How we score a neighbourhood

Urban Index combines practical access (walk, cycling, public transport), everyday essentials, and local risk factors into clear 0–100 scores. This page explains the approach at a high level — what the scores represent, what they consider, and where the underlying ideas come from. New here? Start with what your scores mean.

Walk Score

Measures how easily you can run everyday errands on foot. Nearby essentials matter most, and additional options add smaller incremental value (the second café helps — but less than the first).

  • Each amenity's contribution follows a distance decay curve: full credit within roughly 400 m, fading to near-zero around 1.5 km. A 1.3× walking-network correction is applied before the curve so distances reflect realistic on-street routes, not crow-flies proximity.
  • Considers a balanced set of daily destinations (groceries, food, parks, schools, health services, community). Full-service supermarkets are weighted more heavily than small convenience outlets where that distinction applies.
  • Where street-network data is available, a street-connectivity signal adjusts the score based on how navigable the local grid is. Coastline-aware so waterfront addresses aren't penalised for the water in their catchment.
  • When elevation data is available, walking distances are inflated on hilly terrain — up to ~1.28× on very steep ground, derived from Tobler's hiking function — so amenities that are actually uphill register as farther away.
  • Includes a POI quality signal for selected categories using Google ratings — so a highly rated grocer or café counts more than a poor one nearby. This goes beyond standard walkability tools, adding a layer of real-world quality to the proximity score.

Cycling Score

Measures whether cycling is practical — not just possible. We value nearby infrastructure, prefer protected facilities where available, and consider how useful cycling is for reaching destinations.

  • Rewards closer cycling infrastructure more than far infrastructure.
  • Adjusts for infrastructure quality (e.g., protected vs on-road).
  • Adds network density and destination reach as separate scoring components alongside infrastructure proximity and quality.
  • Accounts for terrain when elevation data is available (hills change the experience).

Public transport score

Measures whether public transport is a dependable option. Where timetable data is available, we use GTFS feeds — the same open timetable standard that powers Google Maps public transport directions — sourced directly from Australian state transit authorities. This is what separates Urban Index from tools that count stops on a map: we score how often services actually run, not just whether they exist.

  • Calibration: the top of the scale is anchored using Melbourne's central transit hub — one of Australia's most multi-modal public transport precincts — so "100" reflects a real, reproducible reference rather than an arbitrary local maximum. Why does this matter for your score?
  • Scores train, ferry, tram, and bus based on nearby availability — each weighted to reflect its typical reliability and network reach. The score adapts to whichever modes are actually present near the address.
  • Rewards frequent service more than infrequent service when GTFS data is available.
  • Where timetable data is missing, the public transport score falls back to proximity-only (no frequency signal), but uses the same overall scaling when calibration data is present so the headline score stays on the same kind of scale.

Livability Score (overall)

A composite index that blends multiple dimensions into a single headline score. It is designed to answer a simple question: how well does this location support day-to-day life?

  • Combines green space, healthcare, education, public transport access (from the public transport score above), and active transport into a single score. The active-transport slice blends 60% Walk Score with 40% Cycling Score.
  • Includes Flood Safety as a first-class factor (currently available in Brisbane only).
  • The dimensions and their relative weights are grounded in independent international research into what makes cities genuinely liveable. Large-scale studies consistently identify the same factors as most influential: ease of getting around without a car, access to healthcare and education, green space, and protection from local environmental risks like flooding. The weights reflect how that body of research ranked each factor's contribution to everyday quality of life (OECD Better Life Index; EIU Global Liveability Index, 2023; Frank et al., 2010).

Value additions: POI quality + flood safety

POI quality signal (Google ratings)

Not all destinations are equal. For selected categories only, we use Google Places ratings as a light-weight quality signal so the score reflects lived experience — while still keeping distance as the primary driver. Ratings are adjusted using a Bayesian shrinkage formula, which prevents a 5-star place with 3 reviews from outscoring a 4.4-star place with 3,000.

Flood Safety (public flood awareness data)

In flood-prone cities, livability includes resilience. We incorporate flood awareness classification as a transparent safety factor and show it clearly in the breakdown (currently available in Brisbane only).

References

Our methodology is aligned with published research and public frameworks.

  1. Carr, L.J. et al. (2011). Walk Score as a global estimate of neighbourhood walkability.
  2. Frank, L.D. et al. (2010). Walkability Index / built environment and walking (connectivity and active transport framing).
  3. Winters, M. et al. (2013). Built environment influences on route selection for bicycle and car travel.
  4. Tobler, W. (1993). Three presentations on geographical analysis and modeling — hiking function for walking speed on slope.
  5. Xu, Y. et al. (2018). Food environment and grocery-access research (major vs convenience differentiation).
  6. Transportation Research Board (2013). Transit Capacity and Quality of Service Manual (TCRP Report 165).
  7. Economist Intelligence Unit (2023). Global Liveability Index.
  8. OECD (2020). Better Life Index Framework.
  9. Efron, B. & Morris, C. (1977). Stein's paradox in statistics — empirical Bayes estimation (basis for Bayesian shrinkage rating adjustment).