I'm building a locator for points of interest and nearby search. What should I use?

For a locator + nearby search feature, the best choice is usually a geospatial database / search engine with distance indexing.

Good options

1. PostgreSQL + PostGIS

Best if you want a solid general-purpose backend.

• Stores points, polygons, routes, etc.
• Supports:
  • “near me” queries
  • radius searches
  • bounding box search
  • geofencing
  • distance sorting
• Very reliable and widely used

Use this if:

• you already use PostgreSQL
• you want accuracy and flexibility
• your scale is moderate to high

---

2. Elasticsearch / OpenSearch

Best if your locator also needs strong text search.

• Great for:
  • POI search by name/category
  • autocomplete
  • full-text search
  • geo-distance filtering
• Supports geospatial queries and ranking

Use this if:

• search relevance matters a lot
• you need fast text + geo combined search
• you want autocomplete and typo tolerance

---

3. MongoDB geospatial indexes

Good for simpler implementations.

• Easy to use if you already use MongoDB
• Supports:
  • nearby queries
  • geo within a shape
  • distance-based filtering

Use this if:

• your app already uses MongoDB
• you want something simpler than PostGIS

---

4. Dedicated mapping/location services

If you want managed infrastructure and map/search data:

• Google Places API
• Mapbox Search / Geocoding
• Here Maps
• Foursquare Places

Use this if:

• you don’t want to manage your own POI database
• you need business/place data out of the box
• you’re okay with API costs and vendor dependence

---

My recommendation

• For your own POI data: use PostgreSQL + PostGIS
• For rich search/autocomplete: pair it with Elasticsearch/OpenSearch
• For ready-made places data: use Google Places / Mapbox / Here

Typical architecture

• PostGIS for storing POIs and geo queries
• Elasticsearch/OpenSearch for text search and ranking
• Optional Redis for caching frequent searches

If you want, I can suggest the best stack based on your app size, budget, and whether you're using SQL or NoSQL.