O-D catchment methodology — primary and secondary trade areas
The Woodfine co-location platform defines trade areas for each cluster using an Origin-Destination (O-D) model based on crow-flies distance rings over a hexagonal spatial grid. Each cluster is assigned two catchment zones that determine which population and spend data is attributed to it. The trade-area inputs to the deterministic ranking system and the V3 catchment ranking methodology flow from this model; population and spend layers are documented in trade-area data sources.
Correction (2026-08-02): this article's own title and body use "catchment" throughout for what is, by its own description, a straight-line crow-flies distance ring. Its sibling Trade-area methodology (a later article on the identical geometry — same 35 km/150 km rings, same H3 resolution-7 cells, numbers verified matching exactly) explicitly bans this: "any geometry derived from a straight-line formula reads as 'distance band (straight-line)' on the map face and in the detail panel — never 'catchment' and never 'trade area.'" The underlying geometry here is not in dispute; the terminology directly violates a sibling article's explicit, disclosure-relevant labeling rule. Flagged, not resolved.
Spatial Framework
Trade areas are computed using the H3 global hexagonal grid at resolution 7. Each H3 resolution-7 cell covers approximately 5.16 km² with a centre-to-centre spacing of approximately 2.11 km. The grid is continuous and consistent worldwide, enabling direct comparison between clusters across all countries in the current dataset.
Catchment Zone Definitions
Primary catchment: All H3 resolution-7 cells whose centre point falls within 35 km (crow-flies) of the cluster centroid. This zone represents the immediate trade area where the majority of regular shopping trips originate.
Secondary catchment: All H3 resolution-7 cells whose centre point falls between 35 km and 150 km (crow-flies) of the cluster centroid. This zone captures the wider regional draw, including occasional shoppers and cross-regional trips.
The 35 km primary boundary is a provisional parameter based on established retail geography conventions. It is subject to refinement once empirical origin-destination data becomes available.
The 150 km outer boundary aligns with the platform's data collection radius, ensuring that every cell contributing to a cluster's catchment has been ingested and verified.
Distance Method
All distances are calculated as the crow-flies (great-circle) distance using the haversine formula. No drive-time routing is used. This approach is:
- Reproducible without map routing infrastructure
- Consistent across urban and rural geographies
- Computationally efficient over millions of H3 cells
- Suitable as a baseline before empirical O-D data is available
H3 ring traversal identifies candidate cells efficiently, with haversine as the definitive distance measure for final inclusion.
HOME and AWAY Perspectives
The platform distinguishes two perspectives on catchment population.
HOME: Population counts derived from residential data (WorldPop 2026). Represents where people live within each catchment zone. This is the default view and is fully implemented.
AWAY: Population counts representing daytime or workplace population. Workplace distribution differs from residential distribution — concentrated in commercial districts and employment centres rather than dispersed across residential areas. The AWAY perspective is planned; the data source is pending.
One Cell, Multiple Clusters
A single H3 cell may fall within the catchment of multiple co-location clusters. This is intentional: trade areas are not exclusive territories. A household within 35 km of two competing clusters contributes to both clusters' primary catchment populations. This reflects the competitive retail landscape and is foundational to the cross-cluster comparison methodology; cluster boundary handling at the same parking lot is documented in cluster deduplication threshold.
Application
Catchment zone membership is the basis for:
- Population aggregation (census data by zone)
- Spend aggregation (grocery, hardware, wholesale spend by zone)
- Cross-cluster competitive ranking (see: Catchment Ranking Methodology V3)
The catchment polygons displayed on the map are generated from the same 35 km / 150 km crow-flies radii, visualised in two distinct colours to distinguish primary from secondary zones.
See also
- catchment-ranking-methodology-v3
- Trade area data sources — population and spend
- Retail co-location methodology
References
- Catchment area — Wikipedia, accessed 2026-06-14
- Trade area — Wikipedia, accessed 2026-06-14
- H3: Uber's Hexagonal Hierarchical Spatial Index — H3 Geo, accessed 2026-06-14
- WorldPop Global High Resolution Population Denominators Project — WorldPop, University of Southampton, accessed 2026-06-14
Wikipedia content reproduced under CC BY-SA 4.0.
Cluster centroids from which catchment distances are measured are derived from OpenStreetMap POI records. OpenStreetMap data © OpenStreetMap contributors, licensed under ODbL.