Geocoding is the process of converting a human-readable address into geographic coordinates (latitude and longitude). It is what happens when you type an address into a mapping app and a pin appears on the map.
How Geocoding Works
A geocoder takes an address as input, parses it into components (street number, street name, city, postal code), and looks it up against a reference database of known addresses and their coordinates. The result is a coordinate pair and a confidence score indicating how well the address matched. A score of 100% means an exact rooftop match; lower scores indicate interpolated or approximate results.
Geocoding Quality Levels
- Rooftop — coordinates for the specific building footprint
- Parcel centroid — centre of the property parcel
- Street interpolation — estimated position along a street segment
- Intersection — coordinates of the nearest intersection
- Postal code centroid — centre of the postal code area
- City centroid — centre of the city (lowest precision)
The quality level matters enormously for the use case. Routing an ambulance requires rooftop precision. Analyzing regional demographic trends may be fine with postal code centroids. Always document the geocoding quality level in your dataset's metadata.
Reverse geocoding does the opposite: it takes coordinates and returns the nearest address. This is used in mobile apps to show your current location as an address and in data processing to add address information to GPS tracks. A delivery company might use reverse geocoding to convert driver GPS logs into readable address histories.
Geocoding in Canada
Canada Post's address database is the authoritative source for Canadian addresses. Statistics Canada publishes the Postal Code Conversion File (PCCF) for postal code level geocoding — it maps each postal code to a representative point and census geography. The National Address Register (NAR), maintained by Natural Resources Canada, is a growing national address dataset.
Common Geocoding Challenges
Address data quality directly affects geocoding results. Misspellings, missing unit numbers, non-standard abbreviations, and outdated addresses all reduce match rates. A data cleaning step before geocoding — standardizing address formats, correcting obvious errors — significantly improves results. See Address Validation for practical techniques.
Key Takeaways
- Geocoding converts addresses to coordinates; reverse geocoding does the opposite
- Quality levels range from rooftop precision to city centroid approximation
- Always document geocoding quality in dataset metadata
- Clean address data before geocoding to maximize match rates
Learn how to analyze spatial data in Spatial Analysis.