An interactive index of retail co-location clusters across North America and Europe — each scored by modelled catchment population and consumer spend.
Each dot is a co-location cluster — where large-format anchors group together.
Click a cluster to inspect its trade area. On mobile, drag the panel up for detail.
This platform is built upon high-fidelity open-source and public-purpose datasets. We gratefully acknowledge the following contributors:
The information provided on Woodfine Location Intelligence is for informational and research purposes only. While we utilize authoritative data sources, the results presented (including co-location scores and synthesized spend metrics) are generated through computational modeling and may contain inaccuracies.
Woodfine Group does not guarantee the accuracy, completeness, or timeliness of the data. Retailer locations and catchment analysis should not be used for high-stakes site selection, navigation, or critical infrastructure planning without independent verification.
We do not utilize real-time individual tracking data. Catchment population and consumer-spend metrics are derived from WorldPop high-resolution population grids assigned to clusters via a distance-decay radius model. No personally identifiable information (PII) is processed or displayed.
This interface operates on a Zero-Cookie and Zero-State Telemetry architecture. It does not deploy tracking cookies, retain session states, or harvest Personally Identifiable Information (PII). System interactions are limited to the collection of anonymized network routing and viewport data strictly for the purpose of auditing infrastructure performance.
Each location on this map is assigned a tier based on the composition of large-format retail anchors present within the co-location cluster — not on retailer count or proximity alone.
Tiers are labelled Regional, District, Local in descending co-location strength.
The framework is applied to 6,493 co-location clusters across North America and Europe, using OpenStreetMap point-of-interest records as the primary source. Clusters are identified by a two-pass spatial clustering algorithm (DBSCAN with a distance threshold calibrated to sub-metropolitan retail-park spacing, followed by an Intersection-over-Union merge pass). Cluster counts are descriptive rather than inferential: the totals shift with the clustering parameters, so the figures indicate market structure rather than precise measurements. Within each tier, clusters are ranked by geometric compactness and catchment characteristics. A “Regional Market” denotes a settlement containing at least one qualifying co-location; the published Top-400 ranks these markets by a composite score.
Population and consumer spend figures are derived from WorldPop 100 m population grids and modelled per-capita spend estimates, aggregated to H3 hexagonal cells within a 150 km radius of each cluster centroid. Spend is estimated by a simple chain: WorldPop cell population → a single, uniform national per-capita retail-spend rate → sum over the catchment cells. The per-capita rate is applied uniformly within a country; it does not adjust for local income, age, or household composition, so cell-level spend is an order-of-magnitude estimate rather than a measured figure.
Because the catchment is built from a fixed radius and then aggregated to H3 hexagons, the totals are sensitive to the choice of radius and hexagon resolution — the modifiable areal unit problem (MAUP). Different but equally reasonable zone definitions would yield different population and spend totals; the figures indicate scale, not precise counts. The intended refinement — replacing the radius-based catchment with mobility-defined catchments derived from parking-lot geo-fencing — is a planned extension as origin-destination panel data is acquired.
This methodology is documented in a technical note prepared by Woodfine Management Corp. (2026). For the full technical note or enquiries about the dataset, contact corporate.secretary@woodfinegroup.com.