Faceted Search APIs: Dynamic, Category-Aware, and Scalable (on Elasticsearch)

This is Part 5 of the "Building a Scalable, Faceted Online Marketplace" series. Read the Introduction here.

Why Faceted Search APIs?

Faceted search is the backbone of modern e-commerce discovery. Users expect to filter by brand, price, color, rating, and more—dynamically, and with blazing speed. To deliver this at scale, we need robust, category-aware APIs powered by Elasticsearch.

What You'll Learn

• How to design dynamic, category-driven facet APIs
• Querying Elasticsearch for both products and facet counts
• Handling dynamic attributes and nested fields
• Real-world performance tips for large datasets

Key Features

• Dynamic Facets: APIs adapt to category and available data
• Scalable Aggregations: Uses Elasticsearch's aggregation framework
• Flexible Filtering: Supports multi-select, range, and nested filters
• Consistent UX: Backend and frontend always in sync

Example API Endpoints

• /api/products?category=Shoes&brand=Nike&color=Red
• /api/facets?category=Shoes

Sample Elasticsearch Query

{
  "query": { ... },
  "aggs": {
    "brand": { "terms": { "field": "brand.keyword" } },
    "color": { "terms": { "field": "attributes.color.keyword" } },
    "price": { "range": { "field": "price", "ranges": [ ... ] } }
  }
}

Real-World Tips

• Use keyword fields for facet values
• Cache facet queries for popular categories
• Limit aggregation size for performance
• Sync facet definitions between backend and frontend

Next up: Zero Downtime Reindexing & Unblocking in Elasticsearch

In the next article, we'll cover how to safely reindex and unblock Elasticsearch indices in production, with zero downtime!