How we measure accessibility
The Accessibility score describes what is actually within reach of a neighbourhood — groceries, schools, transit, healthcare, parks and nine other amenity categories — measured by travel time on the real street network, not by drawing circles on a map. It is a 0–100 percentile within each city.
Last updated
What we measure
Reachability is measured with isochrones: for each dissemination area's centre, the Valhalla routing engine computes the polygon of everything reachable within a time budget, for four travel modes (walking, cycling, driving, public transit) and six time budgets (5, 10, 15, 20, 25 and 30 minutes) — 24 polygons per dissemination area. Transit isochrones incorporate schedules and walking transfers from each agency's GTFS feed.
Isochrones matter because a 1 km radius means different things in a downtown grid and a suburban cul-de-sac. A point 800 m away across a highway with no crossing is inside the circle but outside the isochrone.
Amenities are counted in 13 categories: food and groceries, dining and cafés, schools and education, health and personal care, banking and finance, retail shops, recreation and sports, entertainment venues, public transportation, community services, pet care services, vehicle services, and trails (measured as kilometres of trail, not a count). Amenity locations come from OpenStreetMap's Canada extract; school locations come from official feeds (Calgary, Vancouver, Edmonton open data) where available, with OpenStreetMap as the fallback.
The travel-time matrix you see in the lens UI — how many groceries within 15 minutes on foot, within 10 minutes by bike — is served directly from these isochrone counts, population-weighted from dissemination areas up to the neighbourhood.
Data sources and update cadence
- OpenStreetMap (Geofabrik Canada extract)
Amenity locations for 12 of 13 categories, trail geometry, and the routable street network
CadenceRe-imported with each pipeline run from the national extract
- Municipal school feeds (data.calgary.ca, opendata.vancouver.ca, data.edmonton.ca)
School locations with board and level metadata, richer than OpenStreetMap tags
CadenceRefreshed when the city updates the dataset, typically annually
- GTFS feeds: Calgary Transit, TransLink, Edmonton Transit Service, TTC
Transit stops, routes and schedules for transit isochrones
CadenceAgencies publish continuously; ingested on the pipeline cadence
- Valhalla routing engine
Computes all travel-time isochrones and drive times on the OpenStreetMap network
CadenceRecomputed each pipeline run (roughly 30 minutes of routing per city)
How the score is computed
The composite Access score is computed from a simpler set of inputs than the travel-time matrix, and we publish the difference rather than blur it. For scoring, eight amenity counts are taken within fixed distances of each dissemination area's centre — 1 km for transit stops, groceries, restaurants and retail; 2 km for schools, health services, recreation and trails. Each count is ranked against every other dissemination area in the same city with PERCENT_RANK and the eight percentiles are combined with fixed weights.
The asymmetry is deliberate. The lens UI needs the full mode-by-time matrix, which requires isochrones. The composite score only needs a stable ranking of neighbourhoods, and radius counts produce nearly the same ranking at a fraction of the compute. The two views are computed from the same amenity data.
Dissemination areas roll up to the neighbourhood by population-weighted averaging. A score of 50 means the neighbourhood is at its city's median for amenity access; scores do not transfer between cities.
Share of the Accessibility composite. Weights sum to 100% and are the same for every city.
What this score does not capture
Stated plainly, because a score is only worth trusting if its limits are.
- Counts measure presence, not quality. A grocery store is a grocery store to this score, whatever its prices, hours or shelves; opening hours are not modelled at all.
- OpenStreetMap is community-maintained. Coverage is strong in Canadian cities but not uniform, and a missing tag means a missing amenity.
- Transit isochrones use published schedules, not real-time service. Delays, cancellations and crowding are invisible to the measure.
- The score does not model the pedestrian experience — sidewalk quality, crossings, grade, winter conditions — only network reachability and amenity counts.
- Indoor accessibility (wheelchair access, elevators, stairs) is not measured.
- Walkability comparisons across cities are unreliable: amenity density is captured, but how pleasant or safe the walking network itself is differs between cities in ways the data does not capture.
Definitions
- Isochrone
- The polygon containing every point reachable from a starting location within a time budget for a given travel mode, computed on the routable network. PickYourPlace computes 24 per dissemination area: 4 modes × 6 time budgets.
- Dissemination area (DA)
- Statistics Canada's smallest standard geography, typically 400–700 residents. Accessibility is computed per DA and rolled up to the neighbourhood weighted by population.
- GTFS
- General Transit Feed Specification — the standard format transit agencies use to publish stops, routes and schedules. Powers the transit travel-time calculations.
- POI category
- One of 13 amenity groupings distilled from OpenStreetMap tags and official school feeds, from food and groceries to pet care services. Trails are measured in kilometres rather than counted.
- 15-minute measure
- The count of amenities inside the 15-minute isochrone for a chosen mode — the operational version of the "15-minute city" question.
- Percentile score
- The neighbourhood's rank among all neighbourhoods in the same city, 0–100. A 50 is the city median. Percentiles do not transfer between cities.
Frequently asked questions
How is this different from a walk score?
Two ways. Travel times are computed on the actual routable network with the Valhalla engine — including transit schedules — rather than straight-line distance. And the inputs are published: 13 named amenity categories, fixed radii and weights for the composite, and the full mode-by-time matrix in the lens view.
What does an access score of 80 mean?
The neighbourhood ranks above 80% of neighbourhoods in the same city on a weighted blend of eight amenity counts — transit, groceries, schools, dining, health, recreation, retail and trails. It does not mean the neighbourhood is "walkable" in an absolute sense; it means it out-ranks most of its own city.
Why do the lens page numbers and the composite score use different methods?
The lens page's travel-time matrix uses isochrones because the mode and time dimensions are the point of that view. The composite score uses fixed-radius counts (1 km and 2 km) because a stable within-city ranking is all it needs, and the two methods rank neighbourhoods nearly identically. Both are computed from the same amenity data, and we state the difference rather than imply one method did everything.
Where does the amenity data come from?
OpenStreetMap's Canada extract for 12 of 13 categories, and official municipal feeds for schools (Calgary, Vancouver and Edmonton open data), which carry board and level metadata OpenStreetMap lacks. Transit comes from each agency's GTFS feed: Calgary Transit, TransLink, Edmonton Transit Service and the TTC.
Does the score account for transit delays or service quality?
No. Transit reachability is computed from published GTFS schedules. Real-time performance, crowding and reliability are not in the data.