Data Quality Review: Indian Postal-Code Candidates in OpenStreetMap
Hello OpenStreetMap Community,
I am relatively new to OpenStreetMap. While learning and conducting a data-quality audit, I performed a rule-based comparison of Indian postal-code tags (addr:postcode=* and postal_code=*) against India Post state-level PIN prefix ranges.
The audit identified 2,142 objects for manual review, containing 430 distinct PIN values across 23 states and union territories.
These are review candidates—not confirmed errors. State-prefix mismatches can be legitimate because postal delivery areas may cross administrative borders, and some records may be outdated or affected by historical reorganisations.
Examples requiring verification
-
Undrajavaram and Chilakapadu, Andhra Pradesh —
359687 -
Vaddegudem and Kommara, Andhra Pradesh —
127006 -
Barwani, Madhya Pradesh —
111111 -
Central Vista, Delhi —
020626 -
Mattaya, Kerala —
979303 -
Bihar Sharif, Bihar —
903101
Some may be clear typos or placeholder values, while others require verification against authoritative postal information and local knowledge.
Important review guidance
Please do not change a postcode solely because it does not match the state prefix. Before editing, please verify:
-
The official postal delivery area
-
Nearby post offices
-
Border-area delivery arrangements
-
Current local knowledge
-
Whether the tag belongs to the object itself or was copied from nearby data
The attached public share link contains:
https://filebin.net/osm_india_review
-
State
-
Place name
-
OSM class and place type
-
OSM object ID
-
Direct OSM link
-
Current tagged PIN
-
Expected state prefix
-
Coordinates
Per-state PIN counts may overlap because the same PIN can appear in multiple areas or border regions. Therefore, the all-India total is the deduplicated count of 430 unique PIN values, while 2,142 is the total number of flagged objects.
I am relatively new to OSM and do not yet have enough experience to review and correct all of these records safely. I am sharing this dataset so experienced contributors from any country can help validate the findings, correct confirmed errors where appropriate, or organise the work into manageable review tasks. Guidance on the proper workflow would also be greatly appreciated.
This dataset is intended only as a review aid. No automatic edits should be made from this list.
Thank you to everyone who can review, provide guidance, or help improve the data.