Delivered geospatial data workflow
Urban Geocoded Dataset for Guatemala City
A replicable workflow for turning selected urban areas into a structured, traceable and enriched geocoded dataset.
Problem and approach
Urban address data can be incomplete, fragmented or inconsistent. For a Zona 25 pilot in Guatemala City, BGPD used populated-area selection, systematic point generation, controlled reverse geocoding, batch manifests, result classification, deduplication and neighbor-based enrichment.
- Direct official addresses remain separate from inferred nearby references
- Run manifests support restartability and duplicate prevention
- Administrative and contextual attributes are enriched where appropriate
Documented pilot metrics
The following figures describe the completed first pass and are published with their contextual labels; nearby references are not direct official addresses.
- Unique points
- 27,382
- Direct matches
- 8,795
- Partial results
- 876
- Official addresses
- 611
- Nearby references
- 4,751
- Usable references
- 14,422
- Coverage
- 52.67%
- Duplicate point IDs
- 0
Processed and validated in the pilot.
Points with direct official-address matches.
Points with partial address information.
Unique official addresses represented.
Additional nearby-address references, not direct matches.
Points with a usable address reference.
Address-reference coverage after enrichment.
No duplicate point IDs in the completed first pass.
Outputs and boundaries
The delivered dataset separates direct, partial, missing and nearby-reference results so that users can assess coverage and provenance. It can support urban analytics, logistics, planning, market intelligence and field operations; it does not make an inferred nearby reference equivalent to a verified official address.