Who Uses GH
Geohash is widely used in database proximity searches. Redis has native geohash support via GEOADD/GEOSEARCH. Elasticsearch's geo_point type can store geohashes. MongoDB supports geohash-based geospatial indexing. It is popular for caching weather or location data by area.
- Backend developers
- Database engineers
- Redis / Elasticsearch users
- Location-based apps
How to Read GH Coordinates
- Geohash uses a 32-character alphabet: 0-9 and b-z (excluding a, i, l, o).
- Each character adds ~2.5 bits of precision (interleaving lat and lng bits).
- Characters 1–12: precision from ~5000 km down to ~37 mm.
- Adjacent cells in geohash space share a common prefix (with minor edge exceptions).
Converting GH by Hand
Geohash to DD: decode each base32 character into 5 bits, interleave to separate lat/lng bit strings, then convert each binary fraction to degrees. The algorithm alternates longitude bits (even) and latitude bits (odd).
Frequently Asked Questions
What alphabet does Geohash use?
Geohash uses a 32-character base32 alphabet: 0-9, b, c, d, e, f, g, h, j, k, m, n, p, q, r, s, t, u, v, w, x, y, z. Note the absence of a, i, l, o (visually confusable characters).
How long should a Geohash be?
4 chars = ~39 km, 5 chars = ~4.9 km, 6 chars = ~1.2 km, 7 chars = ~152 m, 8 chars = ~38 m, 9 chars = ~4.8 m, 10 chars = ~1.2 m. For most location apps, 8–9 characters is ideal.
Why do neighboring cells sometimes have very different geohashes?
Geohash has edge discontinuities at meridians (±180°), the equator, and within cells due to the space-filling curve it uses. Two points very close together can have completely different geohashes near these boundaries.