← Back to the map

How WalkUp scores

WalkUp asks one question about a place: how much of daily life is a short walk away? It finds the nearest groceries, food, transit, schools, parks and more, measures the real walking time to each, and turns that into a single 0–100 score with a per-category breakdown. Here's exactly how, so the number is one you can check rather than one you have to trust.

The 15-minute city

The idea behind the score is the 15-minute city: a good place to live is one where the things you reach for in an ordinary week — a grocery run, a coffee, a pharmacy, a bus, a park — are a short walk from your door, not a drive. WalkUp measures how much of that daily life is genuinely within walking reach of a single point on the map.

Real walking time, not distance

Distance on a map lies — a shop 200 metres away across a motorway is not a 200-metre walk. So WalkUp scores the real walking time along the actual street and path network, never the straight-line "crow-flies" distance.

It also corrects for how fast people actually walk errands. The routing engine prices every walk at an idealized 5 km/h on a perfect path with no crossings or waits. Stopwatch-timed real walks run about 1.4× longer — an effective 3.5 km/h once you count pace, detours and signals. WalkUp applies that correction, so the minutes in your report are honest walking minutes. That's why the credit window below reads as ~7.1 and ~21.4 real minutes rather than the 5 and 15 the model was first calibrated against.

Closer counts for more: distance-decay

Not every nearby amenity counts equally. An amenity you can reach in about 7.1 minutes or less (5 minutes at the engine's ideal pace) earns full credit. Past that, credit fades smoothly the further the walk, reaching zero at about 21.4 minutes (the 15-minute outer edge) — beyond that, it's not really a walk you'd make for an errand. So a grocery store 4 minutes away lifts the score far more than one 18 minutes away.

Variety counts: multi-counting

For some categories, one is plenty — a single nearby pharmacy or supermarket covers the need. For others, choice is the point: ten restaurants within a walk is a real amenity that one is not. So a handful of these categories count several nearby instances, each a little less than the last, up to a cap:

Every other category counts only its nearest instance — having two pharmacies next door isn't twice as walkable as one.

What matters most: category weights

The categories don't all pull equally on the final score. Groceries, food and transit are the backbone of a walkable week and carry the most weight; the education tiers and nightlife are lighter completeness signals. The weights are relative — the score is normalized by the categories actually present — and are a tunable default, not a law:

The five bands

The 0–100 score collapses into five plain-language bands, reused from the published Walk Score scale:

Where the data comes from

Every amenity, street and path WalkUp uses comes from OpenStreetMap, the open map of the world built by its contributors. The scoring method re-implements the published Walk Score methodology on that open data — we build our own because the official Walk Score API only covers the US and Canada, and WalkUp is meant to work anywhere OpenStreetMap does.

Map data © OpenStreetMap contributors, available under the Open Database License (ODbL).

The honest limits

WalkUp is only as complete as the map. In a well-mapped city the picture is rich; in a thinly-mapped area a real amenity can be missing simply because no one has added it yet. The score reflects what OpenStreetMap knows, not the ground truth.

So the report is careful to tell two different "zeros" apart. "None nearby" means we looked and there genuinely isn't one within walking reach — that legitimately lowers the score. "Couldn't check" means a data lookup failed, so that category is marked unavailable and left out of the score entirely — never counted as if the area had nothing. A missing answer is never silently scored as a bad one.