

A historical pickup heatmap is a spatial aggregation of where, in a given area, riders and drivers have actually successfully met in past trips. It is used as a recommendation signal for future pickup-spot suggestions: the surface-of-density over historical successful-pickup locations becomes the ranking function for candidate meet-spots.

Why it beats pure topology

The topologically-nearest walkable road point is often wrong in ways the map does not encode:

  • Wrong side of a busy road.
  • A no-stopping zone not tagged in OSM.
  • Across a median or a one-way that adds minutes of driving.
  • A physically-accessible point that no one actually uses because of local knowledge (mall parking is easier on the north side; the apartment complex guard insists on the east gate at night).

Historical pickups have already passed the feasibility filter by real-world selection. The heatmap surfaces the spots that real riders have converged on without having to encode every reason they converged there.

Canonical instance — Lyft gated-community outside-gate spots

Lyft's 2026-04-23 write-up makes the point explicitly:

"The team looked at historical ride patterns to surface spots that real riders have actually used near gates — so we're not just pointing you toward the gate in theory, we're pointing you to where riders in your community actually go."

The article also includes a visual — "Pickup location heatmap of an example gated community" — confirming the signal is literally rendered as a heatmap in the design process.

Seen in

Caveats

  • Cold start: a new community with no historical trips has no heatmap signal — fall back to topology or driver feedback.
  • Privacy: pickup locations are PII-adjacent; the aggregation granularity must be coarse enough not to expose individual rider trips. Not discussed in the Lyft post; this is a known concern for any system building on historical user-location data.
Last updated · 550 distilled / 1,221 read